E FAST Technical Manual

User Manual:

Open the PDF directly: View PDF .
Page Count: 147

Download
Open PDF In Browser	View PDF

Forward

Formative Assessment
System for Teachers™
(FAST™)
Abbreviated Technical Manual for Iowa
Version 2.0, 2015–2016

NOTICE: Information for measures that were not implemented
statewide is omitted from this publication at the request of
The Iowa Department of Education.

Page | 1
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights
reserved.

Forward
Formative Assessment System for Teachers™ (FAST™): Abridged Technical Manual
Version for Iowa 2.0
Copyright © 2015 by Theodore J. Christ and
Colleagues, LLC. All rights reserved.
Warning: No part of this publication may be
reproduced or transmitted in any form or by any
means, electronic or mechanical, now known or later
developed, including, but not limited to,
photocopying, recording, or the process of scanning
and digitizing, transmitted, or stored in a database or
retrieval system, without permission in writing from
the copyright owner.

Published by Theodore J. Christ and Colleagues, LLC (TJCC)
Distributed by TJCC and FastBridge Learning, LLC (FBL)
43 Main Street SE Suite # 509
Minneapolis, MN 55414
Email: help@fastbridge.org
Website: www.fastbridge.org
Phone: 612-424-3714

Prepared by Theodore J. Christ, PhD as Senior Author and Editor with contributions from (alphabetic order)
Yvette Anne Arañas, MA; LeAnne Johnson, PhD; Jessie M. Kember, MA; Stephen Kilgus, PhD; Allyson J. Kiss;
Allison M. McCarthy Trentman, PhD; Barbara D. Monaghen PhD; Gena Nelson, MA; Peter Nelson, PhD; Kirsten
W. Newell, MA; Ethan R. Van Norman, PhD; Mary Jane White, PhD; and Holly Windram, PhD as Associate
Authors and Editors.

Citation:
Theodore J. Christ and Colleagues (2015). Formative Assessment System for Teachers: Abbreviated Technical
Manual for Iowa Version 2.0, Minneapolis, MN: Author and FastBridge Learning (www.fastbridge.org)

Page | 1
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Forward

Forward
Our Story
Over 15 years, our research team received competitive funding from the US Department of Education to build
knowledge and improve educational assessments. The Formative Assessment System for Teachers™ (FAST™) is
how we disseminate that work. It was developed to reduce the gap between research at universities and
results in the classroom, which can take 10 to 30 years in a typical cycle. FAST™ has reduced that to weeks or
months. We innovate today and implement tomorrow.
In 2010, Dr. Zoheb H. Borbora (Co-Founder) and I (Founder) conceptualized and created FAST™ as a cloudbased system at the University of Minnesota. Our goal was to use research and technology to make it easier
for teachers to collect and use data to improve student outcomes. We initially tried to create and distribute for
free. That model was unsustainable. We had no resources to achieve our standard of excellence. The school
leadership and teachers that were our partners preferred an excellent low-cost system over a good free
system. With the University, we made this transition in early 2012. The demand for FAST™ was tremendous
and overwhelming. The demand quickly outpaced what we could support at the University. So, FastBridge
Learning was launched in the spring of 2015 to distribute and support FAST™.
FastBridge Learning is a partnership between the University of Minnesota, “Theodore J. Christ and Colleagues”
(TJCC), and the FAST™ team. In 2014–15, FAST™ was used in more than 30 states, which includes a statewide
adoption in Iowa (92% of schools). FAST™ users exceeded 5 million administrations in 2014–15. The feedback
has been tremendous. In partnership, we continue to strive for our vision: Research to Results™.

Our Mission
The University of Minnesota and TJCC continue with their mission to innovate with research and
development. FastBridge Learning continues to translate those innovations into practice for use by teachers.
We aspire to provide a seamless and fully integrated solution for PreK–12 teaching, learning, and assessment.
We are not just about assessment. We are about learning and optimization of teaching, parenting, and being.
FAST™ was the centerpiece of FastBridge Learning in 2014–15. It will soon be supplemented with teaching
and learning tools (e.g., materials, guides, reports, and automated software-based personalized instruction).
Like-minded researchers and trainers are encouraged to join our cause (ted@fastbridge.org,
www.fastbridge.org). Educators are invited to join FastBridge Learning and challenge us to innovate and
deliver solutions for the most challenging teaching and learning problems.

Our Values
We are values driven. We strive towards our values. Those are: Tell the Truth, Respect the Teacher, and
Deliver High-Value Solutions. These values inform our work, and we measure our successes against them.
We invite others to hold us accountable.

Page | 2
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Forward
We Tell the Truth
Perhaps more than at any time in the past, educators are bombarded with claims of research, evidence, data,
statistics, and assessments. These words relate to very important and lofty subject matter that is undermined
when they are misused or abused for marketing, sales, or self-promotion. We strive to know better and do
better. We strive to tell the truth. And, the truth is that all research has its limitations, as do the various types of
assessment and data. That is true regardless of any misleading claims. So, we acknowledge the limitations of
the tools we deliver, and we do not exaggerate our claims. Instead, we deliver multiple types of assessment in
one suite and provide guidance so users use the right tool for the right purpose. It is an honest and better
solution for teachers.
We Respect the Teacher
At the beginning, FAST™ (Formative Assessment System for Teachers) was named to make the value of
teachers explicit. They are the primary intended user so we include them and value their opinions that guide
our research, development and refinement. We are in service to the professional educator. We aspire to make
their work easier and more effective.
I (Dr. Christ) was a paraprofessional, residential counselor, and special education teacher. I earned my MA and
PhD degrees in school psychology as part of my professional development to be a better teacher for students
who are at risk. I always intended to return to the classroom but was drawn into a research career, which gives
me great joy. Our team respects, responds, and solicits input from teachers who work to solve difficult
problems with limited resources. We try to understand and meet their needs—and yours—with quality
research, engineering, training, and support.
We Deliver High-Value Solutions
We strive to provide systems and services that are effective, efficient, elegant and economical.





An effective solution improves child outcomes.
An efficient solution saves time and resources.
An elegant solution is pleasing and easy to use.
An economical solution is sustainable for us and our users.

Design and user focus are central tenets.

Page | 3
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Forward

Final Note from Dr. Christ
Thank you for considering our work. It is a compilation of efforts by many graduate and undergraduate
students, researchers, teachers, principals, and state agencies. This would not exist without them. I am very
thankful for their contributions. I hope it confers great benefit to the professional educator and the children
they serve. Education has the potential to be the great equalizer and reduce the gaps in opportunity and
achievement; however, we will only realize that potential if education is of high and equitable quality. I hope
we help in the pursuit of that.
Sincerely,
Ted
Theodore J. Christ, PhD
Co-Founder and Chief Scientific Officer

Page | 4
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Table of Contents

Contents
Forward ....................................................................................................................................................................................2
Our Story..................................................................................................................................................................................2
Our Mission .............................................................................................................................................................................2
Our Values ...............................................................................................................................................................................2
We Tell the Truth..............................................................................................................................................................3
We Respect the Teacher ................................................................................................................................................3
We Deliver High-Value Solutions ...............................................................................................................................3
Final Note from Dr. Christ ..................................................................................................................................................4
Table of Figures ..........................................................................................................................................................................8
Table of Tables............................................................................................................................................................................9
Section 1. Introduction to FAST™ and FastBridge Learning.................................................................................... 13
Chapter 1.1: Overview, Purpose, and Description ................................................................................................. 13
Background..................................................................................................................................................................... 13
All in One.......................................................................................................................................................................... 13
Support and Training .................................................................................................................................................. 14
Trusted Results............................................................................................................................................................... 14
Curriculum-Based Measurement (CBM) ............................................................................................................... 14
Prevention and Intervention .................................................................................................................................... 15
Chapter 1.2: Development ............................................................................................................................................. 15
Chapter 1.3: Administration and Scoring.................................................................................................................. 15
Setting Standards ......................................................................................................................................................... 15
Chapter 1.4: Interpretation of Test Results ............................................................................................................... 16
Standard Setting ........................................................................................................................................................... 16
Chapter 1.5: Reliability ..................................................................................................................................................... 20
Chapter 1.6: Validity.......................................................................................................................................................... 21
Chapter 1.7: Diagnostic Accuracy of Benchmarks ................................................................................................. 21
A Conceptual Explanation: Diagnostic Accuracy of Screeners..................................................................... 22
Decisions that Guide Benchmarks Selection: Early Intervention and Prevention................................. 22
Area Under the Curve (AUC) ..................................................................................................................................... 23
Decision Threshold: Benchmark.............................................................................................................................. 23
Section 2. Reading and Language.................................................................................................................................... 25
Chapter 2.1: Overview, Purpose, and Description ................................................................................................. 25
Page | 5
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Table of Contents
earlyReading................................................................................................................................................................... 25
CBMreading .................................................................................................................................................................... 27
aReading .......................................................................................................................................................................... 29
Chapter 2.2: Development ............................................................................................................................................. 37
earlyReading................................................................................................................................................................... 37
CBMreading .................................................................................................................................................................... 38
aReading .......................................................................................................................................................................... 38
Chapter 2.3: Administration and Scoring.................................................................................................................. 43
earlyReading................................................................................................................................................................... 43
CBMreading .................................................................................................................................................................... 43
aReading .......................................................................................................................................................................... 44
Chapter 2.4: Interpreting Test Results ........................................................................................................................ 44
earlyReading................................................................................................................................................................... 44
CBMreading .................................................................................................................................................................... 46
aReading .......................................................................................................................................................................... 47
Chapter 2.5: Reliability ..................................................................................................................................................... 49
earlyReading................................................................................................................................................................... 49
CBMreading .................................................................................................................................................................... 59
aReading .......................................................................................................................................................................... 72
Chapter 2.6: Validation..................................................................................................................................................... 72
earlyReading................................................................................................................................................................... 72
CBMreading .................................................................................................................................................................... 80
aReading .......................................................................................................................................................................... 86
Chapter 2.7: Diagnostic Accuracy ..............................................................................................................................101
earlyReading.................................................................................................................................................................101
CBMreading ..................................................................................................................................................................108
aReading ........................................................................................................................................................................114
Section 6. FAST™ as Evidence-Based Practice ............................................................................................................121
6.1: Theory of Change ....................................................................................................................................................122
6.2: Formative Assessment as Evidence-Based Practice ....................................................................................122
US Department of Education..................................................................................................................................122
Historical Evidence on Formative Assessment.................................................................................................123
Evidence Based: Contemporary Evidence on Formative Assessment.....................................................123
6.3: Evidence-Based: Formative Assessment System for Teachers ................................................................125
FAST™ Improves Student Achievement .............................................................................................................125
Page | 6
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Table of Contents
FAST™ Improves the Practice of Teachers..........................................................................................................126
FAST™ Provides High Quality Formative Assessments..................................................................................126
References...............................................................................................................................................................................127
Appendix A: Benchmarks and Norms Information...................................................................................................145
Appendix B: FastBridge Learning Reading Diagnostic Accuracy ........................................................................146

Page | 7
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Table of Contents

Table of Figures
Figure 1 A priori model for unified reading achievement ....................................................................................... 40
Figure 3. Example of a student's aReading report with interpretations of the scaled score....................... 48
Figure 14 Theory of Change .............................................................................................................................................122

Page | 8
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Table of Contents

Table of Tables
Table 1 Example Standards for Informational Text .................................................................................................... 42
Table 2 Foundational Skill Examples for Kindergarten and First Grade Students........................................... 42
Table 3 Cross-Referencing CCSS Domains and aReading Domains..................................................................... 42
Table 4 Weighting Scheme for earlyReading Composite Scores .......................................................................... 45
Table 5. Demographic Information for earlyReading Alternate Form Sample................................................. 49
Table 6 Alternate Form Reliability and SEm for earlyReading................................................................................ 50
Table 7 Internal Consistency for earlyReading subtests of variable test length.............................................. 52
Table 8 Internal Consistency for earlyReading subtests of fixed test length .................................................... 52
Table 9 Descriptive Information for earlyReading Test-Retest Reliability Sample.......................................... 54
Table 10 Test-Retest Reliability for all earlyReading Screening Measures ......................................................... 55
Table 11 Disaggregated Test Re-Test Reliability for earlyReading Measures ................................................... 56
Table 12 Inter-Rater Reliability by earlyReading Subtest ........................................................................................ 57
Table 13 Demographic Information for earlyReading Reliability of the Slope Sample................................. 57
Table 14 Reliability of the Slope for all earlyReading screening measures........................................................ 58
Table 15. Reliability of the Slope for earlyReading measures, Disaggregated by Ethnicity ........................ 59
Table 16. Demographic Information for CBMreading First Passage Reduction Sample .............................. 61
Table 17 Descriptive Statistics for First Passage Reduction..................................................................................... 62
Table 18. Demographic Information for Second Passage Reduction Sample.................................................. 62
Table 19 Cut-points used for assigning students to CBMreading passage level based on words read
correct per minute (WRC/min)........................................................................................................................................... 63
Table 20 Descriptive Statistics for Second CBMreading Passage Reduction Sample .................................... 64
Table 21 Alternate Form Reliability and SEm for CBMreading (Restriction of Range)................................... 64
Table 22 Internal Consistency for CBMreading Passages......................................................................................... 66
Table 23 Split-Half Reliability for CBMreading passages.......................................................................................... 66
Table 24 Evidence for Delayed Test-Retest Reliability of CBMreading................................................................ 67
Table 25 CBMreading Delayed Test-Retest Reliability Disaggregated by Ethnicity ....................................... 68
Table 26 Evidence of Inter-Rater Reliability for CBMreading.................................................................................. 69
Table 27 Reliability of the Slope for CBMreading........................................................................................................ 69
Table 28 Reliability of the Slope of CBMreading by Passage using Spearman-Brown Split Half
Correlation ................................................................................................................................................................................ 70
Table 29 Reliability of the Slope for CBMreading by Passage using multi-level analyses............................ 71
Table 30 CBMreading Reliability of the Slope - Disaggregated Data................................................................... 72
Table 31 Demographics for Criterion-Related Validity Sample for earlyReading Composite Scores ...... 74
Table 32 Sample-Related Information for Criterion-Related Validity Data (earlyReading).......................... 74
Table 33 Concurrent and Predictive Validity for all earlyReading measures .................................................... 76
Table 34. Criterion Validity of Spring earlyReading Composite (Updated weighting scheme) with
Spring aReading: MN LEA 3 (Spring Data Collection)................................................................................................ 77
Table 35 Predictive Validity of the Slope for All earlyReading Measures ........................................................... 78
Table 36 Discriminant Validity for Kindergarten earlyReading Measures.......................................................... 78
Page | 9
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Table of Contents
Table 37. Discriminant Validity for First Grade earlyReading Subtests ............................................................... 79
Table 38 Concurrent and Predictive Validity for CBMreading................................................................................ 81
Table 39. Criterion Validity of Spring CBMreading with Spring CRCT in Reading: GA LEA 1 (Spring Data
Collection) ................................................................................................................................................................................. 82
Table 40. Criterion Validity of Spring CBMreading with Spring MCA-III in Reading: MN LEA 4 (Spring
Data Collection)....................................................................................................................................................................... 83
Table 41. Criterion Validity of Spring CBMreading with Spring MCA-III in Reading: MN LEA 3 (Spring
Data Collection)....................................................................................................................................................................... 83
Table 42. Criterion Validity of Spring CBMreading with Spring Minnesota Comprehensive Assessment
III (MCA-III) in Reading: MN LEA 2 (Spring Data Collection)..................................................................................... 84
Table 43. Criterion Validity of Spring CBMreading on Spring MAP in Reading: WI LEA 1 (Spring Data
Collection) ................................................................................................................................................................................. 84
Table 44. Criterion Validity of Spring CBMreading with Spring Massachusetts Comprehensive
Assessment (MCA): MA LEA 1 (Spring Data Collection) ............................................................................................ 85
Table 45 Predictive Validity for the Slope of Improvement by CBMreading Passage Level........................ 85
Table 46 Correlation Coefficients between CBMreading Slopes, AIMSweb R-CBM, and DIBELS Next... 86
Table 47 School Data Demographics for aReading Pilot Test ................................................................................ 90
Table 48 Summarization of K–5 aReading Parameter Estimates by Domain.................................................... 90
Table 49 Item Difficulty Information for K-5 aReading Items ................................................................................. 91
Table 50 School Demographics for Field-Based Testing of aReading Items..................................................... 91
Table 51 Sample Sizes for K-5 aReading Field-Testing by Grade and School ................................................... 92
Table 52 Descriptive Statistics of K–12 aReading Item Parameters...................................................................... 96
Table 53 Demographics for Criterion-Related Validity Sample for GMRT-4th and aReading....................... 97
Table 54 Sample-Related Information for aReading Criterion-Related Validity Data.................................... 98
Table 55 Correlation Coefficients between GMRT-4th and aReading Scaled Score ........................................ 98
Table 56 Content, Construct, and Predictive Validity of aReading....................................................................... 99
Table 57. Criterion Validity of Spring aReading with Spring Minnesota Comprehensive Assessment III
(MCA-III) in Reading: MN LEA 1 (Spring Data Collection) ......................................................................................... 99
Table 58. Criterion Validity for Spring aReading with Spring MCA-III in Reading: MN LEA 4 (Spring Data
Collection) ...............................................................................................................................................................................100
Table 59. Criterion Validity for Spring aReading with Spring MCA-III in Reading: MN LEA 3 (Spring Data
Collection) ...............................................................................................................................................................................100
Table 60. Criterion Validity of Spring aReading with Spring CRCT in Reading: GA LEA 1 (Spring to
Spring Prediction).................................................................................................................................................................100
Table 61.Criterion Validity of Spring aReading with Spring Massachusetts Comprehensive Assessment
(MCA): MA LEA 1 (Spring Data Collection)...................................................................................................................101
Table 62 Kindergarten Diagnostic Accuracy for earlyReading Measures.........................................................102
Table 63 First Grade Diagnostic Accuracy for earlyReading Measures .............................................................103
Table 64. Diagnostic Accuracy of Fall earlyReading Concepts of Print Subtest with Winter aReading:
MN LEA 3 (Fall to Winter Prediction)..............................................................................................................................104

Page | 10
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Table of Contents
Table 65. Diagnostic Accuracy of Fall earlyReading Onset Sounds Subtest with Winter aReading: MN
LEA 3 (Fall to Winter Prediction)......................................................................................................................................104
Table 66. Diagnostic Accuracy of Fall earlyReading Letter Names Subtest with Winter aReading: MN
LEA 3 (Fall to Winter Prediction)......................................................................................................................................104
Table 67. Diagnostic Accuracy of Fall earlyReading Letter Sounds Subtest with Winter aReading: MN
LEA 3 (Fall to Winter Prediction)......................................................................................................................................105
Table 68. Diagnostic Accuracy of Fall earlyReading Letter Sounds Subtest with Spring aReading: MN
LEA 3 (Fall to Spring Prediction)......................................................................................................................................105
Table 69. Diagnostic Accuracy of Winter earlyReading Letter Sounds Subtest with Spring aReading: MN
LEA 3 (Winter to Spring Prediction) ...............................................................................................................................105
Table 70. Diagnostic Accuracy of Winter earlyReading Rhyming Subtest with Spring aReading: MN LEA
3 (Winter to Spring Prediction)........................................................................................................................................105
Table 71. Diagnostic Accuracy of Fall earlyReading Word Segmenting Subtest with Winter aReading:
MN LEA 3 (Fall to Winter Prediction)..............................................................................................................................106
Table 72. Diagnostic Accuracy of Fall earlyReading Nonsense Words Subtest with Winter aReading: MN
LEA 3 (Fall to Winter Prediction)......................................................................................................................................106
Table 73. Diagnostic Accuracy of Fall earlyReading Sight Words Subtest with Winter aReading: MN LEA
3 (Fall to Winter Prediction) ..............................................................................................................................................106
Table 74. Diagnostic Accuracy of Fall earlyReading Sentence Reading Subtest with Winter aReading:
MN LEA 3 (Fall to Winter Prediction)..............................................................................................................................106
Table 75. Diagnostic Accuracy of Fall earlyReading Sentence Reading Subtest with Spring aReading:
MN LEA 3 (Fall to Spring Prediction)..............................................................................................................................107
Table 76. Diagnostic Accuracy of Winter earlyReading Sentence Reading Subtest with Spring
aReading: MN LEA 3 (Winter to Spring Prediction) ..................................................................................................107
Table 77. Diagnostic Accuracy of Winter earlyReading Composite with Winter aReading: MN LEA 3
(Fall to Winter Prediction)..................................................................................................................................................107
Table 78. Diagnostic Accuracy of Fall earlyReading Composite with Spring aReading: MN LEA 3 (Fall to
Spring Prediction).................................................................................................................................................................107
Table 79. Diagnostic Accuracy of Winter earlyReading Composite with Spring aReading: MN LEA 3
(Winter to Spring Prediction)............................................................................................................................................108
Table 80. Diagnostic Accuracy of Fall earlyReading Composite (2014–15 Weights) with Spring
aReading: MN LEA 3 (Fall to Spring Prediction).........................................................................................................108
Table 81. Diagnostic Accuracy of Winter earlyReading Composite (2014-15 Weights) with Spring
aReading: MN LEA 3 (Winter to Spring Prediction) ..................................................................................................108
Table 82 Diagnostic Accuracy by Grade Level for CBMreading Passages........................................................109
Table 83 Diagnostic Accuracy for CBMreading and MCA III..................................................................................110
Table 84. Diagnostic Accuracy on Fall CBMreading with Spring CRCT in Reading: GA LEA 1 (Fall to
Spring Prediction).................................................................................................................................................................111
Table 85. Diagnostic Accuracy on Winter CBMreading on Spring CRCT in Reading: GA LEA 1 (Winter to
Spring Prediction).................................................................................................................................................................111

Page | 11
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Table of Contents
Table 86. Diagnostic Accuracy of Fall CBMreading with Spring MCA-III in Reading: MN LEA 3 (Fall to
Spring Prediction).................................................................................................................................................................112
Table 87. Diagnostic Accuracy for Fall CBMreading with Spring Minnesota Comprehensive Assessment
III (MCA-III) in Reading: MN LEA 2 (Fall to Spring Prediction)................................................................................112
Table 88. Diagnostic Accuracy for Winter CBMreading with Spring Minnesota Comprehensive
Assessment III (MCA-III) in Reading: MN LEA 2 (Winter to Spring Prediction) ................................................113
Table 89. Diagnostic Accuracy for Winter CBMreading with MCA-III in Reading: MN LEA 3 (Winter to
Spring Prediction).................................................................................................................................................................113
Table 90. Diagnostic Accuracy of Winter CBMreading with Spring Massachusetts Comprehensive
Assessment (MCA): MA LEA 1 (Winter to Spring Prediction) ................................................................................114
Table 91 Diagnostic Accuracy statistics for aReading and GMRT-4th .................................................................115
Table 92 Diagnostic Accuracy Statistics for aReading and MAP..........................................................................116
Table 93 Diagnostic Accuracy for aReading and MCA-III .......................................................................................116
Table 94. Diagnostic Accuracy of Spring aReading with Spring MAP in Reading: WI LEA 1 (Spring Data
Collection) ...............................................................................................................................................................................117
Table 95. Diagnostic Accuracy of Fall aReading with Spring MCA-III in Reading: MN LEA 3 (Fall to Spring
Prediction)...............................................................................................................................................................................118
Table 96. Diagnostic Accuracy of Winter aReading with Spring MCA-III in Reading: MN LEA 3 (Winter to
Spring Prediction).................................................................................................................................................................118
Table 97. Diagnostic Accuracy Fall aReading with Spring Massachusetts Comprehensive Assessment
(MCA): Cambridge, MA (Fall to Spring Prediction) ...................................................................................................119
Table 98. Diagnostic Accuracy of Winter aReading with Spring Massachusetts Comprehensive
Assessment (MCA): MA LEA 1 (Winter to Spring Prediction) ................................................................................119
Table 99. Diagnostic Accuracy of Fall aReading with Spring CRCT in Reading: GA LEA 1 (Fall to Spring
Prediction)...............................................................................................................................................................................120
Table 100. Diagnostic Accuracy of Winter aReading with Spring CRCT in Reading: GA LEA 1 (Winter to
Spring Prediction).................................................................................................................................................................120
Table 101. Diagnostic Accuracy of Winter aReading with Spring Criterion-Referenced Competency
Tests (CRCT) in Reading: Georgia LEA 1 (Winter to Spring Prediction) .............................................................121
Table 102. Diagnostic Accuracy of Fall aReading with Spring Minnesota Comprehensive Assessment III
(MCA-III) in Reading: MN LEA 2 (Fall to Spring Prediction) ....................................................................................121
Table 103. Estimates of the Increase in the Percentage of Students who are Proficient or above with
the Implementation of Formative Assessment (Kingston & Nash, 2011, p. 35).............................................124
Table 104. FAST™ Statistical Significance and Effect Sizes.....................................................................................125
Table 105. Summary of Diagnostic Accuracy AUC Statistics and Validity Evidence.....................................146

Page | 12
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction

Section 1. Introduction to FAST™ and FastBridge Learning
This document provides a brief overview of FastBridge Learning and a detailed description of the
Formative Assessment System for Teachers™ (FAST™) measures. This document is partitioned into six
major sections:


Introduction to FAST™ and FastBridge Learning



Reading Measures



Math Measures



Social-Emotional-Behavioral Measures



Early Childhood and School Readiness



FAST™ as Evidence-Based Practice

The introduction and measurement sections are organized into chapters: (1) Overview, Purpose, and
Description, (2) Development, (3) Administration and Scoring, (4) Interpretation of Test Results, (5)
Reliability, (6) Validation, and (7) Diagnostic Accuracy of Benchmarks.

Chapter 1.1: Overview, Purpose, and Description
FAST™ was developed by researchers as a cloud-based system for teachers and educators.
Background
FAST™ assessments were developed by researchers at universities from around the country, which
include the Universities of Minnesota, Georgia, Syracuse, East Carolina, Buffalo, Temple, and Missouri.
FAST™ cloud-based technology was developed to support the use of those assessments for learning.
Although there is a broad set of potential uses, the system was initially conceptualized to make it
easier for teachers (see the Forward for more information).
FAST™ is designed for use within Multi-Tiered Systems of Support (MTSS) and Response to
Intervention (RTI) frameworks for early intervention and prevention of deficits and disabilities. It is
research- and evidence-based. FAST™ is distinguished and trusted by educators. It is transforming
teaching and learning for educators and kids nationwide.
All in One
FAST™ is one, comprehensive, simple cloud-based system with CurriculumBased Measurement (CBM) and Computer-Adaptive Tests (CAT) for universal
screening, progress monitoring, MTSS/RTI support, online scoring, and
automated reporting. It is easy to implement with online training and
resources, automated rostering and SIS integration, nothing to install or
maintain, and multi-platform and device support.

Page | 13
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction

Support and Training
Our school support team is accessible and responsive for support via live
chat, e-mail, or phone. When combined with our knowledge base—full of
quick tips, articles, videos, webinars, and flipped training for staff—in
addition to customized online or onsite training, your teachers and
administration are supported at every step.

Trusted Results
FAST™ is an evidence-based formative assessment system that was
developed by researchers at the University of Minnesota in cooperation
with others from around the country. They set out to offer teachers an
easier way to access and use the highest quality formative assessments.
Researchers and developers are continuously engaged with teachers and
other users to refine and develop the best solutions for them. (e.g., better
data, automated assessments, and sensible reports).

Curriculum-Based Measurement (CBM)
Our Curriculum-Based Measures (CBM) are highly sensitive to growth over
brief periods. We offer Common Core-aligned CBM measures with online
scoring and automated skills analysis in earlyReading and earlyMath (K-1),
CBMreading, CBMcomprehension, and CBMmath (1-6).

Automated Assessments
Our Computer-Adaptive Tests (CAT) provide a reliable measure of broad
achievement and predict high-stakes test outcomes with high accuracy.
Automatically adapting to students’ skill levels to inform instruction and
identify MTSS/RtI grouping, we offer aReading (K–12), aMath (K–6), and
Standards-Based Math (6–8).

Page | 14
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction
Prevention and Intervention
Designed for Multi-Tiered Systems of Support (MTSS) and Response to
Intervention (RTI), FAST™ makes program implementation easy and
efficient with automated scoring, analysis, norming and reporting;
customizable screening, benchmarking, instructional recommendations,
and progress monitoring.

Chapter 1.2: Development
FastBridge Learning has a strong foundation in both research and theory. FAST™ assessments were
created to provide a general estimate of overall achievement in reading and math, as well as provide a
tool to identify students at risk for emotional and behavioral problems. For reading and math
assessments, item banks have been created containing a variety of items, including those with
pictures, words, individual letters and letter sounds, sentences, paragraphs, and combinations of these
elements. Overall, FastBridge Learning aims to extend and improve on the quality of currently
available assessments.

Chapter 1.3: Administration and Scoring
FAST™ is supported by an extensive set of materials to support teachers and students, including selfdirected training modules that allow teachers to become certified to administer each of the
assessments. FAST™ assessments can be administered by classroom teachers, special education
teachers, school psychologists, and other individuals such as paraprofessionals with usually less than
an hour of training. Administration time varies depending on which assessment is being administered.
Online administrations require a hard copy of the student materials (one copy per student) and access
to the FAST™ system (i.e., iPad or computer with Internet connection). Paper-and-pencil assessment
administration materials and instructions are available upon request. As with any assessment, only
students who can understand the instructions and can make the necessary responses should be
administered FAST™ assessments. Assessments should be administered in a quiet area conducive to
optimal performance. The brevity of FAST™ assessments aims to minimize examinee fatigue, anxiety,
and inattention. For the majority of assessments, FAST™ produces automated reports summarizing
raw scores, percentile scores, developmental benchmarks, subscale and subtest scores, and composite
scores. The online system provides standardized directions and instructions for the assessment
administrator.
Setting Standards
Overall, FastBridge Learning uses standard-setting processes to summarize student performance.
Standards may be used to inform goal setting, identify instructional level, and evaluate the accuracy of
student performance. For the purpose of this technical manual, standards are the content or skills that
are expected (content standards), which are often defined by a score for purposes of measurement
(performance standards). A number of terms are used to reference performance standards, including:
benchmarks, cut scores, performance levels, frustrational, instructional or mastery levels, and
Page | 15
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction
thresholds. These terms each reference categories of performance with respect to standards and are
used throughout the technical manual. The method of standard setting is described below.

Chapter 1.4: Interpretation of Test Results
The FastBridge Learning software provides various resources to assist administrators with test result
interpretations. For example, a Visual Conventions drop down menu is available to facilitate
interpretation of screening and progress monitoring group and individual reports. Percentiles are
calculated for local school norms unless otherwise indicated. Local school norms compare individual
student performances to their same grade and school peers. For example, a student in the 72nd
percentile performed as well or better than 72 percent of his or her grade level peers at that school.
Methods of notation are also included to provide information regarding those students predicted to
be at risk. Exclamation marks (! and !!) indicate the level of risk based on national norms. One
exclamation mark refers to some risk, whereas two exclamation marks refer to high risk of reading
difficulties or not meeting statewide assessments benchmarks, based on the score. Interpreting
FastBridge Learning assessment scores involves a basic understanding of the various scores provided
in the FAST™ system and helps to guide instructional and intervention development. FAST™ includes
individual, class, and grade level reports for screening, and individual reports for progress monitoring.
Additionally, online training modules include sections on administering the assessments, interpreting
results, screen casts, and videos.
Results should always be interpreted carefully considering reliability and validity of the score, which is
influenced by the quality of standardized administration and scoring. It important to consider the
intended purpose of the assessment, its content, the stability of performance over time, scoring
procedures, testing situations, or the examinee. The FAST™ system automates analysis, scoring,
calculations, reporting and data aggregation. It also facilitates scaling and equating across screening
and progress monitoring occasions.
Standard Setting
It is necessary to address questions such as, “How much skill/ability defines proficiency?” There are
many methods used for standards setting; however, human judgment is inherent to the process
(Hambleton & Pitoniak, 2006) because some person(s) decide “how many” or “how much is enough”
to meet a standard. Because judgment is involved, there are some criticisms that standard setting is
arbitrary (Glass, 1978) and the results of standard setting are very often the source of debate and
scrutiny. The Standards for Educational and Psychological Testing (AERA, APA & NCME, 199) define the
basic requirements to set standards and therein recognize the role of human judgment: “cut scores
embody value judgments as well as technical and empirical considerations” (p. 54). The standard
setting process in designed to ensure those value judgments are well-informed. The Standards along
with the professional literature (e.g., Hableton & Pitoniak, 2006; Swets, Dawes & Monahan, 2000) guide
the standard setting processes for FAST™. A brief description of relevant concepts and methods are
below.

Page | 16
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction
Kane (1994, 2006, 2013) suggests that the rationale and reasons for the selected standard-setting
procedure are often the most relevant and important source of evidence for the validity of standards
to interpret scores. The method should be explicit, practical for the intended interpretation and use,
implemented with fidelity, and documented (Hambleton & Pitoniak, 2006). The convergence of
standards with other sources of information, such as criterion measures, also contributes to validation;
however, such evidence is often limited because the quality and standards from external sources are
often just as limited (Kane, 2001). Moreover, external sources, such as criterion measures, are often
unavailable or misaligned with the experimental measure or intended use for the standard.

Standard Setting Methods
There are methods to set relative or absolute standards. Norm-referenced methods are most familiar
to the general public. They are used to set a relative standard such that a particular proportion of a
population is above or below the standard. For example, if the standard is set at the 40th percentile
then 39% of the population is below and 60% is at or above. Norm-referenced standards are relative to
the performances in the population. As noted by Sereci (2005), “scores are interpreted with respect to
being better or worse than others, rather than with respect to the level of competence of a specific
test taker” (p. 118). Norm-referenced standards are used in FAST™ to guide resource allocation. Gradelevel norms are provided for the class, school, district, and nation.
Absolute- or criterion-referenced methods are less familiar to the general public. They are used to
define “how much is enough” to be above or below a standard. For example, if the standard is that
students should identify all of the letters in the alphabet with 100% accuracy then all of the students
in a particular grade might be above or below that standard. These methods often rely on the
judgment of experts who classify items, behaviors, or individual persons. Those judgments are used to
define the standard. For example, the expert is asked to consider a person whose performance is just
at the standard. They then estimate the probability that person would exhibit a particular behavior or
response correctly to a particular item. Another approach is to have that expert classify individuals as
above or below the standard. Once classified, the performance of the individuals is analyzed to define
the standard. The particular method is carefully selected based on the content and purpose of the
measure and standard. Careful selection of experts and panels, training, procedures, validation and
documentation are all important components of those expert-based approaches.
Norm-Referenced Standards

Norm-referenced methods are used to set a relative standard such that a particular proportion of a
population is above or below the standard. For example, if the standard is set at the 40th percentile
then 39% of the population is below and 60% is at or above. Norm-referenced standards are relative to
the performances in the population. As noted by Sereci (2005), “scores are interpreted with respect to
being better or worse than others, rather than with respect to the level of competence of a specific
test taker” (p. 118). Norm-referenced standards are used in FAST™ to guide resource allocation. Gradelevel norms are provided for the class, school, district and nation.

Page | 17
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction
Norm-Referenced Interpretations
FAST™ calculates and reports the percentile
ranks, or percentiles, of scores relative to
same-grade peer performance in the class,
school, district, and FAST™ users around
the nation. Those percentiles are classified
and color-coded in bands: < 19.99th (red),
20th to 29.99th (orange), 30th to 84.99th
(green) and > 85th percentiles (blue). These
standards were set to guide resource
allocations for early intervention and
prevention within multi-tiered systems of
support (MTSS).
Most schools can provide supplemental and intensive supports for students at risk and enrichment for
the highest achieving students. Schools rarely have resources to support more than 30% of at-risk
learners with supplemental and intensive supports; even if a larger proportion would benefit (Christ,
2008; Christ & Arañas, 2014). The norm-referenced standards are applied to each norm group to
support decisions by individual teachers (class norms), school-based grade level teams (school norms)
and district-wide grade level teams (district norms) as to which students receive supports.
The percentiles and standards should be used to identify the individuals who will receive
supplemental support. The proportion of the population who receive supports depends on the
availability of resources. For example, one school might provide supplemental support to all students
below the 30th percentile (red and orange). Another school might provide supplemental support to all
students below the 20th percentile (red), but monitor those below the 30th percentile (orange). These
are local decisions that should be determined in consideration of the balance between student needs
and system resources.
National norms are used to compare local performance to that of an external group. The standards
(color codes) are applied to support decisions about core and system-level supports. Visual analysis of
color codes are useful to estimate the typicality of achievement in the local population. They are often
used in combination with benchmarks to guide school and district level decisions about instruction,
curriculum and system-wide services (e.g., are the school-wide core reading services sufficient to
prevent deficit achievement for 80% of students). If FAST™ data indicate that much more than 20% of
a school or district’s students are below the 20th percentile on national norms, then remediation efforts
in that area should be considered as the data suggest that the core instruction is not supporting
adequate achievement. If they observe that fewer than 20% of the total school population are below
the 20th percentile on national norms, their population is over-performing relative to others.
Subsequently, the school should continue using effective services, but identify another domain of
focus.

Page | 18
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction

Criterion-Referenced Standards (Benchmarks)
Absolute or criterion-referenced methods are used to define “how much is enough” to be above or
below a standard. For example, if the standard is that students should identify all of the letters in the
alphabet with 100% accuracy then all of the students in a particular grade might be above or below
that standard. These methods often rely on the judgment of experts who classify items, behaviors, or
individual persons. Those judgments are used to define the standard. For example, the expert is asked
to consider a person whose performance is just at the standard. They then estimate the probability
that person would exhibit a particular behavior or response correctly to a particular item. Another
approach is to have that expert classify individuals as above or below the standard. Once classified,
the performance of the individuals is analyzed to define the standard. The particular method is
carefully selected based on the content and purpose of the measure and standard. Careful selection of
experts and panels, training, procedures, validation, and documentation are all important
components of those expert-based approaches.
FAST™ reports provide tri-annual grade-level benchmarks, which generally correspond with the 15th
and 40th percentiles on national norms. Scores below the 15th percentile are classified as “high-risk.”
Those at-or-above the 15th and below the 40th are “some risk;” and those at or above the 40th are “low
risk.” This is consistent with established procedures and published recommendations (e.g., RTI
Network). It is common practice to use norm-reference standards at the 15th and 40th percentiles; or to
use pre-determined standards on state achievement tests. As quoted from the RTI Network:

“Reading screens attempt to predict which students will score poorly on a future
reading test (i.e., the criterion measure). Some schools use norm-referenced test scores
for their criterion measure, defining poor reading by a score corresponding to a
specific percentile (e.g., below the 10th, 15th, 25th, or 40th percentile). Others define
poor reading according to a predetermined standard (e.g., scoring below “basic”) on
the state’s proficiency test. The important point is that satisfactory and unsatisfactory
reading outcomes are dichotomous (defined by a cut-point on a reading test given
later in the students’ career). Where this cut-point is set (e.g., the 10th or 40th
percentile) and the specific criterion reading test used to define reading failure (e.g., a
state test or SAT 10) greatly affects which students a screen seeks to identify”
(retrieved on 1-24-15 from
http://www.rtinetwork.org/essential/assessment/screening/readingproblems)
The procedure used by FAST™ is described in more detail below. Again, FAST™ establishes
benchmarks that approximate the 15th and 40th percentiles on national norms. This report provides
additional evidence on the correspondence with those standards and proficiency on state tests.
Interpreting Criterion-Referenced Standards
Benchmarks are often used to discern whether students are likely to perform sufficiently on a highstakes assessment, such as a state test. FastBridge Learning will estimate specific benchmarks for
states and districts if their state test data are provided (help@fastbridge.org). Another way to interpret
Page | 19
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction
benchmarks is to consider them the minimal level of performance that is acceptable. Anything less
places the student at risk. These standards should be met for all students. They are not based on the
distribution of performance unlike norms so all students can truly meet benchmark standards.
If more than 30% of students are below the “some risk” benchmark standard, then it is necessary to
modify core instruction and general education instruction to better serve all students. This is the most
efficient approach to remediate widespread deficits. If fewer than 15% of are below the “some risk”
benchmark standard in a specific content area, then core instruction is highly effective. It should be
maintained and other content areas should be considered the focus. Schools often focus on reading
and behavior and then move to math and other content areas as they achieve benchmark standards
for 85% of students in each domain.

Chapter 1.5: Reliability
Reliability refers to the stability with which a test measures the same skills across minimal differences
in circumstances. Nunnally and Bernstein (1994) offer a hierarchical framework for estimating the
reliability of a test, emphasizing the documentation of several forms of reliability. First and foremost,
alternate-form reliability with a two-week interval is recommended, assuming that alternate (but
equivalent) forms of the same test with different items should produce approximate scores. The
second recommended form of reliability is test-retest reliability, which also employs a two-week
interval of time. The same test administered at two different points in time (i.e., the difference of a
two-week interval) should produce approximately the same scores. Finally, inter-rater reliability is
recommended and may be evaluated by comparing scores obtained for the same student by two
different examiners. For many FastBridge Learning assessments, there is no threat to inter-rater
reliability because assessments are electronically scored. For the purpose of this technical manual,
error refers to unintended factors that contribute to changes in scores. Other forms of reliability
evidence include internal consistency (the extent to which different items measure the same general
construct and produce similar scores), and reliability of the slope (the ratio of true score variance to
total variance).
Overall, FastBridge Learning assessments show evidence of reliability coefficients that show promise
for producing little test error. Further, evidence supports the use of FastBridge Learning measures for
screening and progress monitoring, and for informing teachers of whether instructional practices
have been effective or if more and what kind of instruction may be necessary to advance student
growth in reading and math skills. Educators can be confident that the FastBridge Learning
assessments provide meaningful instructional information that can be quickly and easily interpreted
and applied to impact student learning. Current research on FastBridge Learning assessments is
encouraging, suggesting that these assessments may be used to reliably differentiate between
students who are or are not at risk for reading problems, math problems, or behavioral or emotional
problems.

Page | 20
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction

Chapter 1.6: Validity
To validate an interpretation or use of test scores is to evaluate the plausibility of the claims based on
those scores (Kane, 2013). According to Kane (2013), interpretations and uses can change over time in
response to evolving needs and new understandings. Additionally, consequences of the proposed
uses of a test score need to be evaluated. Validity refers to the extent to which evidence and theory
support the interpretations of test scores. Types of validity discussed in this technical manual are
content, criterion, predictive, and discriminant validity.
Content validity is the extent to which a test’s items represent the domain or universe intended to be
measured. Criterion-related validity is the extent to which performance on a criterion measure can be
estimated from performance on the assessment procedure being evaluated. Predictive validity is the
extent to which performance on a criterion measure can be estimated from performance across time
on the assessment being evaluated. Finally, discriminant validity is a measure of how well an
assessment distinguishes between two groups of students at different skill levels.
Establishing validity evidence of FastBridge Learning assessments is ongoing. Studies will continue to
provide information regarding the content and construct validity of each assessment. Validity
evidence will be interpreted as data is disaggregated across gender, racial, ethnic, and cultural groups.
All FastBridge Learning assessments were designed to be sensitive to student growth while also
providing instructionally relevant information. Current research supports the validity of FastBridge
Learning assessments across reading, math, and behavioral domains.

Chapter 1.7: Diagnostic Accuracy of Benchmarks
Campbell and Ramey (1994) acknowledged the importance of early identification through the use of
effective screening measures and intervention with those students in need. Early identification,
screening, and intervention have been shown to improve academic and social-emotional/behavioral
outcomes (Severson, Walker, Hope-DooLittle, Kratochwill, & Gresham, 2007). Effective screening is a
pre-requisite for efficient service delivery in a multi-tiered Response to Intervention (RTI) framework
(Jenkins, Hudson, & Johnson, 2007). RTI seeks to categorize students accurately as being at risk or not
at risk for academic failure. Inaccurate categorization can lead to consequences such as ineffective
allocation of already minimal resources.

Page | 21
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction
A Conceptual Explanation: Diagnostic Accuracy of Screeners
Within medicine, a diagnostic test can be used to determine the presence or absence of a disease. The
results of a screening device are compared with a “gold standard” of evidence. For instance, a doctor
may administer an assessment testing whether a tumor is malignant or benign. Based on a gold
standard, or later diagnosis, we can estimate how well the screener identifies cases in which the
patient truly has the ailment and cases in which he or she does not. When using any diagnostic test
with a gold standard there are four possible outcomes: the test classifies the tumor as malignant when
in fact it is malignant (True Positive; TP), the test classifies the tumor as not malignant when in fact it is
not malignant (True Negative; TN), the test classifies the tumor as malignant when in fact it is benign
(False Positive; FP), and the test classifies
the tumor as benign when in fact it is
malignant (False Negative; FN). The rates of
each classification are directly tied to the
decision threshold, or cut-off score, of the
screening measure. The cut-off score is the
score at which a subject is said to be
symptomatic or not symptomatic. The
decision regarding placement of the
decision threshold is directly tied to the
implications of misclassifying a person as
symptomatic versus not-symptomatic
(Swets, Dawes, & Monahan, 2000). In the
case of the tumor, a FN may mean that a
patient does not undergo a lifesaving
procedure. Conversely, a FP may cause undue stress and financial expense for treatments that aren’t
needed.
Decisions that Guide Benchmarks Selection: Early Intervention and Prevention
It should be apparent from a review of the illustration above that decisions based on the screener are
inherently imperfect. The depiction in that particular figure illustrates a correlation of approximately
.70 between the predictor and criterion measure. In this example, CBMreading is an imperfect
predictor of performance on the state test. Regardless of the measures, there will always be an
imperfect relationship. This is also true for test-retest and alternate-form reliability (i.e., performance
on the same test on two occasions). Tests are inherently unreliable, and all interpretations of scores
are tentative. This is especially true for screening assessments, which are designed to be highly
efficient and, therefore, often have less reliability and validity than a more comprehensive, albeit
inefficient, assessment.
For the purposes of screening in education, students who are in need of extra help may be overlooked
(FN), and students who do not need extra help may receive unneeded services (FP). The performance
of diagnostic tests, and corresponding decision thresholds, can be measured via sensitivity, specificity,

Page | 22
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction
positive predictive value, negative predictive value, and area under the curve (AUC). All of the
following definitions are based on the work of Grzybowski and Younger (1997).
i.
ii.
iii.
i.
ii.

Sensitivity: The probability of a student testing positive given the true presence of a
difficulty. (TP / TP + FN)
Specificity: The probability of a student testing negative if the difficulty does not exist.
(TN / TN + FP)
Positive Predictive Power: The proportion of truly struggling students among those with
positive test results. (TP / TP + FP)
Negative Predictive Power: The proportion of truly non-struggling students among all
those with negative test results. (TN / TN + FN)
Area Under the Curve (AUC): Quantitative measure of the accuracy of a test in
discriminating between students at-risk and not at-risk across all decision thresholds.

Previous research in school psychology (e.g., Hintze & Silberglitt, 2005; VanDerHeyden, 2011) derives
decision thresholds by iteratively computing specificity and sensitivity at different cut scores.
Precedence would be given to maximize each criterion by computing sensitivity and specificity for
each point. A more psychometrically sound and efficient method is to compute scores via a receiver
operating characteristic (ROC) curve analysis.
Area Under the Curve (AUC)
Area Under the Curve (AUC) is used as a measure of predictive power. It is obtained by calculating the
sensitivity and specificity values for all possible cutoff points on the screener by fixing a cutoff point
on criterion measure and plotting specificity (or TPP) against sensitivity (or TNP). AUC is expected to
be .5 if the screener provided little or no information. AUC is expected to be 1 for a perfect diagnostic
method to identify the students at risk correctly. Although the criteria that are applied to interpret
AUCs are variable, values are considered excellent (.90 to 1.0), good (.80 to .89), fair (.70 to 79), or poor
(< .69). It seems reasonable and generally consistent with the standards outlined by the National
Center for Response to Intervention that an AUC of at least .85 is required for low-stakes decisions and
that an AUC of at least .90 is required for high-stake decisions.
Decision Threshold: Benchmark
A decision threshold is established to maximize the benefits of the decision process relative to its costs
(Swets, Dawes, & Monahan, 2000). That threshold is adjusted to establish a neutral, lenient, or strict
classification criterion for the predictor. A neutral threshold will balance the proportion of TP and FP,
although not all thresholds should be balanced. For example, screening measures for reading often
over-identify students (increase the rate of TP as well as FP) to ensure that fewer positive cases are
missed. This is a rational choice, because failure to identify TP outweighs the consequences of
increased FP.
Thresholds that are more lenient (over-identify) increase sensitivity, thereby increasing the proportion
of positive classifications (both TP and FP). Thresholds that are more strict (under-identify) increase
Page | 23
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 1. Introduction
specificity, thereby increasing the proportion of negative classifications (both TN and FN; Swets et al.,
2000). The decision threshold is adjusted to obtain the optimal ratio of positive and negative
classifications along with that of true and false classifications. For example, Silberglitt and Hintze
(2005) systematically modified the CBMreading benchmark scores in Third Grade to optimize the cut
score, which improved classification accuracy. In general, FAST™ uses a procedure to balance
sensitivity and specificity so that benchmarks neither over- nor under-identify individuals.
FastBridge Learning assessments predict performance on state accountability tests, including tests
administered in Iowa, Illinois, Vermont, Indiana, New York, Colorado, Minnesota, Georgia,
Massachusetts, and Wisconsin. For diagnostic accuracy analyses, cut scores were selected by
optimizing sensitivity at approximately .70, and then balancing it with specificity using methods
presented by Silberglitt and Hintze (2005). Overall, analyses suggest that current benchmarks for
FastBridge Learning assessments are appropriate, accurate, and reliable.
For a summary of Reading Diagnostic Accuracy statistics, see Appendix B: FastBridge Learning
Reading Diagnostic Accuracy.

Page | 24
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Section 2. Reading and Language
Chapter 2.1: Overview, Purpose, and Description
earlyReading
The earlyReading measure is designed to assess both unified and component skills associated with
Kindergarten and First Grade reading achievement. earlyReading is intended to enable screening and
progress monitoring across four domains of reading (Concepts of Print, Phonemic Awareness, Phonics,
and Decoding) and provide domain-specific assessments of these component skills as well as a
general estimate of overall reading achievement. earlyReading is an extension of CBMreading, which
was initially developed by Deno and colleagues to index the level and rate of reading achievement
(Deno, 1985; Shinn, 1989). The current version of earlyReading has an item bank that contains a variety
of items, including those with pictures, words, individual letters and letter sounds, sentences,
paragraphs, and combinations of these elements.
The research literature provides substantial guidance on instruction and assessment of alphabetic
knowledge, phonemic awareness, and oral reading. The objective of earlyReading measures is to
extend and improve on the quality of currently available assessments.

Aspects of Reading measured by earlyReading
Concepts of Print (COP)
COP is defined as the general understanding of how print works and how it can be used (Snow, Burns,
& Griffin, 1998). Concepts of print is the set of skills used in the manipulation of text-based materials,
which includes effective orientation of materials (directionality), page turning, identifying the
beginning and ending of sentences, identifying words, as well as identifying letters, sentences, and
sentence parts. Concepts of print are normally developed in the emergent literacy phase of
development and enable the development of meaningful early reading skills: “Emergent literacy
consists of skills, knowledge, and attitudes that are developmental precursors to conventional forms
of reading and writing” (Whitehurst & Lonigan, 1998). These skills typically develop from preschool
through the beginning of First Grade— with some more advanced skills that develop through Second
Grade, such as understanding punctuation, standard spelling, reversible words, sequence, and other
standard conventions of written and spoken language. Introductory level of logical and analytical
abilities as in understanding the concepts of print has an impact on early student reading
achievement (Adams, 1990; Clay, 1972; Downing, Ollila, & Oliver, 1975; Hardy et al., 1974; Harlin & Lipa,
1990; Johns, 1972; Johns, 1980; Lomax & McGee, 1987; Nichols et al., 2004; Tumner et al., 1988).
Phonemic Awareness (PA)
Phonemic Awareness involves the ability to identify and manipulate phonemes in spoken words
(National Reading Panel [NRP], 2000). Phonemes are the smallest units of sound in spoken language.
"Depending on what distinctions are counted, there are between 36-44 phonemes in English, which is
about average for languages" (Juel, 2006, p.418). According to Adams, “to the extent that children
Page | 25
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
have learned to ‘hear’ phonemes as individual and separable speech sounds, the system will, through
the associative network, strengthen their ability to remember or ‘see’ individual letters and spelling
patterns” (1990, p. 304). Hearing and distinguishing individual letter sounds comes last (Goswami,
2000). Children who manipulate letters as they are learning to hear specific sounds have been shown
to make better progress in early reading development than those who do not (NRP, 2000, p. 2-4).
Phonemic awareness skills are centrally involved in decoding by processes of blending and
segmenting phonemes (NRP, 2000). Phonemic awareness also helps children learn how to spell words
correctly. Phonemic segmentation is required to help children retain correct spellings in memory by
connecting graphemes (printed letters) to phonemes (NRP, 2000).
Phonics
Phonics is the set of skills readers use to identify and manipulate printed letters (graphemes) and
sounds (phonemes). It is the correspondences between spoken and written language. This connection
between letters, letter combinations, and sounds enable reading (decoding) and writing (encoding).
Phonics skill development “involves learning the alphabetic system, that is, letter-sound
correspondences and spelling patterns, and learning how to apply this knowledge” to reading (NRP,
2000b).
Decoding
“Decoding ability is developed through a progression of strategies sequential in nature: acquiring
letter-sound knowledge, engaging in sequential decoding, decoding by recognizing word patterns,
developing word accuracy in word recognition, and developing automaticity and fluency in word
recognition” (Hiebert & Taylor, 2000, p. 467). When a child has a large and established visual lexicon of
words in combination with effective strategies to decode unfamiliar words, he/she can read fluently—
smoothly, quickly, and more efficiently (Adams, 1990; Snow et al., 1998). The reader can also focus
his/her attention on monitoring comprehension: “If there are too many unknown words in the
passage that require the child to apply more analytic (phonemic decoding) or guessing strategies to
fill in the blanks, fluency will be impaired” (Phillips & Torgesen, 2006, p. 105). According to RAND,
“readers with a slow or an inadequate mastery of word decoding may attempt to compensate by
relying on meaning and context to drive comprehension, but at the cost of glossing over important
details in the text” (2002, p. 104). Decoding is often linked with phonics with the emphasis on lettersound knowledge. Vocabulary contains common characteristics with decoding such as recognizing
word patterns, as in prefixes and suffixes.

Uses and Applications
earlyReading consists of 12 different evidence-based assessments for screening and monitoring
student progress.


Concepts of Print



Word Rhyming



Onset Sounds



Word Blending



Letter Names



Word Segmenting



Letter Sounds



Decodable Words

Page | 26
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice


Nonsense Words



Sentence Reading



Sight Words-Kindergarten (50 words)



Oral Language (Sentence Repetition)



Sight Words-1st Grade (150 words)

There are recommended combinations of subtests for fall, winter, and spring screening aimed to
optimize validity and risk evaluation. Similarly, there are recommended combinations of subtests for
fall, winter, and spring for monitoring of progress. Supplemental assessments may be used to
diagnose and evaluate skill deficits. Results from supplemental assessments provide guidance for
instructional and intervention development. earlyReading is often used by teachers to screen all
students and to estimate annual growth with tri-annual assessments (fall, winter, & spring). Students
who progress at a typical pace through the reading curriculum meet the standards for expected
performance at each point in the year. Students with deficit achievement can be identified in the fall
of the academic year so that supplemental, differentiated, or individualized instruction can be
provided.
earlyReading is designed to accommodate quick and easy weekly assessments, which provide useful
data to monitor student progress and evaluate response to instruction. The availability of multiple
alternate forms for various subtests of earlyReading make it suitable for monitoring progress between
benchmark assessment intervals (i.e., fall, winter, and spring) for those students that require more
frequent monitoring of progress. Onset Sounds has 13 alternate forms, and the following subtests
have a total of 20 alternate forms: Letter Naming, Letter Sound, Word Blending, Word Segmenting,
Decodable Words, Sight Words, and Nonsense Words. Concepts of Print, Rhyming, and Sentence
Reading progress monitoring forms have not yet been developed.
Target Population
earlyReading is designed for all students in the early primary grade levels. This includes students in
Kindergarten through Third Grade. earlyReading subtests are most relevant for students in
Kindergarten and First Grade, but they have application to students in later grades who have yet to
master early reading skills.
CBMreading
Curriculum-Based Measures of Reading (CBMreading) is a particular version of Curriculum-Based
Measurement of Oral Reading fluency (CBM-R), which was originally developed by Deno and
colleagues to index the level and rate of reading achievement (Deno, 1985; Shinn, 1989). The tool is an
evidence-based assessment for use to screen and monitor student progress in reading competency in
primary grades (1–6). CBMreading uses easy, time-efficient assessment procedures to determine a
student’s general reading ability across short intervals of time (i.e., weekly, monthly, or quarterly).
Students read aloud for one minute from grade or instructional- level passages. The words read
correct per minute (WRCM) functions as a robust indicator of a reading health and a sensitive indicator
of intervention effects. CBMreading includes standardized administration and scoring procedures
along with proprietary instrumentation, which was designed and developed to optimize the
Page | 27
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
consistency of data collected across progress monitoring occasions. CBMreading provides teachers
with a direct link to instruction and allows them to determine if and when instructional adaptations
are needed, set ambitious but attainable goals for students, and monitor progress toward those goals
(Fuchs & Fuchs, 2002).
CBMreading emerged from a project funded by the Institute for Education Sciences in the US
Department of Education. That project was entitled Formative Assessment Instrumentation and
Procedures for Reading (FAIP-R), so they are sometimes described as the FAIP-R passages. Early
versions of those passages were used in published research (Ardoin & Christ, 2008; Christ & Ardoin,
2009). The goal in creating the CBMreading measures was to systematically develop, evaluate and
finalize research-based instrumentation and procedures for accurate, reliable, and valid assessment
and evaluation of reading rate.
For the remainder of the manual, CBM-R will refer to the general concept of Curriculum-Based
Measurement of Oral Reading while CBMreading will refer to the assessment in FastBridge Learning.

Aspects of Reading Measured by CBMreading
The Common Core State Standards for English Language Arts and Literacy in History/Social Studies,
Science and Technical Subjects (National Governors Association Center for Best Practices & Council of
Chief State School Officers, 2010) is a synthesis of information gathered from state departments of
education, assessment developers, parents, students, educators, and other pertinent sources to
develop the next generation of state standards of K–12 students to ensure that all students are college
and career literacy ready by the end of their high school education. This process is headed by the
Council of Chief State School Officers (CCSO) and the National Governors Association (NGA). The
Standards are an extension of a previous initiative by the CCSSO and NGA titled the College and
Career Readiness (CCR) Anchor Standards. The CCR Anchor Standards are numbered from one to ten.
The Standards related to fluency are found within Foundational Skills in Reading. These standards are
relevant to K–5 children and include the working knowledge of the following subcategories:
(1) Print Concepts: the ability to demonstrate the organization and basic feature of print.
(2) Phonological Awareness: demonstrate understanding of spoken words, syllables, and
sounds or phonemes.
(3) Phonics and Word Recognition: the skill of applying grade-level phonics and word analysis
skills in decoding words.
(4) Fluency: Reading on-level texts with sufficient purpose, accuracy, and fluency to support
comprehension.
Oral Reading Fluency
Reading involves simultaneous completion of various component processes. In order to achieve
simultaneous coordination across these component processes, instantaneous execution of each
component skill is required (Logan, 1997). Reading fluency is achieved so that performance is
speeded, effortless, autonomous, and achieved without much conscious awareness (Logan, 1997).
Oral reading fluency represents the automatic translation of letters into coherent sound
Page | 28
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
representations, unitizing those sound components into recognizable wholes, and automatically
accessing lexical representations, processing meaningful connections within and between sentences,
relating text meaning to prior information, and making inferences in order to supply missing
information. Logan (1997) described oral reading fluency as the complex orchestration of these skills,
establishing it as a reliable measure of reading expertise.
As previously mentioned, CBMreading is a particular version of an oral reading fluency measure.
CBMreading is an effective tool used to measure rate of reading. Indeed, reading disabilities are most
frequently associated with deficits in accurate and efficient word identification. Although reading is
not merely rapid word identification or the “barking at words” (Samuels, 2007), the use of rate-based
measures provide a general measure of reading that can alert teachers to students who have
problems and are behind their peers in general reading ability. Overall, CBMreading provides a global
indicator of reading.

Uses and Applications
CBMreading is an evidence-based assessment for use to screen and monitor students’ progress in
reading achievement in the primary grades (1–6). Each assessment is designed to be highly efficient
and give a broad indication of reading competence. The automated output of each assessment gives
information on the accuracy and fluency of passage reading which can be used to determine
instructional level to inform intervention. At the school level, student growth can be tracked and
monitored, allowing administrators to look at improvements both across grades and academic years
for the purpose of accountability. Teachers and administrators may use this information to help
parents better understand their children’s reading needs.
Target Population
CBMreading is designed for all students in grades 1 through 6. For elementary grades 2 through 6,
measures of fluency with connected text (curriculum-based measure of oral reading; CBM-R) are often
used as a universal screeners for grade-level reading proficiency. Although strong evidence exists in
the literature to support the use of CBM-R (Fuchs, Fuchs, & Maxwell, 1988; Kranzler, Brownell, & Miller,
1998; Markell & Deno, 1997), support for CBM-R as a universal screener for students who are not yet
reading connected text is less robust (Fuchs, Fuchs, & Compton, 2004; National Research Council,
1998). Thus, CBMreading may not be appropriate for students not yet reading connected text with
some degree of fluency. For those students not yet reading connected text with fluency, CBMreading
results and scores should be interpreted with caution.
aReading
The Adaptive Reading (aReading) assessment is a computer-adaptive measure of broad reading ability
that is individualized for each student. aReading provides a useful estimate of broad reading
achievement from Kindergarten through twelfth grade. The question-and-response format used in
aReading is substantially similar to many statewide, standardized assessments. Browser-based
software adapts and individualizes the assessment for each child so that it essentially functions at the

Page | 29
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
child’s developmental and skill level. The adaptive nature of the test makes it more efficient and more
precise than paper-and-pencil assessments.
The design of aReading has a strong foundation in both research and theory. During the early phases
of student reading development, the component processes of reading are most predictive of future
reading success (Stanovich, 1981, 1984, 1990; Vellutino & Scanlon, 1987, 1991; Vellutino, Scanlon,
Small, & Tanzman, 1991). Indeed, reading disabilities are most frequently associated with deficits in
accurate and efficient word identification. Those skills are necessary but not sufficient for reading to
occur. After all, reading is comprehending and acquiring information through print. It is not merely
rapid word identification or the “barking at words” (Samuels, 2007). As such, a unified reading
construct is necessary to enhance the validity of reading assessment and inform balanced instruction
throughout the elementary grades. aReading was developed based on a skills hierarchy and unified
reading construct (presented later in the technical manual).

Computer-Adaptive Testing (CAT)
Classroom assessment practices have yet to benefit from advancements in both psychometric theory
and computer technology. Today, almost every school and classroom in the United States provides
access to computers and the Internet. Despite this improved access to computer technology, few
educators use technology to its potential. Within an IRT based Computer-Adaptive Test (CAT), items
are selected based on the student’s performance on all previously administered items. As a student
answers each item, the item is scored in real time, and his or her ability (theta) is estimated. When a
CAT is first administered, items are selected via a “step rule” (Weiss, 2004). That is, if a student answers
an initial item correctly, his or her theta estimate increases by some value (e.g., .50). Conversely, if an
item is answered incorrectly, the student’s theta estimate decreases by that same amount. As testing
continues, the student’s ability is re-estimated, typically via Maximum Likelihood Estimation (MLE).
After an item is administered and scored, theta is re-estimated and used to select the subsequent
item. Items that provide the most information—based on the item information function—at that
theta level that have not yet been administered are selected for the examinee to complete. The test is
terminated after a specific number of items have been administered (a fixed-length test) or after a
certain level of precision—measured by the standard error of the estimate of theta—is achieved.
Subsequent administrations begin at the previous theta estimate and only present items that have
not been administered to that particular student. Research using simulation methods and live data
collections has been performed on aReading to optimize the length of administrations, the level of the
initial step size, and item selection algorithms to maximize the efficiency, and psychometric properties
of the assessment.
There are multiple benefits of CAT as compared to traditional paper-and-pencil tests or non-adaptive
computerized tests. The benefits that are most often cited in the professional literature include: (a)
individualized dynamic assessment, which does not rely on a fixed set of items across
administrations/individuals; (b) testing time that is reduced by one-half to one-third (or more) of
traditional tests because irrelevant items are excluded from the administration; (c) test applicability
Page | 30
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
and measurement precision across a broad range of skills/abilities, and (d) more precise methods to
equate assessment outcomes across alternate forms or administrations (Kingsbury & Houser, 1999;
Weiss, 2004; Zickar, Overton, Taylor, & Harms, 1999).
IRT-based CAT can be especially useful in measuring change over time. CAT applications that are used
to measure change/progress have been defined as adaptive self-referenced tests (Weiss, 2004; Weiss &
Kingsbury, 1984) or, more recently, adaptive measurement of change (AMC; Kim-Kang & Weiss, 2007;
Kim-Kang, & Weiss, 2008). AMCs can be used to measure the change of an individual’s skills/abilities
with repeated CATs administered from a common item bank. Since AMC is a CAT based in IRT, it
eliminates most of the problems that result when measuring change (e.g., academic growth) using
traditional assessment methods that are based in classical test theory. Kim-Kang and Weiss (2007)
have demonstrated that change scores derived from AMC do not have the undesirable properties that
are characteristic of change scores derived by classical testing methods. Research suggests that
longitudinal measurements obtained from AMC have the potential to be sensitive to the effects of
treatments and interventions at the single-person level, and are generally superior measures of
change when compared to assessments developed within a classical test theory framework (VanLoy,
1996). Finally, AMC compiles data and performance estimates (θ) from across administrations to
enhance the adaptivity and efficiency of CAT.

Aspects of Reading measured by aReading
Concepts of Print (COP)
The assessment of Concepts of Print in aReading focuses on assessing types of instruction outlined in
the state and national standards and is based on relevant reading research for developing readers
including skills synthesized from the work of Marie Clay (e.g., 2007), Barbara Taylor (i.e., 2011), Levy et
al., (2006), and NWEA goals for Concepts of Print development (2009).
Concepts of Print is defined as the general understanding of how print works and how it can be used
(Snow, Burns, & Griffin, 1998). Concepts of print is the set of skills used in the manipulation of textbased materials, which include effective orientation of materials (directionality), page turning, identify
the beginning and ending of sentences, identify words, identify letters, sentences, and sentence parts.
Concepts of print are normally developed in the emergent literacy phase of development and enable
the development of meaningful early reading skills: “Emergent literacy consists of skills, knowledge,
and attitudes that are developmental precursors to conventional forms of reading and writing”
(Whitehurst & Lonigan, 1998). These skills typically occur prior to or early in school years and are based
on the child’s exposure to printed materials and reading skills modeled by others, especially adults.
Development in this area from age 4 to age 6 has been documented using word and sentence
discrimination tasks that violated elements of word shape, word elements, and spelling (Levy, Gong,
Hessels, Evans, & Jared, 2006).
By age 4, few children are able to read any single words, but most can distinguish drawings from
writing and can detect abstract print elements such as letter spacing. These latter skills are related to
Page | 31
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
letter reading skill but not phonological awareness. This suggests that print conventions may develop
before word reading skills (Rayner, et al., 2012).
At age 5, most can detect word-shape and letter-orientation violations and letter sequencing. At age
6, knowledge of spelling is a stronger predictor of word reading than for 5 year olds (Rayner et al.,
2012). According to Levy et al., “The data show clear development of print concepts from 48 to 83
months of age. This development begins with an understanding of figural and spatial aspects of
writing (word shape). Next, or in conjunction with the first development, comes development of more
abstract notions of word constituents, including letter orientation, and finally comes an understanding
of more detailed aspects of acceptable spelling patterns” (2006, p. 89).
Some more advanced skills can develop through Second Grade, such as understanding punctuation,
standard spelling, reversible words, sequence, and other standard conventions of written and spoken
language. Introductory level of logical and analytical abilities as in understanding the concepts of print
has an impact on early student reading achievement (Adams, 1990; Clay, 1972; Downing, Ollila, &
Oliver, 1975; Hardy et al., 1974; Harlin & Lipa, 1990; Johns, 1972; Johns, 1980; Lomax & McGee, 1987;
Nichols et al., 2004; Tumner et al., 1988).
Phonological Awareness (PA)
The assessment of Phonological Awareness in aReading focuses on assessing types of instruction
outlined in the state and national standards and is based on relevant reading research for developing
readers including skills that are generally ordered from broader segments of word sounds to smaller
sound distinctions and the ability to manipulate these smaller sounds.
Phonological Awareness is a broad term involving the ability to detect and manipulate the sound
structure of a language at the level of phonemes (i.e., smallest units of sound in spoken language),
onset-rimes, syllables, and rhymes. It is used to refer to spoken language rather than letter-sound
relationships, which are the focus of phonics. Most students, especially in preschool, Kindergarten, and
First Grade, benefit from systematic and explicit instruction in this area (Adams, 1990; Carnine et al.,
2009; NRP, 2000; Rayner et al., 2012; Snow, et al., 1998).
Phonemic awareness refers to the ability to know, think about, and use phonemes—individual sounds
in spoken words. It is a specific type of phonological skill dealing with individual speech sounds that
has been studied extensively and predicts success in reading development in languages that use
alphabetic writing systems (Adams, 1990; NRP, 2000; Rayner, et al., 2012). The conscious awareness of
phonemes as the basic units of sound in a language allows the reader to identify, segment, store, and
manipulate phonemes in spoken words and it is required for proficient reading—when phonemes are
linked to letters and letter combinations in the language’s orthography. According to Adams, “to the
extent that children have learned to ‘hear’ phonemes as individual and separable speech sounds, the
system will, through an associative network, strengthen their ability to remember or ‘see’ individual
letters and spelling patterns” (1990, p. 304). Unfortunately, this is a difficult task because the English
language does not follow an explicit one-to-one correspondence between phonemes and letters
(graphemes). English phonemes may be associated with various letters. Similarly, a single letter may
Page | 32
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
be associated with several phonemes. This lack of one-to-one correspondence between phonemes
and graphemes creates a sense of ambiguity. Although the English alphabet is generally structured so
that many morphemes and words can be generated from relatively few letters (see Perfetti, 1985), the
simplicity of the grapheme-phoneme relations in English (i.e., 26 letters to 41 phonemes—across
numerous word combinations, letters or letter combinations with multiple sounds) makes it a lesstransparent system for learners to decode phonologically (Liberman, Cooper, Shankweiler, & StuddertKennedy, 1967; Rayner et al., 2012). Rayner et al., (2012), provides a good example of this lack of
transparency, “American English has over a dozen vowel sounds but only five standard vowel letters.
That means that a, e, i, o, and u have to do double and triple duty…For example, cat, cake, car, and call
each use the letter a for a different vowel phoneme” (2012, p.311).
Hearing and distinguishing individual letter sounds comes later in development (Goswami, 2000).
Children who manipulate letters as they are learning to hear the sounds make better progress in early
reading development (NRP, 2000, p. 2-4). Phonemic awareness skills are centrally involved in decoding
by blending and segmenting phonemes (NRP, 2000). Phonemic awareness also helps children learn
how to spell words correctly. Phonemic segmentation is required to help children retain correct
spellings in memory by connecting graphemes to phonemes (NRP, 2000).
Phonics
The assessment of phonics in aReading focuses on assessing types of instruction outlined in the state
and national standards and is based on relevant reading research about a student’s ability to identify
and manipulate printed letters (graphemes) and sounds (phonemes).
The correspondences between spoken and written language. This connection between letters, letter
combinations, and sounds enable reading (decoding) and writing (encoding). Phonic skill
development “involves learning the alphabetic system, that is, letter-sound correspondences and
spelling patterns, and learning how to apply this knowledge” to reading (NRP, 2-89).
Phonics most often refers to an instructional approach—a systematic, planned, explicit, sequential
method to teach beginning readers how to link written letters in words to the sounds of those letters
(i.e., understand the alphabetic principle) to decode regular words. This instruction is helpful to most
beginning readers in early primary grades (Christensen & Bowey, 2005; Juel & Minden-Cupp, 2000:
Mathes et al., 2005; NRP, 2000; Stahl, 2001) and helps “…foster full alphabetic processing to enable
children to handle the orthography” (Juel, 2006, p. 422). Indeed, early and systematic instruction in
phonics results in better achievement in reading (; NRP, 2000; and others).
For the purpose of aReading assessment, we operationalize phonics as skills associated with the
awareness and use of letter-sound (i.e., grapheme-phoneme) correspondence in relation to
development of successful spelling and reading using the language’s orthography. Assessment and
instruction of phonics explores how these skills are applied to decode (read) and encode (spell/write)
the language (NRP, 2000).

Page | 33
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Orthography and Morphology
The assessment of orthography and morphology in aReading focuses on assessing types of instruction
outlined in the state and national standards and is based on relevant reading research for readers
including development of correct spelling, word identification and discrimination, and application of
morphological and phonological knowledge.
Measures of orthography and morphology assist readers to recognize and decode or decipher words
in isolation and during reading. The ability to quickly recognize words and access their meanings
allows readers to focus their limited cognitive resources (e.g., attention, memory) on meaning instead
of decoding (e.g., Bear, Invernizzi, Templeton, & Johnston, 2012). For example, students whose reading
difficulties or disabilities persist into the secondary grades need explicit instruction in word
recognition skills to help them identify and understand new words—particularly across different
content areas. These skills contribute substantively to vocabulary and reading comprehension
development, therefore assessing students in these areas allows aReading to determine if a student is
able to accurately use and apply these skills.
Orthography

Orthography is the relationship between a script (i.e., a set of symbols) and the structure of a language
(Katz & Frost, 1992). It typically refers to the use of letters and sounds to depict (i.e., write) a language.
In relation to word learning and vocabulary development however, it refers to the reader’s ability to
identify, develop, store and retrieve orthographic representations of words or word parts using
underlying orthographic/visual representations and phonological structures (Burt, 2006; Perfetti, 1992,
1997; Stanovich & West, 1989). Although phonological elements have been identified as central to
development of letter and word skills, they do not explain all of the variance in word recognition such
that, as Stanovich and West (1989) point out, “…phonological sensitivity is a necessary but not
sufficient condition for the development of word recognition processes” (p.404). Thus the underlying
visual element of orthographic representation is also important. Given older students’ exposure to
new words throughout the upper grade levels, both elements of sound and orthographic/visual
representations of letter/word identification need to be considered when discussing assessment of
orthography in older students.
Measures of orthography and morphology can assist readers in recognizing and decoding or
deciphering words in isolation and during reading. The ability to quickly recognize words and access
their meanings allows readers to focus their limited cognitive resources (e.g., attention, memory) on
meaning instead of decoding (Bear et al., 2012). Students whose reading difficulties persist into the
secondary grades need explicit instruction in word recognition skills to help them identify and
understand new words – particularly across different content areas. These skills contribute
substantively to vocabulary and reading comprehension development. Therefore, assessing students
in these areas using aReading orthography items helps to determine whether a student is able to
accurately use and apply these skills.

Page | 34
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Morphology

Morphology is unique from orthography because it emphasizes recognition and manipulation of
word parts that help readers better understand a word’s meaning. Morphemes are the smallest unit of
meaning in a word and morphology is the study of these forms or parts of words. In relation to word
learning and vocabulary development, morphology refers to the reader’s ability to be morphologically
aware of word roots and affixes—suffixes and prefixes—and word origins (etymology). It also refers to
the structural analysis readers may use to segment and manipulate parts of words to identify new
words and help determine unknown word meanings (i.e., morphological analysis) (Carnine et al.,
2009).

Morphological awareness is formally defined as “…awareness of morphemic structures of words and
the ability to reflect on and manipulate that structure” (Carlisle, 1995, p. 194, see also Carlisle 2011)
and morphological processing involves underlying cognitive processes involved in understanding and
using morphological information (Bowers, Kirby, & Deacon, 2010; Deacon, Parrila, & Kirby, 2008). In
their review of the literature on instruction in morphology, Bowers, Kirby, and Deacon (2010) use
morphological knowledge instead of either the awareness or processing terms frequently used, due to
the ambiguity of the learning processes that may or may not be used by students in relation to
morphological instruction. aReading therefore typically refers to morphological knowledge and this
background guided development of items to assess students’ knowledge of the meaning behind parts
of a word.
Vocabulary
The assessment of vocabulary in aReading focuses on assessing word knowledge and vocabulary
outlined in the state and national standards and based on relevant reading research for K–12 readers
including understanding and recognition of words in context that are appropriate for students at
grade-level as well as appropriate for mature readers and writers to convey concepts, ideas, actions,
and feelings (NAEP, 2011). These words include academic and content-specific words, word
categories, word relations, and different parts of speech. The goal of vocabulary assessment should be
to measure word knowledge in context rather than in isolation due to the integrated nature of
reading comprehension in relation to vocabulary development.
Vocabulary is an oral language skill that involves understanding the semantic and contextual relations
of a word(s) in both general and content-specific domains (Storch & Whitehurst, 2002; Whitehurst &
Lonigan, 1998). Vocabulary knowledge develops through general oral communication, direct
instruction using contextual or decontextualized words and through reading connected texts in a
variety of contexts (Nagy, 2005; Stahl & Kapunis, 2001). As new words are incorporated, word
knowledge and efficiency of access is increased (Perfetti, 1994). Vocabulary is related to other reading
skills (Cain, Oakhill, & Lemon, 2004; Ricketts, Nation, & Bishop, 2007), and a particularly strong
interactive and reciprocal relation occurs between vocabulary and reading comprehension across
ages (Carroll, 1993; McGregor, 2004; Muter, Hulme, Snowling, & Stevenson, 2004; Nation, Clarke,
Marshall, & Durand, 2004; Nation & Snowling, 2004; Oakhill, et al., 2003).

Page | 35
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Indeed, “Vocabulary knowledge is fundamental to reading comprehension; one cannot understand
text without knowing what most of the words mean” (Nagy, 1988, p. 1). Developing a reading
vocabulary typically enlists one’s oral vocabulary: as a beginning reader comes across words that are
less familiar in print, these are decoded and “mapped onto the oral vocabulary the learner brings to
the task” (NRP, 2000, p. 4-3). If the word is not in the learner’s oral vocabulary, decoding becomes less
helpful and contextual information more important. Learning new words from text involves
connecting the orthography to multiple contexts and establishing a flexible definition. Vocabulary
knowledge, then, includes both definitional knowledge and contextual knowledge (Stahl, 1999). Some
words in text are so familiar that they no longer require explicit processing; these are referred to as a
sight word vocabulary (NRP).
Comprehension
The assessment of reading comprehension in aReading focuses on comprehension processes outlined
in the state and national standards and based on relevant reading research for K–12 readers including
the reader’s development of an organized, coherent, and integrated representation of knowledge and
ideas in the text through use of inferential processes and identification of key ideas and details in the
text as well as understanding its craft and structure.
The goal of reading comprehension is to understand the meaning of what is heard and read.
Comprehension is the process of understanding what is heard and read. Comprehension, or
constructing meaning, is the purpose of reading and listening. The NRP noted that “Comprehension
has come to be viewed as the ‘essence of reading’ (Durkin, 1993), essential not only to academic
learning but to lifelong learning as well” (NRP, 2000, p. 4-11). “Good readers have a purpose for
reading” (Armbruster, 2002, p. 34), like learning how to do something, finding out new information, or
for the enjoyment and entertainment that reading for pleasure brings. Good readers actively process
the test “to make sense of what they read, good readers engage in a complicated process. Using their
experiences and knowledge of the world, their knowledge of vocabulary and language structure, and
their knowledge of reading strategies . . ., good readers make sense of the text and know how to get
the most out of it. They know when they have problems with understanding” and they know “how to
resolve these problems as they occur” (Armbruster, 2002, p. 48). aReading items for grades 6–12
depended heavily on comprehension skills. Thus, the aReading team consulted with Dr. Paul van den
Broek in spring 2012 to learn from his expertise in cognitive processes involved in reading
comprehension. After meeting with Dr. van den Broek, the team ensured that questions about the
reading passages should ask students to (a) locate and recall broad and specific information in the
text, (b) integrate and interpret beyond the explicit information in the text, and (c) critique and
evaluate the quality of the author’s writing.

Uses and Applications
Each aReading assessment is individualized by the software and, as a result, the information and
precision of measurement is optimized regardless of whether a student functions at, above, or
significantly below grade level.

Page | 36
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Target Population
aReading is intended for use from Kindergarten through Twelfth Grades for screening. The aReading
item bank consists of approximately 2000 items that target the reading domains described in the
previous section. Items developed for Kindergarten through Grade Five target Concepts of Print,
Phonological Awareness, Phonics, Vocabulary, and Comprehension. Items developed for middle and
high school grade levels target Orthography, Morphology, Vocabulary, and Comprehension. Please
note, however, that the importance and emphasis on each reading domain will vary across children.

Chapter 2.2: Development
earlyReading
The results of the National Assessment of Educational Progress (NAEP) for 2011 suggest that among
Fourth Grade students, 3% perform below a basic level (partial mastery of fundamental skills) and 68%
perform below a proficient level of achievement (demonstrated competency over challenging subject
matter) (National Center for Education Statistics, 2013). Among eighth grade students, 25% perform
below basic and 68% perform below proficiency (Aud, Wilkinson-Flicker, Kristapovich, Rathbun, Wang,
& Zhang, 2013). Approximately 32% of students demonstrate reading proficiency at grade level. The
relatively low levels of reading proficiency and achievement within the general population are
reflected within the population receiving services under the Individuals with Disabilities Education Act
(IDEA), originally enacted in 1975. Among students who receive special education services, 91% of
Fourth Graders and 95% of eighth graders fail to achieve grade-level proficiency in reading. Moreover,
the majority of these students scored in the below basic range in reading achievement (National
Center for Educational Statistics, 2003). The incidence of reading-related disabilities is
disproportionately large when compared to other categories of disability under IDEA. Government
data suggests that almost 60% of the students who are served under IDEA are identified with a
specific learning disability, and 80% of those students are identified with a reading disability (U.S.
Department of Education, 2001, 2002). Of the nearly 3 million students served under IDEA and the 1.5
million students identified with a specific learning disability, approximately 1.3 million are identified
with a reading disability.
Reading instruction and reading development has never been better understood. Nevertheless, there
is a great deal of progress to be made in the future by building on our present knowledge-base. The
National Reading Panel identified five essential component skills that support reading development:
phonemic awareness, phonics, fluency, vocabulary, and comprehension. They did not, however,
define the relative scope and importance of each component within or across developmental phases.
It is likely that specific skill-sets are most relevant and more salient at particular points in the
developmental continuum (Paris, 2005). For elementary grades two through six, measures of fluency
with connected text (curriculum based measure of oral reading; CBM-R) are often used as a universal
screeners for grade-level reading proficiency. CBM-R requires students to read from a grade level
passage for one minute while the number of words read correctly is recorded. Strong evidence exists
in the literature to support the use of CBM-R (L.S. Fuchs, Fuchs, & Maxwell, 1988; Kranzler, Brownell, &
Miller, 1998; Markell & Deno, 1997); however, support for CBM-R as a universal screener for students
Page | 37
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
not yet reading connected text is less robust (Fuchs, Fuchs, & Compton, 2004; National Research
Council, 1998). The research literature provides substantial guidance on instruction and assessment of
alphabetic knowledge, phonemic awareness, and oral reading. The objective of earlyReading
measures is to extend and improve on the quality of currently available assessments.
There is growing evidence that successful reading in elementary school grades depends on a
combination of code-related and language comprehension skills. Code-related skills include the
awareness that printed text is meaningful and the ability to translate the letters and words of the text
into such meaningful concepts. Language comprehension skills concern the ability, knowledge and
strategies necessary to interpret concepts and connect these concepts into a coherent mental
representation of the text (Gough & Tunmer, 1986; Oakhill & Cain, 2007; Whitehurst & Lonigan, 1998).
A long tradition of research support indicates that early code-related skills predict later reading
achievement (Adams, 1990; Chall, 1987; Juel, 1988; LaBerge & Samuels, 1974; National Reading Panel,
2000a; Stanovich, 1984). The design of earlyReading has a strong foundation in both research and
theory. During the early phases of student reading development, the component processes of reading
are most predictive of future reading success (Stanovich, 1981, 1984, 1990; Vellutino & Scanlon, 1987,
1991; Vellutino, Scanlon, Small, & Tanzman, 1991). Indeed, reading disabilities are most frequently
associated with deficits in accurate and efficient word identification.
CBMreading
The National Reading Panel identified five essential component skills that support reading
development: phonemic awareness, phonics, fluency, vocabulary and comprehension. Fluency (or
rate) in particular, is important because it establishes a connection between word identification and
comprehension. Despite research establishing CBM-R as a valid measure of general reading
achievement, as well as an effective tool for predicting later reading performance, research also
provides evidence for the necessity to improve CBM-R instrumentation and procedures. Estimates of
longitudinal growth for individual students can be more a function of the instrumentation (i.e.,
inequitable passages) as opposed to students’ actual response to intervention. Development of the
CBMreading passages, therefore, was based on findings from existing research and theory. This
provided clear guidance for ways to improve progress monitoring measures and outcomes that could
yield substantial and significant improvements for a widely used approach to progress monitoring.
These benefits are especially relevant within the evolving context of response to intervention, which
relies substantially on CBM-R and rate-based measures of reading. The goal in creating the
CBMreading measures, therefore, was to systematically develop, evaluate and finalize research-based
instrumentation and procedures for accurate, reliable, and valid assessment and evaluation of reading
rate.
aReading
Similar to CBMreading, aReading item development followed the process and standards presented by
Schmeiser and Welch (2006) in the fourth edition of Educational Measurement (Brennan, 2006).
Research assistants, teachers from each grade level (1st through 12th), and content experts in the area
Page | 38
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
of reading served as both item writers and reviewers for those items at the Kindergarten through 5th
grade level. Items for grades 6 through 12 were constructed to reflect the Common Core State
Standards’ (National Governors Association Center for Best Practices & Council of Chief State School
Officers, 2010) specifications for various skills of interest, as well as the National Assessment of
Educational Progress’ (NAEP, 2011) guidelines for reading assessment items. After items were written
at all grade levels, they were reviewed for feasibility, construct relevance, and content balance. A
stratified procedure was used to recruit a diverse set of item writers from urban, suburban and rural
areas. The item writers wrote, reviewed, and edited assessment materials.
Item writing for aReading was a multi-year, collaborative, and iterative process. First the literature on
item writing guidelines used when developing assessments was reviewed. Next, the literature on
multiple-choice item writing was reviewed. Once the literature was reviewed, the guidelines were
applied to aReading to examine relevance and utility. Extensive guidelines and practice were provided
to item writers and the process outlined above was followed.

The Item Development Process: An a-priori Model
aReading targets the essential skills that enable reading. In its current form aReading provides general
estimate of overall reading achievement (i.e., a screening measure), which we define as the Unified
Measure of Reading Achievement (see Figure 1 below). The research team established an a priori
model of component skills and a unified measurement construct based on previous data and
assumptions regarding typical rates of reading achievement. The use of grade levels is a convenience
because the actual assessment is individualized to each student at the time of assessment (see later
section regarding computer adaptive testing). The grade levels indicate only the likely relevant
domains. These target skill components for assessment are derived from the research literature and
are consistent with the recommendation of the National Reading Panel (2000) and critical
components of most state standards for instruction and assessment.

Page | 39
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Figure 1 A priori model for unified reading achievement

The Item Development Process: Alignment with Common Core State Standards
The Common Core State Standards for English Language Arts and Literacy in History/Social Studies,
Science and Technical Subjects (furthered referred to as the Standards) is a synthesis of information
gathered from state departments of education, assessment developers, parents, students, educators,
and other pertinent sources to develop the next generation of state standards of K–12 students to
ensure that all students are college and career literacy ready by the end of their high school education.
This process is headed by the Council of Chief State School Officers (CCSO) and the National Governors
Association (NGA). The Standards are an extension of a previous initiative by the CCSSO and NGA
titled the College and Career Readiness (CCR) Anchor Standards. The CCR Anchor Standards are
numbered from one to ten, and are as follows: (1) Read closely to determine what the text says
explicitly and to make logical inferences from it; cite specific textual evidence when writing or
speaking to support conclusions drawn from the text. (2) Determine central ideas or themes of a text
and analyze their development; summarize the key supporting details and ideas. (3) Analyze how and
why individuals, events, and ideas develop and interact over the course of a text. (4) Interpret words
and phrases as they are used in a text, including determining technical, connotative, and figurative
meanings, and analyze how specific word choices shape meaning or tone. (5) Analyze the structure of
texts, including how specific sentences, paragraphs, and larger portions of the text (e.g., a section,
chapter, scene, or stanza) relate to each other and the whole. (6) Assess how point of view or purpose
shapes the content and style of a text. (7) Integrate and evaluate content presented in diverse media
and formats, including visually and quantitatively as well as in words. (8) Delineate and evaluate the
Page | 40
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
argument and specific claims in a text, including the validity of the reasoning as well as the relevance
and sufficiency of the evidence. (9) Analyze how two or more texts addresses similar themes or topics
in order to build knowledge or to compare the approaches the authors take. (10) Read and
comprehend complex literary and informational texts independently and proficiently. These anchor
standards were designed with three themes in mind: craft and structure, integration of knowledge
and ideas, and range of reading and level of text complexity.
The Standards add a level of specificity in the form of end of grade expectations for each of these ten
anchor standards. To do this, the Standards organize the 10 anchor standards in three ways. First, a
distinction is made between Literature and Information text. Secondly, the ten items are grouped into
relevant clusters which are the same for both literature and information text. Those clusters are: Key
Ideas and Details 1–3, Craft and Structure 4–6, Integration of Knowledge and Ideas 7–9, and Range of
Reading and Level of Text Complexity 10. Further, the Standards provide a corresponding end of
grade skill expectation by grade for each number within the cluster. A portion of the Readings
Standards for Information Text is presented inRelevant to reading standards for students in
Kindergarten through Fifth Grade, the CCSO and NGA identify foundational skills, including a working
knowledge of concepts of print, the alphabetic principle, and other basic conventions of the writing
system. Sub-categories under these foundational skills include print concepts, phonological
awareness, phonics and word recognition, and fluency. It is important to acknowledge that at this
point in time, fluency is not yet applicable to aReading. Print concepts encompass the ability to
demonstrate the organization and basic feature of print. Phonological awareness is the ability to
demonstrate an understanding of spoken words, syllables, and sounds or phonemes. Finally, phonics
and word recognition includes applying grade-level phonics and word analysis skills in the process of
decoding words. Within each category, there are specific end of the year expectations for each grade
level. Examples are shown in
Table 2. Table 3 specifies cross-references between Common Core State Standards and aReading item
domains.
Table 1 below. Similar, grade-level standards for all K–12 grade levels are available to view in the
Common Core State Standards (2010).
Relevant to reading standards for students in Kindergarten through Fifth Grade, the CCSO and NGA
identify foundational skills, including a working knowledge of concepts of print, the alphabetic
principle, and other basic conventions of the writing system. Sub-categories under these foundational
skills include print concepts, phonological awareness, phonics and word recognition, and fluency. It is
important to acknowledge that at this point in time, fluency is not yet applicable to aReading. Print
concepts encompass the ability to demonstrate the organization and basic feature of print.
Phonological awareness is the ability to demonstrate an understanding of spoken words, syllables,
and sounds or phonemes. Finally, phonics and word recognition includes applying grade-level
phonics and word analysis skills in the process of decoding words. Within each category, there are
specific end of the year expectations for each grade level. Examples are shown in
Page | 41
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 2. Table 3 specifies cross-references between Common Core State Standards and aReading item
domains.
Table 1. Example Standards for Informational Text
Reading Standards for Informational Text K–2
Kindergarten
Key Ideas and Details
1. With prompting and support,
ask and answer questions about
key details in a text.

Grade 1

Grade 2

1. Ask and answer questions about
key details in a text.

2. With prompting and support,
identify the main topic and retell
key details of a text.

2. Identify the main topic and
retell key details of a text.

3. With prompting and support,
describe the connection between
two individuals, events, ideas, or
pieces of information in a text.

3. Describe the connection
between two individuals, events,
ideas, or pieces of information in a
text.

1. Ask and answer such questions
as who, what, where, when, why
and how to demonstrate
understanding of key details in a
text.
2. Identify the main topic of a
multi-paragraph text as well as the
focus of specific paragraphs within
the text.
Describe the connection between
a series of historical events,
scientific ideas or concepts, or
steps in technical procedures in a
text.

Table 2. Foundational Skill Examples for Kindergarten and First Grade Students
Reading Standards: Foundational Skills (K–1)
Kindergarten
Print Concepts
1. Demonstrate understanding of the organization
and basic features of print.
a. Follow words from left to right, top to bottom, and
page by page.

Grade 1
1. Demonstrate understanding of the organization
and basic features of print.
a. Recognize the distinguishing features of a
sentence (e.g., first word, capitalization, ending
punctuation).

b. Recognize that spoken words are represented in
written language by specific sequences of letters.
c. understand that words are separated by spaces in
print.
d. Recognize and name all upper- and lowercase
letters of the alphabet.

Table 3. Cross-Referencing CCSS Domains and aReading Domains
Common Core Subgroups / Clusters
aReading Domains

Page | 42
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Foundational Skills
Print Concepts

Concepts of Print

Phonological Awareness

Phonemic Awareness

Phonetic Awareness

Phonetic Awareness

Vocabulary

Vocabulary

College and Career Readiness Reading Standards for Literature / Informational Text
Key Ideas and Details

Comprehension

Craft and structure

Comprehension & Vocabulary

Integration of Knowledge and Ideas

Comprehension & Vocabulary

Chapter 2.3: Administration and Scoring
earlyReading
Administration time varies depending on which earlyReading assessment subtest is being
administered. A timer is built into the software and is required for all subtests. For those assessments
that calculate a rate-based score (i.e., number correct per minute), the default test duration is set to
one minute. These subtests include Letter Names, Letter Sounds, Sight Words, Decodable Words, and
Nonsense Words. For those subtests that do not calculate a rate-based score (number correct), the
default test duration is set to open-ended. This includes Concepts of Print, Onset Sounds, Word
Rhyming, Word Segmenting, and Word Blending subtests. Although it is not recommended, those
administering the assessments can change the test duration by selecting options from a drop-down
menu. earlyReading is individually administered, and each subtest can take approximately one to
three minutes to complete; administration of the composite assessments for universal screening takes
approximately 5 minutes.
CBMreading
CBMreading includes standardized administration and scoring procedures along with proprietary
instrumentation, which was developed with funding from the US Department of Education and
Institute for Education Sciences. CBMreading takes approximately one minute to administer a single
passage. The administration of three passages takes approximately five minutes per student.
If a student stops or does not say a word aloud for three seconds, tell the student the word, mark the
word as incorrect, and instruct the student to continue. Aside from the three second rule, do not
provide the student with correct responses, or correct errors that the student makes. Alternate scoring
methods include word-by-word error analysis and performance analysis to evaluate the types of errors
committed by students.

Page | 43
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
aReading
aReading can be group administered in a computer lab setting, or a student can complete an
administration individually at a computer terminal set up in a classroom. The aReading assessment
terminates on its own, informing students they have completed all items. A typical aReading
administration is approximately 30 items. Students in grades K–5 take an average of 10–15 minutes to
complete an assessment, and students in grades 6–12 take an average of 20–30 minutes.
Administration time may vary by student. Instructions for completing aReading are provided via
headphones to students. In addition to audible instructions, students are provided with an animated
example. No verbal instructions are required on behalf of the administrator.

Chapter 2.4: Interpreting Test Results
earlyReading

Raw Scores
Each earlyReading subtest produces a raw score. The primary score for each subtest is the number of
items correct and/or the number of items correct per minute. These raw scores are used to generate
percentile ranks.

Composite Scores
The best estimate of students’ early literacy skills is the earlyReading composite score. The composite
score consists of multiple subtest scores administered during a universal screening period. The
earlyReading composite scores were developed as optimal predictors of spring broad reading
achievement in Kindergarten and First Grade. A selected set of individual subtest scores were
weighted to optimize the predictive relationship between earlyReading and broad reading
achievement scores (See Table 4 below). The weighting is specific to each season. It is important to
emphasize that the weighting is influenced by the possible score range and the value of the skill. For
example, letter sounds is an important skill with a score range of 0 to 60 or more sounds per minute.
This represents a broad range of possible scores with benchmark scores that are fairly high (e.g.,
benchmarks for fall, winter, and spring might be 10, 28, and 42, respectively). In contrast, Concepts of
Print has a score range from 0 to 12 and benchmarks are relatively low in value (e.g., benchmarks for
fall and winter might be 8 and 11, respectively). As a result of both the score range and the relative
value of Concepts of Print to overall early reading performance, the subtest score is more heavily
weighted in the composite score. The weightings are depicted in Table 4 (below). The high (H),
moderate (M), and low (L) weights indicate the relative influence of a one point change in the subtest
on the composite score. A one point change for an H weighting is highly influential. A one point
change in an L weighting has low influence.
The composite scores should be interpreted in conjunction with specific subtest scores. A variety of
patterns might be observed. It is most common for students to perform consistently above or below
benchmark on the composite and subtests; however, it is also possible to observe that a particular
student is above benchmark on one or more measures but below the composite benchmark. It is also
Page | 44
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
possible for a student to be below benchmark on one or more subtests but above the composite
benchmark. Although atypical, this phenomenon is not problematic. The recommendation is to
combine the use of composite and subtest scores in order to optimize the decision-making process.
Overall, composite scores are the best predictors of future reading success.

Table 4. Weighting Scheme for earlyReading Composite Scores
earlyReading Subtests
Concepts of Print
Onset Sounds
Letter Names
Letter Sounds
Word Segmenting
Nonsense/Decodable Words
Sight Words
Sentence Reading
CBMreading

Kindergarten
F
W
H
M
H
L
L
L
L
M

S

L
M
M
L

First Grade
F
W

L
H
M
L

S

L
H
M

L
H
M

L

L

Broad Score

Note. The weighting of subtests for the composite is represented above.
H – high weighting, M – moderate weighting, L – low weighting.

Kindergarten
The composite score for Kindergarten students in the fall includes Concepts of Print, Onset Sounds,
Letter Sounds, and Letter Naming. The composite score for winter includes Onset Sounds, Letter
Sounds, Word Segmenting and Nonsense Words. Finally, for spring of the Kindergarten year, the
following subtests are recommended in order to compute an interpretable composite score: Letter
Sounds, Word Segmenting, Nonsense Words, and Sight Words (50). The Decodable Words score may
be used in place of Nonsense Words for computing any of the composite scores specified.
First Grade
The composite score for First Grade students in the fall includes Word Segmenting, Nonsense Words,
Sight Words (150), and Sentence Reading. The composite score for winter includes Word Segmenting,
Nonsense Words, Sight Words (150), and CBMreading. Finally, for spring of First Grade, the following
subtests are recommended in order to compute an interpretable composite score: Word Segmenting,
Nonsense Words, Sight Words (150), and CBMreading. The Decodable Words score may be used in
place of Nonsense Words for computing any of the composite scores specified.

Page | 45
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Benchmark Scores
Benchmark scores are available for each earlyReading subtest for the specific grade level and month
for which they are intended for use. Thus, a benchmark is purposefully not provided for every subtest,
for each month, during Kindergarten and First Grade. Benchmarks were established for earlyReading
to help teachers accurately identify students who are at risk or not at risk for academic failure. These
benchmarks were developed from a criterion study examining earlyReading assessment scores in
relation to scores on the Group Reading Assessment and Diagnostic Evaluation (GRADE; Williams,
2001). Measures of diagnostic accuracy were used to determine decision thresholds using criteria
related to sensitivity, specificity, and area under the curve (AUC). Specificity and sensitivity was
computed at different cut scores in relation to maximum AUC values. Decisions for final benchmark
percentiles were generated based on maximizing each criterion at each cut score (i.e., when the cut
score maximized specificity ≥ .70, and sensitivity was also ≥ .70; see Silberglitt & Hintze, 2005). Based
on these analyses, the values at the 40th and 15th percentiles were identified as the primary and
secondary benchmarks for earlyReading, respectively. These values thus correspond with a prediction
of performance at the 40th and 15th percentiles on the GRADE™ (Group Reading Assessment and
Diagnostic Evaluation), a nationally normed reading assessment of early reading skills. Performance
above the primary benchmark indicates the student is at low risk for long-term reading difficulties.
Performance between the primary and secondary benchmarks indicates the student is at some risk for
long-term reading difficulties. Performance below the secondary benchmark indicates the student is
at high risk for long-term reading difficulties. These risk levels help teachers accurately monitor
student progress using the FAST™ earlyReading measures. Benchmarks are reported in the FastBridge
Learning: Benchmarks and Norms Guide.
CBMreading
Interpreting CBMreading scores involves a basic understanding of the various scores provided in the
FAST™ software.

Total Words Read
Total Words Read refers to the total number of words read by the student, including correct and
incorrect responses.

Number of Errors
This is the total number of errors the student made during the one minute administration time.

Words Read Correct per Minute (WRC/min)
This is the number of Words Read Correct per minute. This is computed by taking the total number of
words read and subtracting the number of errors the student made.

Benchmark Scores
Benchmark scores are available for CBMreading by grade level and time of the year (i.e., fall, winter,
spring) for which they are intended for use. A benchmark is provided for every grade level (1-6), for
each of the three time points throughout the school year. The assessment of oral reading rate with
Page | 46
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
CBM-R is well established in the literature for use to benchmark student progress (Wayman et al.,
2007). Benchmarks were established for CBMreading to help teachers accurately identify students
who are at risk or not at risk for reading difficulties. All CBMreading forms are divided into Levels A, B
and C, which correspond to 1st grade (A), 2nd and 3rd grade (B), and 4th through 6th grade (C)
reading levels, respectively. There are 39 Level A passages, 60 Level B, and 60 Level C passages. Each
passage is assigned as a screening/benchmarking form for each grade level (1st to 6th) and a variety
of progress monitoring forms. The weekly passage set options include: one unique passage weekly (1x
weekly), two unique passages weekly (2x weekly), or three unique passages weekly (3x weekly).
Analyses were conducted to link CBMreading scores with DIBELS Next and AIMSweb benchmark and
target scores.
Measures of diagnostic accuracy were used to determine decision thresholds using criteria related to
sensitivity, specificity, and area under the curve (AUC). Specifically, specificity and sensitivity was
computed at different cut scores in relation to maximum AUC values. Decisions for final benchmark
percentiles were generated based on maximizing each criterion at each cut score (i.e., when the cut
score maximized specificity ≥ .70, and sensitivity was also ≥ .70; see Silberglitt & Hintze, 2005).
Benchmarks generally correspond to the 40th and 15th percentiles on nationally normed assessments
and FAST™ norms. Benchmarks are reported in the FastBridge Learning: Benchmark and Norms Guide.
aReading
Interpreting aReading scores involves a basic understanding of the various scores provided in the
FAST™ software.

Scaled Scores
Scores generated by the aReading computer-adaptive test (CAT) yield scores based on an IRT logit
scale. This type of scale is not useful to most school professionals; in addition, it is difficult to interpret
scores on a scale for which everything below the mean value yields a negative number. Therefore, it
was necessary to create an aReading scale more similar to existing educational measures. Such scales
are arbitrarily created with predetermined basal and ceiling scores. For instance, the Measure of
Academic Progress (MAP; NWEA) uses a scale from 150 to 300 and STAR Early Literacy (Renaissance
Learning) uses a scale from 300 to 850.
The aReading scale yields scores that are transformed from logits using the following formula:
y = 500 + (50 x Logit Score)
The logit scale has an M = 0 and SD = 1 and y is the new aReading scaled score, and (theta) is the
initial aReading logit theta estimate. Scores were scaled with a lower bound of 350 and a higher

Page | 47
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
bound of 650. The mean value is 500 and the standard deviation is 50. There are several shortcomings
in reporting logit scores to educational professionals. Among these are: (a) teachers are unfamiliar
with a measure ranging in six points with decimals and (b) using the current logit reporting scheme,
negative values demarcate ability estimates below average. Displaying negative values for ability
estimates may carry with it off-putting connotations. Thus, the researchers for aReading chose to
adopt an arbitrary scale upon which to report logit scores and theta estimates. The idea of reporting
scores on the same scale as standard intelligence tests was considered but ultimately dismissed. The
reason for dismissing this idea was two-fold; the researchers wanted to avoid educators and parents
inaccurately equating aReading scores with IQ scores. In addition, creating a novel and arbitrary scale
would encourage educators to refer to the current technical manual for assistance in interpreting such
scores accurately. Details on interpreting aReading scaled scores for instructional purposes is
delineated in the following section.
Interpreting Scaled Scores
aReading scaled scores
have an average of 500
and standard deviation of
50 across the range of
Kindergarten to twelfth
grades. Scores should be
interpreted with
reference to the
benchmarks and norms.
In addition, aReading has
descriptions regarding
the interpretation of a
student’s scaled score
with respect to mastered,
developing, and future
Figure 2. Example of a student's aReading report with interpretations of the scaled
skill development. These
score
are intended to help
teachers better understand the developmental progression and student needs. FAST™ generates
individual reports to describe the reading skills that the student has mastered, is developing, and will
develop based on their scaled score Figure 2 shows an example of a student’s score report and her
mastered, developing, and future skills for Concepts of Print.

Benchmark Scores
Benchmark scores for aReading are available for Kindergarten through Twelfth Grade at three time
periods: fall, winter, and spring. Benchmarks were established for aReading to help teachers accurately
identify students who are at risk or not at risk for academic failure. Benchmarks are reported in the
FastBridge Learning: Benchmark and Norms Guide.

Page | 48
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Chapter 2.5: Reliability
earlyReading
earlyReading measures were administered to Kindergarten and First Grade students from nine
elementary schools across three school districts in a metropolitan area in the Midwest. Kindergarten
students who participated in the study were enrolled in all-day Kindergarten programming. The
demographic information for each school district is provided in Table 5.
Table 5. Demographic Information for earlyReading Alternate Form Sample
Category

District A

District B

District C

White

56.1%

93%

79.5%

Black

13.5%

4%

6.8%

Hispanic

10.3%

3%

4.5%

Asian/Pacific Islander

19.4%

4%

10.5%

American Indian/Alaska Native
Free/Reduced Lunch

>.1%
44.9%

1%
17%

.25%
9%

LEP

15.8%

6%

6%

Special Education

12.6%

10%

10%

Classrooms were recruited by the reading coordinator within each school district. Teachers received a
$20.00 gift card for participating. Five progress monitoring alternate forms were randomly chosen for
each earlyReading measure (for which progress monitoring forms exist). Students were administered
five forms of one to two earlyReading measures consecutively in one sitting. Administration was
conducted by trained administrators who all attended a two-hour training session in addition to
completing integrity checks while working with students.
Evidence of reliability is available for Alternate Forms for all earlyReading subtests (see
Table 6). In order to effectively examine reliability coefficients, Standard error of measurement (SEm)
has also been provided. The SEm is an index of measurement error representing the standard
deviation of errors attributable to sampling. The SEm provides information about the confidence with
which a particular score can be interpreted, relative to a single individual’s true score. Thus, a small
SEm represents greater confidence that a score reflects the individual’s true performance and skill
level. The SEm is based on the formula
=
√1 − where SD represents the standard deviation
of the distribution and r represents the reliability of the measure. The SEm can be used by those
administering the measure to help interpret the score obtained by a student. The SEm for both
Kindergarten and First Grade are available for each subtest in
Table 6. To determine parallel form construction, ANOVAs were conducted to compare alternate forms
for each individual subtest. See below for a complete description of parallel form construction for the

Page | 49
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
following earlyReading subtests: Onset Sounds, Letter Naming, Letter Sounds, Word Blending, Word
Segmenting, Decodable Words, and Nonsense Words.
Table 6. Alternate Form Reliability and SE m for earlyReading
Grade
Kindergarten
Onset Sounds
Letter Naming
Letter Sound
Word Blending
Word Segmenting
Decodable Words
Nonsense Words
Sight Words (50)
Nonsense Words
Grade 1
Word Blending
Word Segmenting
Decodable Words
Nonsense Words
Sight Words (150)

N (range)

Coefficient
Range

Median

SEm (SD)

25–29
36–37
34–36
36–37
37–38
29
28
24–28
28

.77–.89
.82–.92
.85–.94
.59–.79
.68–.92
.96–.98
.86–.96
.94–.99
.86–.96

.83
.88
.89
.71
.82
.97
.93
.97
.93

.99 (.86)
5.07 (3.77)
5.56 (4.89)
.97 (.82)
8.07 (6.21)
2.93 (2.71)
2.15 (1.91)
4.40 (4.13)
2.15 (1.91)

30–31
40
36–37
26–27
37

.15–.59
.67–.87
.97–.98
.69–.96
.91–.96

.26
.82
.98
.85
.94

9.83
2.98
3.05 (3.04)
4.14

Note. SD = Standard Deviation.

Alternate Form Reliability
Onset Sounds. To determine parallel form construction, a one-way, within-subjects (or repeated
measures) ANOVA was conducted to compare the effect of Onset Sounds alternate forms (n = 5) on
the number of correct responses within individuals. There was not a significant effect for forms F
(1,109) = 1.81, p =.18. This indicates that different forms did not result in significantly different mean
estimates of correct responses.
Letter Names. To determine parallel form construction, a one-way, within-subjects (or repeated
measures) ANOVA was conducted to compare the effect of Letter Names alternate forms (n = 5) on the
number of correct responses within individuals. There was not a significant effect for forms F (1,146) =
.71, p=.40. This indicates that different forms did not result in significantly different mean estimates of
correct responses.
Letter Sounds. To determine parallel form construction, a one-way, within-subjects (or repeated
measures) ANOVA was conducted to compare the effect of Letter Sounds alternate forms (n = 5) on
the number of correct responses within individuals. There was not a significant effect for forms F
(1,139) = .96, p =.33. This indicates that different forms did not result in significantly different mean
estimates of correct responses.

Page | 50
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Word Blending. To determine parallel form construction, a one-way, within-subjects (or repeated
measures) ANOVA was conducted to compare the effect of Word Blending alternate forms (n = 5) on
the number of correct responses within individuals. There was not a significant effect for forms F
(1,121) = 1.60, p =.21. This indicates that different forms did not result in significantly different mean
estimates of correct responses.
Word Segmenting. To determine parallel form construction, a one-way, within-subjects (or repeated
measures) ANOVA was conducted to compare the effect of Word Segmenting alternate forms (n=5)
on the number of correct responses within individuals in both Kindergarten and grade 1. There was
not a significant effect for forms as used in either grade: Kindergarten = F (1,150) = 3.24, p=.07; grade 1
= F (1,121) = 1.60, p =.21. This indicates that different forms did not result in significantly different
mean estimates of correct responses.
Decodable Words. To determine parallel form construction, a one-way, within-subjects (or repeated
measures) ANOVA was conducted to compare the effect of Decodable Words alternate forms (n = 5)
on the number of correct responses within individuals. There was not a significant effect for forms F
(1,145) = 1.72, p =.19. This indicates that different forms did not result in significantly different mean
estimates of correct responses.
Nonsense Words. To determine parallel form construction, a one-way, within-subjects (or repeated
measures) ANOVA was conducted to compare the effect of Nonsense Words alternate forms (n = 5) on
the number of correct responses within individuals. For Kindergarten, there was not a significant effect
for forms F (1,107) = .03, p =.86. For First Grade, there was not a significant effect for forms F (1,106) =
2.34, p =.13. This indicates that different forms did not result in significantly different mean estimates
of correct responses.

Internal Consistency (Item-Total Correlations)
Some earlyReading measures have fixed test lengths and are subject to typical internal consistency
analyses. Some earlyReading measures, however, are timed. Different students will therefore have
tests of different lengths. Internal consistency measures of reliability are inflated on timed measures
because of the high percentage of incomplete items at the end of the assessment, which are those for
which examinees did not respond (Crocker & Algina, 1986). As a solution to both illustrate the
potential inflation and also reduce it, estimates of internal consistency (reliability) were run on the
items completed by approximately 16% of students, the items completed by 50% of students, and
items completed by approximately 84% of students. Items not completed were coded as incorrect.
For both fixed test -length and inconsistent test-length analyses, data were derived from a random
sample of students from the FAST™ database from the 2012–13 academic year. Reliability of measures
with variable test length is reported in .

Page | 51
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Table 7. Reliability of measures with fixed test length is reported in Table 8.

Table 7. Internal Consistency for earlyReading Subtests of Variable Test Length
Measure

Grade

N

Alpha

Split-Half

Letter Names

K

444

18 items

.95

.96

35 items

.98

.99

52 items

.98

.99

10 items

.93

.93

30 items

.98

.98

50 items

.98

.99

6 items

.76

.75

23 items

.95

.96

40 items

.98

.98

5 items

.74

.73

18 items

.93

.95

31 items

.96

.98

11 items

.90

.91

29 items

.97

.98

47 items

.99

.99

12 items

.90

.91

53 items

.99

.99

94 items

.99

.99

Letter Sounds

K

Decodable Words

683

K-1

Nonsense Words

434

K–1

Sight Words (50)

501

K–1

Sight Words (150)

505

1

678

Table 8. Internal Consistency for earlyReading Subtests of Fixed Test Length
Measure

Grade

N

# of items

Alpha

Split-Half

Concepts of Print

K

336

12

.75

.76

Onset Sounds

K

597

16

.87

.91

Rhyming

K

586

16

.94

.91

Page | 52
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Word Blending

K–1

480

10

.90

.91

Word Segmenting

K–1

500

10

.95

.96

Page | 53
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Test-Retest Reliability
In fall 2012, data were collected to determine test-retest reliability for all earlyReading screening
measures. Participants included 85 Kindergarten and 71 First Grade students from two elementary
schools in a metropolitan area in the Midwest. Kindergarten students who participated in the study
were enrolled in all-day Kindergarten at two elementary schools within the same school district.
Table 9. Descriptive Information for earlyReading Test-Retest Reliability Sample
Grade

Measure

N

Min

Time 1 (Note. Time 1 is based on fall screening data)
K
Concepts of Print
39
3
K
Onset Sounds
39
2
K
Letter Naming
39
2
K
Letter Sound
39
0
K
Rhyming
39
0
K
Word Blending
39
0
Time 2 (~3 weeks later)
K
Concepts of Print
40
5
K
Onset Sounds
40
5
K
Letter Naming
40
2
K
Letter Sound
40
1
K
Rhyming
40
4
K
Word Blending
34
0
Time 1 (Note. Time 1 is based on fall screening data)
1st
Word Blending
37
0
1st
Word Segmenting
37
4
1st
Decodable Words
37
0
1st
Nonsense Words
37
1
1st
Sight Words (150)
37
1
1st
Sentence Reading
37
1
st
1
Composite
33
79
Time 2 (~3 weeks later)
1st
Word Blending
37
0
1st
Word Segmenting
37
14
1st
Decodable Words
37
0
1st
Nonsense Words
37
0
1st
Sight Words (150)
37
0
1st
Sentence Reading
37
0
1st
Composite
33
80

Max

SD

Mean

Max Possible

12
16
52
50
16
10

1.96
2.94
15.91
13.36
4.60
3.81

9.46
13.74
32.23
19.62
11.77
3.92

12
16
52/min
38/min
16
10

12
16
52
39
16
9

2.07
2.35
15.73
11.23
3.74
3.13

9.7
14.15
31.70
19.43
12.18
3.71

12
16
52/min
38/min
16
10

10
34
49
34
91
181
128

7.19
55.81
12.82
10.65
35.35
42.68
11.99

2.94
7.50
14.10
8.39
24.21
38.60
104.5

10
32
50/min
50/min
150/min

10
34
50
50
109
220
138

11.00
27.57
17.22
17.97
46.41
55.81
13.58

14.75
5.28
13.95
12.53
29.48
50.13
110.03

10
32
50/min
50/min
150/min

All First Grade students who participated in the study were from a single school. The majority of
students within the school district were White (78%), with the remaining students identified as either
Black (19%), or other (3%). Forty to fifty percent of students at each school were on free and reduced
lunch. For details regarding the demographic sample for this data collection, see Table 9. Coefficients
Page | 54
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
in Table 10 with larger sample sizes (i.e., larger than the above specified sample) were derived from a
convenience sample from the FAST™ database. This sample was approximately 85% White.
Teachers randomly selected three to five students and sent home passive consent forms. The first
administration of earlyReading measures were given by classroom teachers during a two-week
screening period. All teachers attended a two-hour training session on earlyReading measures and
were observed by the lead teacher at each school for a percentage of the time to confirm
administration integrity. The second administration of the earlyReading measures were given by a
team of school psychology graduate students. All graduate students also attended a two-hour
training session related to earlyReading administration. This second administration took place two to
three weeks after the termination of the initial screening period. Due to ongoing data collections, testretest reliability evidence has been provided for additional time intervals: fall (F) to winter (W), winter
to spring (S), and fall (F) to spring (S). Sample sizes vary by time interval. Test-retest reliabilitis are
reported in Table 10.
Table 10. Test-Retest Reliability for all earlyReading Screening Measures
Test-Retest Coefficient (N)
Measure

Grade

2-3 Weeks

F to W

W to S

F to S

Concepts of Print

K

.42 (39)

.66 (89)

.58 (90)

.51 (168)

Onset Sounds

K

.79 (67)

.75 (89)

.67 (90)

.58 (167)

Letter Names

K

.94* (45)

.65 (1781)

.55 (951)

.45 (1141)

Letter Sounds

K

.92 (75)

.51 (1241)

.61 (1282)

.35 (1600)

Rhyming

K

.74 (39)

.68 (917)

.62 (946)

.46 (1130)

Word Blending

K

.73 (70)

.59 (832)

.59 (856)

.34 (1069)

Word Segmenting

K

.86 (37)

--

.61 (834)

--

Decodable Words

K

.98 (29)

.70 (56)

.68 (168)

--

Nonsense Words

K

.94 (27)

.70 (119)

.74 (321)

--

Sight Words (50)

K

.97 (34)

--

.73 (169)

--

Composite

K

--

.91 (191)

.68 (185)

.71 (220)

Word Blending

1

.77 (67)

.61 (592)

.78 (579)

.54 (568)

Word Segmenting

1

.83 (77)

.52 (589)

.70 (582)

.48 (573)

Decodable Words

1

.97 (73)

.80 (2152)

.84 (604)

.69 (1194)

Nonsense Words

1

.76 (64)

.78 (1977)

.82 (439)

.65 (1046)

Sight Words (150)

1

.94 (74)

.84 (913)

.82 (432)

.60 (1137)

Sentence Reading

1

.98 (37)

Pending

Pending

Pending

Composite

1

.97 (33)

.90 (153)

.92 (104)

.88 (118)

Note. Sample size provided in parentheses. F = Fall. S = Spring. W = Winter.
*Outliers that were +/- 2 standard deviations from the mean were removed from the test-retest reliability sample.
In this case 2 cases, making up 3% of the sample, were removed.

Page | 55
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 11. Disaggregated Test Re-Test Reliability for earlyReading Measures
Test-Retest Coefficient (N)
Measure

Grade

2-3 Weeks

F to W

W to S

F to S

Letter Names

Ethnicity
Black

K

--

.69 (293)

--

.57 (274)

Letter Names

Hispanic

K

--

.61 (194)

--

.44 (179)

Letter Names

White

K

--

.61 (1129)

--

.40 (1293)

Letter Sounds

Black

K

--

.53 (409)

--

.45 (408)

Letter Sounds

Hispanic

K

--

.46 (270)

--

.29 (292)

Letter Sounds

White

K

--

.50 (1410)

--

.31 (1687)

Nonsense Words

Black

K

--

.52 (23)

--

.28 (38)

Nonsense Words

Hispanic

K

--

.73 (91)

--

.54 (102)

Onset Sounds

Black

K

--

.53 (424)

--

--

Onset Sounds

Hispanic

K

--

.51 (274)

--

--

Onset Sounds

White

K

--

.49 (1418)

--

--

Word Segmenting

Black

K

--

.61 (49)

--

.32 (52)

Word Segmenting

Hispanic

K

--

.54 (15)

--

.28 (20)

Word Segmenting

White

K

--

.61 (163)

--

.24 (230)

Word Blending

Black

K

--

.56 (219)

--

.34 (222)

Word Blending

Hispanic

K

--

.48 (129)

--

.30 (134)

Word Blending

White

K

--

.56 (533)

--

.33 (778)

Nonsense Words

Black

1

--

.74 (337)

--

.60 (179)

Nonsense Words

Hispanic

1

--

.74 (225)

--

.66 (121)

Nonsense Words

White

1

--

.78 (1156)

--

.63 (624)

Decodable Words

Black

1

--

.80 (375)

--

.73 (206)

Decodable Words

Hispanic

1

--

.75 (260)

--

.63 (138)

Decodable Words

White

1

--

.79 (1220)

--

.67 (707)

Sight Words (150)

Black

1

--

.84 (172)

--

.62 (194)

Sight Words (150)

Hispanic

1

--

.73 (123)

--

.52 (133)

Sight Words (150)

White

1

--

.87 (501)

--

.59 (674)

Word Segmenting

Black

1

--

.60 (171)

--

.53 (205)

Word Segmenting

Hispanic

1

--

.57 (128)

--

.52 (142)

Word Segmenting

White

1

--

.48 (447)

--

.46 (692)

Word Blending

Black

1

--

.66 (172)

--

.51 (205)

Word Blending

Hispanic

1

--

.58 (130)

--

.50 (142)

Word Blending

White

1

--

.48 (452)

--

.36 (705)

Inter-Rater Reliability
earlyReading measures involve a small degree of subjectivity, given clear scoring guidelines and
software-assisted scoring mechanisms. Unreliable scoring in regards to earlyReading may be the result
Page | 56
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
of clerical errors or differences in the interpretation of a student’s response. To alleviate such error,
examples and detailed responses are provided in the earlyReading Screening Administration
Technical Manual.
Evidence of inter-rater reliability is provided in Table 12. All coefficients represent Pearson productmoment correlation coefficients (Pearson r). For demographic information on the sample from which
inter-rater reliability coefficients were derived, see Table 5.
Table 12. Inter-Rater Reliability by earlyReading Subtest
Subtest
Onset Sounds
Letter Sounds
Letter Names
Word Blending
Word Segmenting
Sight Words (50)
Word Blending
Word Segmenting
Decodable Words
Sight Words (150)
Nonsense Words

Grade
K
K
K
K
K
K
1
1
1
1
1

Correlation Coefficient
.98
.99
.99
.98
.85
.99
.89
.83
.99
.97
.99

N
40
47
69
95
90
9
159
85
120
125
51

Reliability of the Slope
Data collected during a normative information-aimed study was used to determine reliability of the
slope for earlyReading measures. Participants included Kindergarten and First Grade students from
various elementary schools. Students were administered one or more earlyReading measures at three
time points throughout the school year (see below)
Table 13. Demographic Information for earlyReading Reliability of the Slope Sample
Category
Female

Kindergarten (N)
2086

1st Grade (N)
1604

Male

2196

1555

White

2710

2114

Black

688

429

Hispanic

439

288

Asian/Pacific Islander

322

252

Other
General Education

123
2656

77
1727

Special Education

1345

1296

Unspecified

1626

1433

Reliability of the slope was calculated for earlyReading screening and progress monitoring data. This
data is shown in Table 14. Reliability of the slope data has also been disaggregated by ethnicity.
Disaggregated information is also provided (See Table 15).
Page | 57
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 14. Reliability of the Slope for All earlyReading Screening Measures
Subtest
Onset Sounds
Letter Names
Letter Sounds
Rhyming
Word Blending
Word Blending
Word Segmenting
Word Segmenting
Decodable Words
Decodable Words
Sight Words (50)
Sight Words (150)
Nonsense Words
Nonsense Words

Grade
K
K
K
K
K
1
K
1
K
1
K
1
K
1

N
2129
1627
2229
904
958
824
235
824
52
918
167
624
116
664

Coefficient
.91
.81
.88
.38
.73
.77
.60
.78
.59
.86
.22
.77
.75
.87

Note. All of the above information is based on three time points: fall, winter, and spring data.

Page | 58
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 15. Reliability of the Slope for earlyReading measures, Disaggregated by Ethnicity
Subtest
Onset Sounds

Grade
K

Letter Sounds

K

Letter Names

K

Nonsense Words

K

Word Blending

K

Word Blending

1

Word Segmenting

K

Word Segmenting

1

Nonsense Words

1

Decodable Words

1

Sight Words (150)

1

N
342
253
1253
366
247
1332
256
177
1049
22
89
206
125
515
156
123
420
156
122
418
48
15
157
153
92
328
199
136
449
130
103
303

Coefficient
.90
.89
.92
.93
.86
.89
.80
.76
.83
.70
.81
.77
.57
.74
.93
.74
.77
.78
.77
.73
.60
.36
.65
.93
.89
.85
.88
.91
.83
.71
.85
.79

Ethnicity
Black
Hispanic
White
Black
Hispanic
White
Black
Hispanic
White
Black
White
Black
Hispanic
White
Black
Hispanic
White
Black
Hispanic
White
Black
Hispanic
White
Black
Hispanic
White
Black
Hispanic
White
Black
Hispanic
White

Note. All of the above information is based on three time points: fall, winter, and spring data.

CBMreading
The CBMreading passages in FAST™ have been systematically developed and field tested over a
number of years to address the problems with pre-existing passage sets that introduced error into
measurement of student reading rate during screening and progress monitoring. The goal in creating
the CBMreading measures was to systematically develop, evaluate, and finalize research-based
instrumentation and procedures for reliable assessment and evaluation of reading rate. Christ and
Ardoin (2009) described their initial method for FAIP-R field testing and passage-set development,
which was designed to minimize variance due to instrumentation/passages and optimize progress
Page | 59
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
monitoring reliability/precision. In a follow-up study, Ardoin and Christ (2009) directly compared
FAIP-R passages to DIBELS and AIMSweb passage sets in a peer-refereed publication. In the only
published study of its kind, they concluded that FAIP-R passages yielded less error and more precise
estimates than either AIMSweb or DIBELS—with the most substantial improvements over the DIBELS
system. Progress monitoring requires a set of equivalent reading passages, a schedule, graphing
procedures, and trend line or data-point decision rules due to the increased frequency of CBM-R
administration. The materials and decision rules guiding material use, selection, and schedule of
administration in CBMreading have all been developed with these core elements in mind.
The CBMreading passages in FAST™ were initially developed and field tested with 500 students per
level. All passages were designed with detailed specifications and in consultation with educators and
content experts. The researchers analyzed data from three rounds of field testing and edited passages
to optimize the semantic, syntactic, and cultural elements. Evidence of test-retest reliability and
reliability of the slope was derived from the same sample of participants.

Alternate Form Reliability
First Passage Reduction
Alternate forms of FAIP-R oral reading passages were administered in order to identify the best set of
passages to use when measuring oral reading rate with students at differing levels of ability in first
through Fifth Grades. Three passage sets of different difficulty level were constructed.
Student participants were from urban and suburban schools located in the Southeast, the upper
Midwest, and Northeastern regions of the US. A passive-consent procedure was used so that students
whose parents opted out of participation were excluded from the sample. The sample consisted of
177 participants from Kindergarten through Fifth Grades. Fifteen students were sampled from
Kindergarten and First Grades in the upper Midwest site. Across all three sites, 40 students were
selected from second and Third Grade classrooms (n=80) and 40 students were selected from fourth
and Fifth Grade classrooms (n=80; two extra participants were seen at the Northeastern site).
Information about participant characteristics in the overall and disaggregated sample is provided in
Table 16.

Page | 60
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 16. Demographic Information for CBMreading First Passage Reduction Sample
Grade

Sex
F

Overall
K
3
1
3
2
15
3
15
4
21
5
14
Southeast
2
10
3
6
4
8
5
4
Upper Midwest
K
3
1
3
2
7
3
5
4
8
5
4
Northeast
2
2
3
4
4
5
5
6

M

Age

Race
A

B

H

Am

W

SESa

Special
Education

2
7
21
27
21
26

6 years 5 months
7 years 2 months
7 years 11 months
8 years 10 months
10 years 2 months
10 years 10 months

0
0
1
1
2
3

1
1
7
3
7
4

0
0
3
1
0
1

0
0
1
0
0
1

4
9
28
37
33
31

0
1
14
14
12
10

0
0
0
0
1
1

3
11
9
11

8 years 1 month
8 years 11 months
10 years 10 months
11 years 1 month

0
0
0
0

2
0
2
2

0
0
0
0

0
0
0
0

11
17
15
13

1
0
0
0

0
0
0
0

2
7
8
10
7
11

6 years 5 months
7 years 2 months
8 years 4 months
9 years 3 months
10 years 1 month
11 years 0 months

0
0
1
0
2
0

1
1
0
1
3
2

0
0
1
0
0
1

0
0
0
0
0
0

4
9
13
14
10
12

0
1
1
4
2
0

0
0
0
0
0
1 (not specified)

10
6
5
4

7 years 5 months
8 years 4 months
9 years 7 months
10 years 6 months

0
1
0
3

5
2
2
0

2
1
0
0

1
0
0
1

4
6
8
6

12
10
10
10

0
0
1 (speech)
0

Note: A=Asian, B= Black, H= Hispanic, Am= American Indian, W= White.
a
SES indicates the number of students receiving free and reduced price lunch.

Experimenters worked individually with students in separate, quiet areas within the schools at each
site. At the beginning of each session the experimenter provided a general introduction from a
prepared set of directions. Students were assigned to particular levels of passages based on grade
level. Students in Kindergarten and First Grades read passages in Level A. Students in second and
Third Grades read passages in Level B and Fourth and Fifth Graders read passages in Level C. Passage
order was random within student; each student read all passages to facilitate analysis of passage
specific performances and prepare for equating and linking of passages within and across levels.
Students in Kindergarten through Fifth Grades were seen for approximately 10 successive days of
testing. On each day of testing, students in second through Fifth Grades read 12 different passages
and students in Kindergarten and First Grades read six passages each day. This resulted in students at
Level A reading 63 total passages, and students at Levels B and C reading 120 total passages (i.e., three
linking passages plus the number of progress monitoring passages for each level).

Page | 61
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
For each passage read, an appropriate paper version was placed in front of the student so s/he could
read aloud. The experimenter gave directions and then followed along and scored errors on his or her
paper or electronic copy for one minute. Each session lasted approximately 20 minutes. See Table 17
for descriptive statistics for the first passage reduction sample.
Table 17. Descriptive Statistics for First Passage Reduction
M

SD

All 60

66.63

33.08

Reduced 40

65.47

30.97

All 117

117.76

Reduced 80

Median

Trimmed

Min

Max

Skew

Kurtosis

SE

57.30

64.89

23.88

132.0

57.83

63.88

24.83

126.8

0.51

-0.97

8.54

0.50

-0.96

8.0

42.40

117.80

117.36

23.88

219.47

0.07

-0.26

5.43

114.92

43.90

116.20

114.92

10.00

225.0

-0.01

-0.07

5.0

All 117

153.59

40.19

151.09

152.84

69.42

251.74

0.21

-0.34

5.37

Reduced 80

151.55

40.33

146.84

149.70

70.00

251.0

0.40

-0.29

4.79

Level A Passages

Level B Passages

Level C Passages

Note. M = Mean; SD = Standard Deviation; Min = Minimum; Max = Maximum; SE = Standard Error

Second Passage Reduction

Screening. Screening took place at the beginning of the year. Due to time constraints, only one
of the three screening passages were used to collect alternate form reliability coefficients.
Experimenters worked individually with students in separate, quiet areas within the schools at
each site. At the beginning of each session, the experimenters provided an introduction
dialogue from a standardized set of directions. Each student was assessed during a single
session. The appropriate paper version of the screening story was placed in front of the student
so that s/he could read aloud. After instructions were delivered, the examiner followed along
and scored errors on his or her copy for one minute. Each student session lasted approximately
three to five minutes. Participants included students from urban and suburban schools located
in the Southeast, the upper Midwest, and Northeastern regions of the United States. A passive
consent procedure was used so that those students whose parents opted out of participation
were excluded from the sample. The sample consisted of 1,250 students from first through Fifth
Grades. Participant characteristics in the overall and disaggregated sample are provided in
Table 18.
Table 18. Demographic Information for Second Passage Reduction Sample
Grade
Level
Overall
K
1
2

Sex
F

M

5
159
154

5
157
164

Age

5 years 11 months
6 years 9 months
7 years 9 months

Page | 62
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Race
A

B

N/A*
10
51
9
41

H

Am

W

49
44

3
3

203
221

SESa

21
23

Special
Education

14
13

Section 6. FAST as Evidence-Based Practice
3
78
4
80
5
80
Southeast**
K
5
1
67
2
67
3
18
4
19
5
10
Upper Midwest
1
58
2
57
3
44
4
38
5
41
Northeast
1
34
2
30
3
16
4
23
5
29

118
107
83

8 years 7 months
9 years 8 months
10 years 6 months

3
5
7

40
21
33

43
47
36

2
1
2

108
113
85

33
18
23

14
20
15

5
83
84
30
12
16

5 years 11 months
6 years 11 months
7 years 10 months
8 years 10 months
9 years 11 months
11 years 2 months

N/A*
2
3
0
1
1

9
8
3
1
0

16
16
5
2
4

0
0
0
1
0

116
120
40
26
21

N/A
N/A
N/A
N/A
N/A

7
10
4
2
4

56
58
54
69
48

6 years 10 months
7 years 8 months
8 years 9 months
9 years 10 months
10 years 5 months

3
6
3
3
4

14
16
13
8
13

28
28
33
40
30

0
2
2
1
2

69
63
47
55
40

9
16
23
6
17

17
14
30
25
21

18
22
34
26
19

6 years 6 months
7 years 1 month
8 years 2 months
9 years 6 months
10 years 6 months

5
0
0
1
2

21
15
24
12
20

5
0
5
5
2

0
0
0
0
0

21
37
21
31
24

12
7
10
12
6

7
3
10
18
11

Note: A=Asian, B= Black, H= Hispanic, Am= American Indian, W= White.
a
SES indicates the number of students receiving free or reduced price lunch.
*Not all schools were able to provide complete information about Race, SES, or Special Education.
**About 5% of the southeast demographics were not available and are not included in this table.

Leveling/Anchor Process. On the first day of testing, three passages were sequentially administered to
the student at the beginning of the testing session. The experimenter immediately determined the
median score from these passages and compared it against the criterion points (established during
screening) to determine which level of progress monitoring passages to administer. Table 19 provides
the cut points used to assign students to passage difficulty level (i.e. Level A, B or C). Although
students were selected across grade level, student reading levels were restricted for each passage
level. Experimenters worked individually with students in separate, quiet areas within the schools at
each site.
Table 19. Cut-points Used for Assigning Students to CBMreading Passage Level Based on Words
Read Correct per Minute (WRC/min)
Level
A
B
C

Cut Point (WRC/min)
0 (5) to 20 (25)
26 to 70
71 to 140

Page | 63
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Process for Administering Progress Monitoring Passages. All students read all passages to facilitate
analysis of passage-specific performances and prepare for equating and linking of passages within
and across levels respectively. Experimenters worked individually with students in separate, quiet
areas within the schools. Students completed approximately eight successive assessment sessions.
On the first day of testing, after the child had been leveled according to the anchor passages, the
experimenter also administered four to nine progress monitoring passages at the student’s
appropriate level. All other days of testing were used to complete administration of the progress
monitoring set at the appropriate level. This resulted in students at Level A reading a total of 42
passages, and students at Levels B and C reading a total of 83 passages. All progress monitoring
passages were administered randomly to each student. For each passage read, an appropriate paper
version was placed in front of the student so that he or she could read aloud. The experimenter gave
directions and then followed along and scored errors on the electronic or paper copy for one minute.
Each session lasted approximately 15–20 minutes. See Table 20 for Descriptive Statistics.
Table 20. Descriptive Statistics for Second CBMreading Passage Reduction Sample
Passages

Mean

SD

Median

Trimmed

Min

Max

Skew

Kurtosis

SE

Level A
Reduced 39

20.38

10.46

18.33

19.39

4.54

56.18

0.94

0.87

1.04

Reduced 30

19.54

9.96

17.55

18.60

4.40

55.30

0.99

1.16

0.99

Level B
Reduced 80

52.20

16.40

52.19

51.92

15.96

102.65

0.20

-0.18

1.13

Reduced 60

51.24

15.83

50.90

50.93

16.00

101.00

0.24

-0.06

1.09

Reduced 30

52.33

15.34

52.08

52.13

17.00

99.50

0.17

-0.20

1.06

Level C
Reduced 80

103.04

26.09

103.05

102.65

42.24

195.83

0.20

-0.21

1.32

Reduced 60

102.96

25.25

102.95

102.54

43.87

194.53

0.21

-0.15

1.28

Reduced 30

104.68

24.41

104.70

104.29

46.80

195.37

0.22

-0.10

1.24

Table 21 summarizes information accumulated across several studies. Data was collected from three
states: Minnesota, New York, and Georgia. The information represents evidence for alternate form
reliability of CBMreading, and overall reliability of the performance level score.
Table 21. Alternate Form Reliability and SE m for CBMreading (Restriction of Range)
Coefficient a
Grade and Passage
Passage Level A (Grade 1)
Grade 1 (206)
Grade 2 (21)
Grade 3 (4)
Total N 231

# of
passages
39

Page | 64
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

# of
Weeks
<2

Range

Median

SEm

.62 - .86

.74

5.40

Section 6. FAST as Evidence-Based Practice
Passage Level B (Grades 2–3)
60
<2
.65 - .82 .75
8.54
Grade 1 (138)
Grade 2 (179)
Grade 3 (126)
Grade 4 (32)
Grade 5 (13)
Total N 488
Passage Level C (Grades 4–6)
60
<2
.78 - .88 .83
10.41
Grade 1 (3)
Grade 2 (135)
Grade 3 (79)
Grade 4 (156)
Grade 5 (140)
Total N 513
Passage Level A (Grade 1)
39
<2
.89 - .94 .92
3.03
Grade (206)
Grade (21)
Grade 3 (4)
Total N 231
Passage Level B (Grades 2–3)
60
<2
.87 - .92 .90
4.97
Grade 1 (138)
Grade 2 (179)
Grade 3 (126)
Grade 4 (32)
Grade 5 (13)
Total N 488
Passage Level C (Grades 4–6)
60
<2
.92 - .95 .94
7.06
Grade 1 (3)
Grade 2 (135)
Grade 3 (79)
Grade 4 (156)
Grade 5 (140)
Total N 513
Note: Alternate-Form Correlation – Individual Passages (Average Fisher z-transformed Inter-Passage Correlations). Sample
sizes by grade level are provided in parentheses. *Coefficients are reduced due to restriction of range. SEm estimates are
equivalent to published research (viz. Christ & Silberglitt, 2007; Wayman et al., 2007).

Internal Consistency (Item-Total Correlation)
Data collection for internal consistency is ongoing. Evidence of CBMreading internal consistency
across passages is provided in the table below. Similar to alternate-form reliability, information was
gathered across three states: Minnesota, New York, and Georgia. Table 22 provides evidence of the
reliability of the performance level score.

Page | 65
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 22. Internal Consistency for CBMreading Passages
# of
# of
Coefficient
Weeks
Passage and Grade
Passages
Range
Median
Passage Level A (Designed for Grade 1)
60
<2
.91 - .92
.92
Grade 1(206)
Grade 2 (21)
Grade 3 (4)
Total N 231
Passage Level B (Designed for Grades 2 & 3)
60
<2
.89 - .91
.90
Grade 1 (138)
Grade 2 (179)
Grade 3 (126)
Grade 4 (32)
Grade 5 (13)
Total N 488
Passage Level C (Designed for Grades 4-6)
60
<2
.88 - .93
.91
Grade 1 (3)
Grade 2 (135)
Grade 3 (79)
Grade 4 (156)
Grade 5 (140)
Total N 513
Note. Sample sizes by grade level are provided in parentheses.
a
Coefficients are reduced due to restriction of range. Participants were selected from a very narrow (low) ability range to
evaluate reliability with the intended population. SEm estimates are equivalent to those observed in published research (viz.
Christ & Silberglitt, 2007; Wayman et al., 2007).

See Table 23 below for evidence of split-half reliability for CBMreading passages.
Table 23. Split-Half Reliability for CBMreading passages
Grade
Grades 1st
Grades 2 & 3
Grades 4 to 6

N
500
500
500

Range
.90 to .98
.90 to .98
.90 to .98

Median
> .95
> .95
> .95

Test-Retest Reliability (Delayed)
Table 24 provides evidence of the reliability of the performance level score. Data was gathered across
three states: Minnesota, New York, and Georgia. The reliability range coefficients have been computed
with a 95% confidence interval. In addition, for the time lag, the mean number of weeks between data
collections is reported.

Page | 66
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 24. Evidence for Delayed Test-Retest Reliability of CBMreading
Grade
1
2
3
4
5
6
1
2
3
4
5
6

N
428
414
435
475
481
220
408
386
403
406
411
218

Time Period
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring

# Weeks Lag
Mean
18.68
18.56
18.98
19.00
19.00
17.45
35.57
35.93
35.79
35.67
35.67
35.01

SD
3.03
2.44
2.50
2.32
2.51
0.86
2.02
1.61
1.47
1.41
1.41
0.96

Coefficient
Range
.88 - .92
.91 - .94
.92 - .94
.93 - .95
.92 - .94
.92 - .95
.79 - .85
.87 - .91
.89 - .93
.91 - .94
.92 - .94
.90 - .94

Median
.90
.93
.93
.94
.93
.94
.82
.90
.91
.93
.93
.92

Note. SD = Standard Deviation.

Table 25 provides evidence of delayed test-retest reliability, disaggregated by ethnicity. The sample
included students from urban, suburban, and rural areas in Minnesota.

Page | 67
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 25. CBMreading Delayed Test-Retest Reliability Disaggregated by Ethnicity
# Weeks Lag
Grade
2
2
2
2
3
3
3
3
4
4
4
4
5
5
5
5
2
2
2
2
3
3
3
3
4
4
4
4
5
5
5
5

N
1518
369
210
308
1439
442
197
314
1384
353
204
268
1309
378
205
247
1518
369
210
308
1439
442
197
314
1384
353
204
268
1309
378
205
247

Time Period
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Winter
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring
Fall to Spring

14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
31
31
31
31
31
31
31
31
31
31
31
31
31
31
31
31

Coefficient
Range
.91 - .92
.90 - .94
.92 - .95
.91 - .94
.91 - .92
.90 - .93
.87 - .92
.91 - .94
.88 - .91
.89 - .92
.89 - .94
.87 - .92
.91 - .93
.91 - .94
.89 - .93
.91 - .95
.83 - .86
.75 - .83
.81 - .88
.76 - .84
.88 - .90
.82 - .87
.80 - .88
.80 - .87
.85 - .88
.82 - .88
.80 -.88
.81 - .88
.87 - .90
.77 - .84
.83 - .89
.84 - .90

Median
.91
.92
.94
.93
.92
.91
.90
.93
.90
.91
.92
.90
.92
.93
.91
.93
.85
.79
.85
.80
.89
.85
.84
.84
.87
.85
.85
.85
.88
.81
.86
.87

Ethnicity
White
Black
Asian
Hispanic
White
Black
Asian
Hispanic
White
Black
Asian
Hispanic
White
Black
Asian
Hispanic
White
Black
Asian
Hispanic
White
Black
Asian
Hispanic
White
Black
Asian
Hispanic
White
Black
Asian
Hispanic

Inter-Rater Reliability
Inter-rater reliability evidence was collected across three states: Minnesota, New York, and Georgia.
See Table 26.

Page | 68
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 26. Evidence of Inter-Rater Reliability for CBMreading
Passage
Passage Level A

Sample Size
146

Coefficient
Range
.83 – 1.00

Median
.97

Passage Level B
Passage Level C

1391
1345

.93 - 97
.83 – 1.00

.97
.98

Reliability of the Slope
Some may argue that alternate-form reliabilities may not accurately capture reliability of the slope due
to the small amount of variation in slope values, represented by low standard error of the estimate
and standard error of the slope values. This might be a result of the structure of passage
administration (Levels vs. Grades). By using passage levels as groups instead of grades, we may be
reducing variability within grades, decreasing the reliability of slope estimates. The following analysis
was conducted using HLM 7 software and used random slopes and random intercepts (See Table 27).
Table 27. Reliability of the Slope for CBMreading
Weeks

Observations

Coefficient
Range
Median

N=34

~ 27 - 30

~25-30

NA

.95

N=39

~ 7 -10

~7-10

NA

.78

~ 27 - 30

~25-30

NA

.98

~ 7 -10

~7-10

NA

.97

~ 27 - 30

~25-30

NA

.98

~ 7 -10

~7-10

NA

.97

Sample Size by Passage
Passage Level A

Passage Level B
N=53

Passage Level C

Table 28 provides a summary for reliability of the slope by grade (passage) level. Reliability of the
slope for multi-level analyses may be biased when standard error of the estimate and standard error of
the slope is minimal. CBMreading growth estimates are less prone to error than comparable progress
monitoring materials. As a result, increased precision (less error) is paradoxically detrimental to multilevel reliability estimates (Raudenbush & Bryk, 2002). In such circumstances, the spearman brown
correlation is more appropriate. The following information includes participants across three states:
Minnesota, New York, and Georgia.

Page | 69
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 28. Reliability of the Slope of CBMreading by Passage using Spearman-Brown Split Half
Correlation
Passage Level
Passage Level A
Grade 1 (68)
Grade 2 (12)
Grade 3 (2)
Total N = 82
Passage Level B
Grade 1 (7)
Grade 2 (72)
Grade 3 (53)
Grade 4 (12)
Grade 5 (6)
Grade 6 (1)
Total N = 151
Passage Level C
Grade 2 (3)
Grade 3 (31)
Grade 4 (81)
Grade 5 (68)
Grade 6 (28)
Total N = 211
Passage Level A
Grade 1 (42)
Grade 2 (15)
Grade 3 (4)
Total N = 61
Passage Level B
Grade 1 (6)
Grade 2 (41)
Grade 3 (38)
Grade 4 (15)
Grade 5 (8)
Grade 6 (1)
Total N = 109
Passage Level C
Grade 2 (2)
Grade 3 (19)
Grade 4 (49)
Grade 5 (44)
Grade 6 (23)
Total N = 137

Weeks (range)
Short Term
10

Coefficient
.71

SEm
.40

Short Term
10 - 20

.74

.31

Short Term
6 - 20

.65

.30

Long Term
14 - 30

.95

.21

Long Term
14 - 30

.70

.31

Long Term
18 - 30

.66

.32

Note. Sample sizes by grade level are provided in parentheses. SEm = |Even – Odd| / sqrt(2)

Page | 70
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 29 provides reliability of the slope evidence derived from multi-level analyses, strictly using true
slope variance divided by total slope variance. The following information includes participants across
three states: Minnesota, New York, and Georgia.
Table 29. Reliability of the Slope for CBMreading by Passage Using Multi-Level Analyses
Passage Level

Weeks (range)

Coefficient

Passage Level A
Grade 1 (68)
Grade 2 (12)
Grade 3 (2)

Short Term
10

.75

Short Term
10–20

.74

Short Term
6–20

.63

Long Term
14–30

.94

Long Term
14–30

.86

Long Term
18–30

.45

Total N = 82
Passage Level B
Grade 1 (7)
Grade 2 (72)
Grade 3 (53)
Grade 4 (12)
Grade 5 (6)
Grade 6 (1)

Total N = 151
Passage Level C
Grade 2 (3)
Grade 3 (31)
Grade 4 (81)
Grade 5 (68)
Grade 6 (28)

Total N = 211
Passage Level A
Grade 1 (42)
Grade 2 (15)
Grade 3 (4)

Total N = 61
Passage Level B
Grade 1 (6)
Grade 2 (41)
Grade 3 (38)
Grade 4 (15)
Grade 5 (8)
Grade 6 (1)

Total N = 109
Passage Level C
Grade 2 (2)
Grade 3 (19)
Grade 4 (49)
Grade 5 (44)
Grade 6 (23)

Total N = 137

Page | 71
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 30 provides evidence of reliability of the slope disaggregated by ethnicity. Participants included
those from Minnesota. Reliability of the slope for multi-level analyses may be biased when few
observations are used to estimate slope. In this instance, slopes were estimated from tri-annual
assessments (3 observations). Coefficients should be interpreted with caution (Raudenbush & Bryk,
2002). The sample included students from urban, suburban, and rural areas of Minnesota.
Table 30. CBMreading Reliability of the Slope - Disaggregated Data
Grade

N (range)

Grades 2–5
Grades 2–5
Grades 2–5
Grades 2–5

1308–1518
353–442
197–210
247– 314

Coefficient
Range
.25 - .43
.32 - .60
.38 - .52
.21 - .52

Median
.28
.43
.40
.45

Ethnicity
White
Black
Asian
Hispanic

aReading
The following sections provide a discussion of types of reliability obtained for aReading, as well as
sources of error. Data collection regarding reliability of the slope is ongoing.

Alternate-Form Reliability
Given the adaptive nature of aReading tests, a proxy for alternate-form reliability is provided by
Samejima (1994), based on the standard error of measurement of an instrument. Using this proxy, the
alternate-forms reliability coefficient for aReading is approximately .95 (based on approximately 2,333
students).

Internal Consistency (Item-Total Correlations)
Given the adaptive nature of aReading tests, a proxy for internal consistency is provided by Samejima
(1994), based on the standard error of measurement of an instrument. Using this proxy, the internal
consistency reliability coefficient for aReading is approximately .95 (based on approximately 2,333
students).

Test-Retest Reliability
Three month test-retest reliability resulted in the following coefficients for 2,038 students in grades 1 5 (Kindergarten and grades 6–12 results are coming soon). Growth was measured four times over the
academic year. The results by grade: one .71, two .87, three .81, four .86, five .75.

Chapter 2.6: Validation
earlyReading
Evidence for validity of the earlyReading subtest measures was examined using the Group Reading
Assessment and Diagnostic Evaluation (GRADE; Williams, 2001). The GRADE™ is a norm-referenced
diagnostic reading assessment that assists teachers in measuring pre-literacy, emerging reading and
core reading skills, as well as providing teachers with implications for instruction and intervention.

Page | 72
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Content Validity
The design specifications for earlyReading measures relate directly to their evidence of content
validity. Each subtest was designed with the intent to address specific criteria aimed to maximize both
utility and sensitivity.

earlyReading

Measures
Concepts of Print

F

K

W

S

F

1st
W

S

Common Core
State Standards

Letter Names

RF.K1, RF.K.1.a, RF.K.1.b, RF.K.1.c,
RF.1.1, F.1.1.a
RF.K.1.d

Letter Sounds

RF.K.3.a

Decodable Words

O

O

O

O

O

Nonsense Words

R.F.K.3, RF.1.3, RF.1.3.b, RF.2.3,
RF.3.3
R.F.K.3, RF.1.3, RF.1.3.b, RF.2.3,
RF.3.3

Sight Words (50)
Sight Words (150)

RF.K.3.c, RF.1.3.g, R.2.3.f, RF.3.3.d

Sentence Reading
(CBM W, S)

RF.K.4, RF.1.4, RF.1.4.b, RF.2.4,
RF.2.4.b, RF.3.4

Onset Sounds

Available upon
request

RF.K.2.c, RF.K.2.D, RF.1.2.c

Rhyming

RF.K.2.a

Word Blending

RF.K.2.b, RF.K.2.c, RF.1.2.b

Word Segmenting

RF.K.2.b, RF.K.2.d, RF.1.2.c,
RF.1.2.d

Oral Repetition
Composite Broad Score

State Specific
Standards

SL.K.6, SL.1.6
– recommended screening tools and composition of broad composite score
O – Optional measure to replace nonsense words

Criterion-Related Validity
Criterion-related validity of earlyReading subtests was examined using the GRADE™. The GRADE is an
untimed, group-administered, norm-referenced reading achievement test that is intended for children
in preschool through grade 12. Comprised of 16 subtests categorized within five components, the
GRADE utilizes particular subtest scores, depending on the testing level, to form the Total Test
composite score. Evidence for the validity of earlyReading is presented below on the external criterion
measure of the GRADE Total Test composite score. Validity is most often represented as a correlation
between the assessment and the criterion. Both concurrent and predictive validity are reported for all
earlyReading measures, where available (See Table 33).

Page | 73
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
In order to establish criterion-related validity, students were recruited from school districts. In School
District 1, three elementary schools participated. Kindergarten students from District 1 who
participated in the study were enrolled in all-day or half-day Kindergarten. The majority of students
within the school district were White (78%), with the remaining students identified as either Black
(19%), or other (3%). Forty to fifty percent of students at each school were eligible for free and reduced
lunch. In school District 2, the majority of students within the school district were White (53%), with
the remaining students identified as Black (26%), Hispanic (11%), Asian (8%), or other (2%). Forty to
fifty percent of students at each school are on free and reduced lunch.
Students in Kindergarten were administered FAST™ earlyReading Concepts of Print, Onset Sounds,
Letter Naming, Letter Sound, Rhyming, Word Blending, Word Segmenting, Sight Words (50), Nonsense
Words, and Decodable Words subtests. Students in First Grade were administered FAST™ earlyReading
Word Blending, Word Segmenting, Sight Words (150), Decodable Words, Nonsense Word, and
Sentence Reading subtests. Teachers administered six to nine measures at each screening period (fall,
winter, and spring). See Table 31 for demographic information about the sample from which
earlyReading composite validity coefficients were derived, including predictive and concurrent
validity. Sample-related information for criterion-related validity data is provided in
Table 32. Predictive and concurrent validity coefficients are reported in Table 33.
Table 31. Demographics for Criterion-Related Validity Sample for earlyReading Composite
Scores
Category

District A

District B

White
Black
Hispanic
Asian/Pacific Islander
Other
Free/Reduced Lunch

78%
19%
--3%
40-50%

53%
26%
11%
8%
2%

Table 32. Sample-Related Information for Criterion-Related Validity Data (earlyReading)
Measure
Kindergarten
Concepts of Print
Onset Sounds
Letter Naming
Letter Sound
Rhyming
Word Blending
Word Segmenting
Decodable Words
Nonsense Words

Fall
N

Mean

SD

Winter
N
Mean

SD

Spring
N
Mean

SD

230
230
230
230
230
230
91
---

8.41
12.28
28.57
15.56
9.47
3.19
6.34
---

2.41
4.17
15.60
11.69
4.97
3.67
9.68
---

58
216
210
210
224
227
228
-2281

2.14
2.04
13.35
13.74
4.62
3.36
12.89
-6.73

-155
230
230
229
228
228
228
229

-1.61
18.40
15.51
3.15
1.79
6.17
15.15
10.71

Page | 74
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

9.43
15.06
39.51
29.20
12.01
6.76
19.35
-7.19

-15.70
54.80
43.10
14.28
9.14
30.11
16.04
13.16

Section 6. FAST as Evidence-Based Practice
Sight Words 50
-----Composite
173 40.78 9.91
173
49.41
First Grade
Word Blending
179 7.42
2.41
169
8.80
Word Segmenting
179 26.84 4.17
168
30.10
Decodable Words
179 14.43 15.60
168
26.19
Nonsense Words
179 10.91 11.69
166
20.69
Sight Words 150
179 34.57 4.97
168
60.10
Sentence Reading
179 43.89 3.67
30
67.62
CBM-R
---183
75.48
Composite
100 39.06 10.39
100
48.07
(median
of
3)
1
Note. Nonsense words in the winter used partially imputed values.

Page | 75
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

-13.74

227
173

44.29
59.79

29.38
14.37

2.14
2.04
13.35
13.74
4.62
3.36
42.55
10.50

172
172
186
131
173
-188
100

9.30
30.91
40.80
24.26
77.03
-104.00
62.87

1.38
4.40
21.12
14.31
21.93
-43.19
12.94

Section 6. FAST as Evidence-Based Practice
Table 33. Concurrent and Predictive Validity for all earlyReading Measures
Type of Validity

Grade

Criterion

N

Coefficient

Information

Concurrent

K

GRADEP

85

.62

Data collected in Fall

Predictive

K

GRADE

K

230

.55

Fall to Spring prediction

Predictive

K

GRADE

K

216

.60

Winter to Spring prediction

Concurrent

K

GRADE

K

140

.03

Data collected in Spring

Concurrent

K

GRADEP

85

.41

Data collected in Fall

Predictive

K

GRADE

K

230

.47

Fall to Spring prediction

Predictive

K

GRADE

K

210

.63

Winter to Spring prediction

Concurrent

K

GRADE

K

214

.18

Data collected in Spring

Concurrent

K

GRADEP

85

.53

Data collected in Fall

Predictive

K

GRADE

K

230

.44

Fall to Spring prediction

Predictive

K

GRADE

K

210

.63

Winter to Spring prediction

Concurrent

K

GRADE

K

214

.19

Data collected in Spring

Predictive

K

GRADEK

227

.66

Winter to Spring prediction

Predictive

K

GRADEK

230

.41

Fall to Spring prediction

Concurrent

K

GRADEK

213

.23

Data collected in Spring

Concurrent

1

GRADE1

71

.22

Data collected in Fall

Predictive

1

GRADE

1

179

.56

Fall to Spring prediction

Predictive

1

GRADE

1

169

.53

Winter to Spring prediction

Concurrent

1

GRADE

1

165

.12

Data collected in Spring

Predictive

K

GRADEK

228

.58

Winter to Spring prediction

Concurrent

K

GRADE

K

213

.25

Data collected in Spring

Concurrent

1

GRADE

1

71

.49

Data collected in Fall

Predictive

1

GRADE

1

179

.32

Fall to Spring prediction

Predictive

1

GRADE

1

168

.60

Winter to Spring prediction

Concurrent

1

GRADE

1

165

.07

Data collected in Spring

Concurrent

K

GRADEK

214

.27

Data collected in Spring

Concurrent

1

GRADE

1

71

.22

Data collected in Fall

Predictive

1

GRADE

1

179

.59

Fall to Spring prediction

Predictive

1

GRADE

1

168

.78

Winter to Spring prediction

Concurrent

1

GRADE

1

124

.46

Data collected in Spring

K

GRADEK

213

.19

Data collected in Spring

Onset Sounds

Letter Names

Letter Sounds

Word Blending

Word Segmenting

Decodable Words

Sight Words 50
Concurrent

Page | 76
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Type of Validity

Grade

Criterion

N

Coefficient

Information

Concurrent

1

GRADE1

71

.59

Data collected in Fall

Predictive

1

GRADE

1

179

.66

Fall to Spring prediction

Predictive

1

GRADE

1

168

.80

Winter to Spring prediction

Predictive

1

GRADE

1

179

.66

Fall to Spring prediction

Concurrent

1

GRADE

1

166

.43

Data collected in Spring

Predictive

K

GRADEK

105

.44

Winter to Spring prediction

Concurrent

K

GRADE

K

215

.27

Data collected in Spring

Predictive

1

GRADE

1

179

.60

Fall to Spring prediction

Predictive

1

GRADE

1

168

.67

Winter to Spring prediction

Concurrent

1

GRADE

1

179

.43

Data collected in Spring

Predictive

K

GRADEK

173

.68

Fall to Spring prediction

Predictive

K

GRADE

K

173

.69

Winter to Spring prediction

Concurrent

K

GRADE

K

173

.67

Data collected in Spring

Predictive

1

GRADE

1

100

.72

Fall to Spring prediction

Predictive

1

GRADE

1

100

.81

Winter to Spring prediction

Concurrent

1

GRADE

1

100

.83

Data collected in Spring

Sight Words 150

Nonsense Words

Composite

Note. All criterion coefficients were determined using the composite of the GRADE. Level is indicated in
superscript. For example, GRADEP represents GRADE Composite Level P.

More recently, criterion-related validity analyses was estimated for spring earlyReading composite
scores to predict spring aReading scores. These findings are summarized in the table below. Students
were recruited from several school districts in Minnesota. Cut score was selected by optimizing
sensitivity at about .70 and balancing sensitivity with specificity (Silberglitt & Hintze, 2005). In the table
below, dashes indicate unacceptable sensitivity and specificity due to low AUC. Criterion coefficients
ranged from .74 to .77.
Table 34. Criterion Validity of Spring earlyReading Composite (Updated weighting scheme) with
Spring aReading: MN LEA 3 (Spring Data Collection)
Grade

N

Composite

Some Risk (<40Mth (SD)
percentile)

aReading

r(x,y)

Cut

AUC

Sens.

Spec.

M (SD)

KG

515

62.41 (12.53)

415.59 (27.13)

.74**

64.5

.87

.77

.80

1

169

56.08 (15.79)

459 (27.34)

.77**

--

.65

--

--

High Risk (<20 th percentile)
KG

515

62.41 (12.53)

415.59 (27.13)

.74**

61.5

.84

.76

.73

1

169

56.08 (15.79)

459 (27.34)

.77**

51.5

.72

.70

.59

Page | 77
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Predictive Validity of the Slope
Validity of earlyReading subtests were examined using the GRADE. Table 31 presents the
demographic information from which the sample was derived, as this sample served multiple
purposes in establishing validity evidence for earlyReading.
Table 35 presents the correlation between the slope of performance using screening data (i.e.,
students were assessed three times per year, fall, winter and spring) and performance on the GRADE.
All correlations account for initial level of performance.
Table 35. Predictive Validity of the Slope for All earlyReading Measures
Measure

Grade

Criterion

N

Coefficient

Onset Sounds

K

GRADEK

217

.29

Letter Names

K

GRADEK

231

.44

Letter Sounds

K

GRADEK

231

.54

Word Blending

K

GRADEK

230

.48

Word Blending

1

GRADE1

178

.16

Word Segmenting

K

GRADEK

224

.49

Word Segmenting

1

GRADE1

178

.23

Decodable Words

1

GRADE1

179

.62

Sight Words (150)

1

GRADE1

180

.59

Nonsense Words

1

GRADE1

174

.61

Note. All coefficients were determined using the composite of the GRADE. Level is indicated in superscript. For
example, GRADEP represents GRADE Composite Level P.

Discriminant Validity
See Table 13 for demographic information on the sample. This study provided data for both reliability
of the slope and discriminant validity evidence for earlyReading measures. Table 36 and Table 37
display discriminant validity for earlyReading subtests in Kindergarten and First Grade, respectively.
Table 36. Discriminant Validity for Kindergarten earlyReading Measures
Measure by
Time of Year

Below 40th Percentile
N
Mean SD

Above 40th Percentile
N
Mean SD

Difference Stats
t
d

Concepts of Print
Beginning

204

6.42

1.60

240

10.29

1.08

30.24

2.88

Middle

-

-

-

-

-

-

-

-

End

-

-

-

-

-

-

-

-

Beginning

185

7.78

3.41

259

15.09

1.04

32.46

3.09

Middle

417

14.78

2.55

0

NA

NA

-

-

End

-

-

-

-

-

-

-

-

Beginning

182

10.38

7.76

262

38.68

9.01

34.42

3.27

Middle

182

25.45

11.34

242

49.20

3.54

30.66

2.99

Onset Sounds

Letter Naming

Page | 78
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
End

193

34.62

12.55

271

65.98

28.85

14.17

1.32

Beginning

187

3.77

3.48

257

23.81

10.40

25.33

2.41

Middle

173

14.16

7.85

251

36.82

8.58

27.66

2.69

End

212

26.13

10.65

252

51.85

10.90

25.59

2.38

Beginning

191

4.34

2.73

253

13.12

2.33

36.50

3.47

Middle

194

8.29

4.17

222

15.41

0.74

25.00

2.46

End

224

11.38

3.97

238

16.00

0.00

17.95

1.67

Beginning

209

0.00

0.00

234

5.85

2.92

28.96

2.76

Middle

179

2.29

2.18

254

8.93

1.09

41.72

4.02

End

206

6.50

2.97

256

10.00

0.00

18.86

1.76

Beginning

-

-

-

-

-

-

-

-

Middle

174

5.63

5.68

260

29.04

4.31

48.73

4.69

End

194

21.50

9.34

268

33.03

1.10

20.03

1.87

Beginning

-

-

-

-

-

-

-

-

Middle

-

-

-

-

-

-

-

-

End

184

2.72

2.43

275

23.32

16.03

17.29

1.62

Beginning

-

-

-

-

-

-

-

-

Middle

107

1.01

1.01

142

12.23

9.47

12.20

1.55

End

191

3.53

2.86

269

19.80

12.33

17.96

1.68

Beginning

-

-

-

-

-

-

-

-

Middle

-

-

-

-

-

-

-

-

End

177

9.58

7.29

253

57.66

22.59

27.33

2.64

Letter Sound

Rhyming

Word Blending

Word Segmenting

Decodable Words

Nonsense Words

Sight Words (50)

Table 37. Discriminant Validity for First Grade earlyReading Subtests
Measure by
Time of Year
Word Blending
Beginning
Middle
End
Word Segmenting
Beginning
Middle
End
Decodable Words
Beginning
Middle
End
Nonsense Words
Beginning
Middle
End

Below 40th Percentile
N
Mean
SD

Above 40th Percentile
N
Mean
SD

Difference Stats
t
d

253
276
-

4.39
7.78
-

2.47
2.10
-

357
344
-

9.13
10
-

0.81
0
-

33.79
19.61
-

2.74
1.58
-

252
278
269

17.89
27.38
26.84

7.62
5.47
5.60

358
340
356

30.67
33.22
33.35

2.15
10.79
0.78

30.09
8.23
21.66

2.44
0.66
1.74

256
265
262

2.92
11.29
17.88

1.93
4.92
7.74

343
353
386

19.65
34.22
51.37

13.78
10.79
15.19

19.28
32.17
32.9

1.58
2.59
2.59

283
221
240

3.88
10.39
12.78

2.29
3.82
5.00

316
292
343

16.06
26.98
35.31

9.02
10.82
12.52

22.09
21.79
26.44

1.81
1.93
2.19

Page | 79
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Sight 150
Beginning
Middle
End
Sentence Reading
Beginning
Middle
End

248
191
253

6.48
30.66
48.48

4.06
15.19
17.54

351
268
379

44.94
73.71
88.47

19.92
15.90
14.44

29.96
29.13
31.27

2.45
2.73
2.49

223
-

12.77
-

4.98
-

332
-

57.05
-

37.68
-

17.44
-

1.48
-

CBMreading
Evidence for validity of the CBMreading passages was examined using the Test of Silent Reading
Efficiency and Comprehension (TOSREC), the Group Reading Assessment and Diagnostic Evaluation
(GRADE), Measures of Academic Progress (MAP), AIMSweb Reading CBM (R-CBM), and the Dynamic
Indicators of Basic Early Literacy Skills (DIBELS) Next. The TOSREC is a brief test of reading that assesses
silent reading of connected text for comprehension. The TOSREC can be used for screening purposes,
as well as for monitoring progress. The GRADE is a norm-referenced diagnostic reading assessment
that assists teachers in measuring pre-literacy skills, emerging reading skills, and core reading skills, as
well as providing teachers with implications for instruction and intervention. MAP is a compilation of
computerized adaptive assessments used to benchmark student growth and to serve as a universal
screener. AIMSweb is web-based and may be used for universal screening, monitoring progress, and
managing data for students in Kindergarten through twelfth grade. Like CBMreading, AIMSweb
Reading CBM probes are intended to measure oral reading fluency and provide an indicator of general
reading achievement. Finally, DIBELS Next are a set of procedures and measures for assessing the
acquisition of early literacy skills from Kindergarten through Sixth Grade. DIBELS Next was designed to
serve as a measure of oral reading fluency and a method for monitoring the development of early
literacy and early reading skills. Data collection to gather evidence of discriminant validity is ongoing.

Content Validity
The design specifications for CBMreading relate directly to their evidence of content validity. Each
passage set was designed with the intent to address specific criteria aimed to maximize both utility
and sensitivity. Specific guidelines were provided for paragraph and sentence structure. This was
necessary to ensure a parallel text structure across the passages. Each writer was instructed to use
three or four paragraphs within each passage and, when possible, include a main idea sentence at the
beginning of each paragraph that would introduce and help organize content for the reader. Writers
were also instructed to not use complex punctuation such as colons and semi-colons in order to
reflect text that is familiar to primary grade levels as well as to encourage a more direct style of writing.
The passages developed for the Grade 1 passages (Level A) could include 150–200 words overall in 2–
5 paragraphs. Sentences were structured to stay within a range of 3–7 words. Each paragraph was
strictly structured to stay within a range of 7–15 sentences. The number of words per sentence and
sentences per paragraph were varied across the story to result in the appropriate total number of
words.
Page | 80
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Passages developed for Grades 2 and 3 (Level B) could include between 230–300 words overall.
Sentences were structured to stay within a range of 6–11 words. Each paragraph was strictly
structured to stay within a range of 3–7 sentences. The number of words per sentence and sentences
per paragraph were varied across the story to result in the appropriate total number of words (i.e.,
230–300).
The passages developed for Grades 4 through 6 (Level C) could include between 240–300 words
overall. Sentences were strictly structured to stay within a range of 7–11 words. Each paragraph was
strictly structured to stay within a range of 3–7 sentences. In addition, the 2nd or 3rd sentence of each
paragraph of these passages was required to be a longer sentence, specifically from 12–19 words in
length. Once again, the number of words per sentence and sentences per paragraph were varied
across the story to result in the appropriate total number of words (i.e., 240–300). Overall evidence
supports that guidelines for development were accurately addressed and provided the FAST™ team
with passages that were consistent at each level in the full and reduced passage sets.

Criterion-Related Validity
Predictive and concurrent criterion validity for each grade level are available using a number of
different tests or criterion (i.e., TOSREC, MAP, AIMSweb and DIBELS Next), providing evidence of
criterion-related validity. Where applicable, the delay between CBMreading administration and
criterion administration is stated. Students scoring in the lower 40th percentile during screening were
assigned to either the long- or short-term condition. Approximately 20% of the students were
targeted within Level 1, 40% in Level 2, and 40% in Level 3. When possible, participants were selected
to ensure they read at the lower end of each score range for Level 1, 2, and 3, respectively. This
methodological constraint ensured that the students wouldn’t grow out of the range of equitable
scores across the time of data collection, which spanned two years.
Concurrent and Predictive Validity for CBMreading Grade-Level Passages is provided in Table 38. All
coefficients were derived from students across three states: Minnesota, New York, and Georgia.
Table 38. Concurrent and Predictive Validity for CBMreading
Type of
Validity

Grade

Criterion

N

Concurrent

1
2
3
4
5
6
1
2
3
4
5
6
1

TOSREC

218
246
233
228
244
222
399
463
483
485
503
225
399

Concurrent

Concurrent

DIBELS
NEXT

AIMSweb

Page | 81
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Time Lapse (Weeks)
M
SD

Coefficient
.86
.81
.81
.79
.81
.82
.95
.92
.96
.95
.95
.95
.95

Section 6. FAST as Evidence-Based Practice

Concurrent

Predictive

Predictive

Predictive

Predictive

2
3
4
5
6
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
1
2
3
4
6
2
3
4
5
6

MAP

AIMSweb

DIBELS
Next

TOSREC

MAP

425
402
445
447
229
237
231
233
219
212
385
413
391
427
431
220
425
80
76
74
85
44
35
33
35
18
240
233
235
220
212

18.68
18.56
18.98
19
19
17.45
35.57
35.93
35.79
35.67
35.67
35.57
35.94
35.79
35.67
35.06
35.23
35.47
35.23
35.29
35.06

.97
.95
.96
.96
.95
.81
.78
.73
.66
.69
.91
.93
.91
.94
.93
.94
.82
.74
.91
.90
.93
.47
.56
.69
.52
.87
.76
.73
.69
.65
.71

3.04
2.44
2.50
2.32
2.51
.86
2.02
1.61
1.47
1.41
1.41
2.02
1.61
1.48
1.41
.96
1.42
1.27
.88
1.13
.96

Note. SD = Standard Deviation; M = Mean.

Additional criterion-related validity evidence for CBMreading is summarized below.
Table 39. Criterion Validity of Spring CBMreading with Spring CRCT in Reading: GA LEA 1
(Spring Data Collection)
Grade

N

CBMreading
CRCT
M (SD)
M (SD)
Some Risk (Meets Standards)
3
324
115.74 (43)
848.62 (28)
4
310
135.74 (39)
848.30 (27)
5
343
148.01 (38)
841.27 (25)
High Risk (Does Not Meet Standards)
3
324
115.74 (43)
848.62 (28)
4
310
135.74 (39)
848.30 (27)
5
343
148.01 (38)
841.27 (25)

Page | 82
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

r(x,y)

Cut

AUC

Sens.

Spec.

.61*
.61*
56*

113.50
131.50
151.50

.77
.79
.73

.71
.71
.69

.70
.73
.69

.61*
.61*
56*

79.00
99.50
122.50

.89
.89
.81

.80
.83
.77

.84
.83
.77

Section 6. FAST as Evidence-Based Practice
Table 40. Criterion Validity of Spring CBMreading with Spring MCA-III in Reading: MN LEA 4
(Spring Data Collection)
Grade

CBM-R
M (SD)

N

MCA-III
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

Some Risk (Does not Meet or Partially Meets Standards)
3

852

139 (40)

348 (20)

.76

141.5

.86

.78

.76

4

818

165 (39)

447 (15)

.71

164.5

.83

.75

.71

5

771

165 (40)

552 (16)

.70

163.5

.84

.77

.76

High Risk (Does Not Meet Standards)
3

852

139 (40)

348 (20)

.76

131.5

.88

.80

.79

4

818

165 (39)

447 (15)

.71

153.5

.87

.80

.78

5

771

165 (40)

552 (16)

.70

151.5

.89

.80

.79

Table 41. Criterion Validity of Spring CBMreading with Spring MCA-III in Reading: MN LEA 3
(Spring Data Collection)
Grade

N

CBMreading
M (SD)

MCA-III
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

Some Risk (Does Not Meet or Partially Meets Standards)
3

502

137.30 (44)

352.97 (22)

.74

131.5

.87

.80

.79

4

505

160.98 (48)

451.54 (18)

.74

161.5

.87

.78

.78

5

505

172.70 (41)

554.09 (15)

.70

164.5

.85

.76

.75

6

472

176.57 (41)

654.35 (19)

.64

173.5

.81

.72

.73

Some Risk (Partially Meets Standards)
3

502

137.30 (44)

352.97 (22)

.74

--

.58

--

--

4

505

160.98 (48)

451.54 (18)

.74

165.5

.63

.70

.55

5

505

172.70 (41)

554.09 (15)

.70

172.5

.66

.69

.53

6

472

176.57 (41)

654.35 (19)

.64

184.5

.60

.69

.56

High Risk (Does Not Meet Standards)
3

502

137.30 (44)

352.97 (22)

.74

121.5

.91

.82

.82

4

505

160.98 (48)

451.54 (18)

.74

144.5

.88

.78

.79

5

505

172.70 (41)

554.09 (15)

.70

151.5

.91

.83

.81

6

472

176.57 (41)

654.35 (19)

.64

165.5

.87

.77

.78

Page | 83
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 42. Criterion Validity of Spring CBMreading with Spring Minnesota Comprehensive
Assessment III (MCA-III) in Reading: MN LEA 2 (Spring Data Collection)
Grade

N

CBMreading
M (SD)

MCA
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

Some Risk (Does Not Meet or Partially Meets Standards)
3

252

142.95 (39)

353.34 (23)

.69**

142.5

.83

.76

.77

4

240

165.98 (39)

451.83 (16)

.69**

165.5

.82

.75

.75

5

234

175.33 (35)

558.69 (15)

.62**

167.5

.83

.73

.72

Some Risk (Partially Meets Standards)
3

252

142.95 (39)

353.34 (23)

.69**

--

.58

--

--

4

240

165.98 (39)

451.83 (16)

.69**

--

.58

--

--

5

234

175.33 (35)

558.69 (15)

.62**

174.5

.66

.67

.54

High Risk (Does Not Meet Standards)
3

252

142.95 (39)

353.34 (23)

.69**

127.5

.85

.73

.71

4

240

165.98 (39)

451.83 (16)

.69**

150.5

.90

.76

.84

5

234

175.33 (35)

558.69 (15)

.62**

151.5

.93

.93

.88

Table 43. Criterion Validity of Spring CBMreading on Spring MAP in Reading: WI LEA 1
(Spring Data Collection)
Grade

N

CBMreading
M (SD)

MAP
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

Some Risk (≤ 40 percentile)
th

2

33

76.88 (45)

181.61 (19)

.87**

66

.97

.91

.91

3

26

115.31 (54)

195.65 (17)

.89**

76

.99

.86

.95

4

31

132.55 (41)

208.23 (13)

.76**

123

.89

.86

.85

5

28

154.50 (33)

211.11 (13)

.66**

140

.89

.78

.79

155.76 (38)

215.08 (12)

.74**

149

1.00

1.00

.83

6

25

Some Risk (20 to 40 percentile)
th

th

2

33

76.88 (45)

181.61 (19)

.87**

58

.68

.67

.74

3

26

115.31 (54)

195.65 (17)

.89**

108.5

.85

1.00

.82

4

31

132.55 (41)

208.23 (13)

.76**

123

.72

.75

.74

5

28

154.50 (33)

211.11 (13)

.66**

144

.69

.80

.60

6

25

155.76 (38)

215.08 (12)

.74**

149

.92

1.00

.71

High Risk (≤ 20 th percentile)
2

33

76.88 (45)

181.61 (19)

.87**

32

.99

.86

.96

3

26

115.31 (54)

195.65 (17)

.89**

62

1.00

1.00

.87

4

31

132.55 (41)

208.23 (13)

.76**

94

1.00

1.00

.93

5

28

154.50 (33)

211.11 (13)

.66**

127

.97

1.00

.92

6

25

155.76 (38)

215.08 (12)

.74**

140

.92

1.00

.77

Page | 84
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 44. Criterion Validity of Spring CBMreading with Spring Massachusetts Comprehensive
Assessment (MCA): MA LEA 1 (Spring Data Collection)
Grade

N

CBMreading
M (SD)

MCA
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

Some Risk (“Warning” and “Needs Improvement”)
3

93

138.82 (38)

241.89 (14)

.72**

132.5

.86

.79

.78

4

94

145.85 (38)

238.40 (15)

.71**

143.5

.81

.73

.78

5

72

158.99 (28)

243.63 (13)

.59**

157

.87

.74

.76

Some Risk (“Needs Improvement”)
3

93

138.82 (38)

241.89 (14)

.72**

132.5

.74

.72

.69

4

94

145.85 (38)

238.40 (15)

.71**

144.5

.69

.69

.64

5

72

158.99 (28)

243.63 (13)

.59**

155.5

.86

.73

.76

High Risk (“Warning”)
3

93

138.82 (38)

241.89 (14)

.72**

103.5

.96

1.00

.93

4

94

145.85 (38)

238.40 (15)

.71**

128

.89

.78

.78

5

72

158.99 (28)

243.63 (13)

.59**

137.5

.80

1.00

.79

Predictive Validity of the Slope
Validity of CBMreading passages were examined using the TOSREC, AIMSweb R-CBM, Measures of
Academic Progress (MAP), and DIBELS Next. Table 45 depicts correlations between the slope and the
achievement outcome. Coefficients provided in
Table 46 were derived from progress monitoring data. Students were monitored with grade level
passages for AIMSweb and DIBELS Next. Correlation coefficients in
Table 46 may be underestimated due to differences in error (i.e., Standard Error of the Estimate and
Standard Error of the Slope) between passage sets (see Ardoin & Christ, 2009 and Christ & Ardoin,
2009). The increased precision of CBMreading passages may lead to less variable slopes compared to
more error prone progress monitoring passages. This in turn may deflate the measure of association
between the two measures.
Table 45. Predictive Validity for the Slope of Improvement by CBMreading Passage Level
Passage
Level
Level A
Level B
Level C
Level A
Level B
Level C
Level A
Level B
Level C

Test or Criterion

N

TOSREC Mid-Year

58
98
158
58
98
158
85
130
186

TOSREC End of Year

TOSREC*

Page | 85
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

# CBMreading
Data Points

Weeks of
Monitoring

1-24
1-29
1-29

Coefficient
.43
.45
.36
.58
.22
.14
.46
.56
.16

Section 6. FAST as Evidence-Based Practice
Level A
Level B
Level C
Level A
Level B
Level C
Level A
Level B
Level C
Level A
Level B
Level C

DIBELS Next

57
197
152
33
39
70
49
71
112
33
42
78

AIMSWEB

MAP

MAP

10-30
10-30
10-30
10-30
10-30
10-30
10
10-20
6-20
14-30
14-30
18-30

10-30
10-30
10-30
30
30
30
10
10
10
30
30
30

.89
.82
.60
.98
.84
.78
.21
.23
.21
.03
.41
.17

Note. Reported coefficients are partial Pearson correlation coefficients. Or rxy.z where x is the slope of
improvement, y is the outcome measure and z is the intercept (initial level). TOSREC* coefficients are based on a
partial data set from students in Georgia and Minnesota.

Table 46. Correlation Coefficients between CBMreading Slopes, AIMSweb R-CBM, and
DIBELS Next
AIMSweb Slope
Passage
Level

N

A

Grade 1 (42)
Grade 2 (15)
Grade 3 (4)

B

C

DIBELS Next Slope
Weeks of
Monitoring
(range)

Coefficient
(95% CI)

10–30

.95
(.92 - .97)

N
Grade 1 (42)
Grade 2 (27)
Grade 3 (6)

Total N = 59

Total N = 75

Grade 1 (6)
Grade 2 (41)
Grade 3 (38)
Grade 4 (15)
Grade 5 (7)
Grade 6 (1)

Grade 1 (6)
Grade 2 (113)
Grade 3 (91)
Grade 4 (27)
Grade 5 (14)
Grade 6 (2)

10–30

.85
(.79 - .90)

Total N = 108

Total N = 253

Grade 4 (49)
Grade 5 (44)
Grade 6 (23)

Grade 4 (130)
Grade 5 (112)
Grade 6 (51)

10–30

.64
(.52 - .74)

Total N = 116

Weeks of
Monitoring
(range)

Coefficient
(95% CI)

10–30

.76
(.65 - .85)

10–30

.75
(.69 - .80)

10–30

.50
(.38 - .61)

Total N = 293

Note. CI = Confidence Interval. Samples are disaggregated by grade level.

aReading

Content Validity
The development and design of aReading has a strong basis in reading research and theory. Items
were created and revised by reading teachers and experts. See previous section on item writing
Page | 86
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
processes. Extensive field testing has been conducted to establish appropriate items for aReading. For
instance, the National Reading Panel (NRP; 2000a) submitted a report detailing the current knowledge
base of reading, as well as information on effective approaches to teach children how to read and the
important dimensions of reading to be emphasized. Factor analysis of preliminary data provided
evidence for a large primary factor (i.e., unidimensionality) and several smaller factors, across the
aReading items. Thus, aReading is designed to provide both a unified and a component assessment of
these dimensions, specifically focusing on five main areas (as put forth by the NRP, 2000a): Concepts
of Print, Phonological Awareness, Phonics, Vocabulary, and Comprehension.
The following describes the field testing process in order to derive item parameters to examine both
difficulty and content areas that aReading items are addressing.
Common Core State Standards: Foundational Skills
Common Core Subgroups / Clusters

aReading Domains

Print Concepts

Concepts of Print

Phonological Awareness

Phonemic Awareness

Phonetic Awareness

Phonetic Awareness

Vocabulary

Vocabulary

Common Core State Standards: College and Career Readiness Reading Standards for
Literature / Informational Text
Common Core Subgroups / Clusters

aReading Domains

Key Ideas and Details

Comprehension

Craft and structure

Comprehension & Vocabulary

Integration of Knowledge and Ideas

Comprehension & Vocabulary

aReading
Concepts of Print:
Familiarity with print/books
Understands appropriate directionality and tracking in print
Identifies organizational elements of text
Letter recognition
Word recognition
Sentence recognition
Phonological Skills:
Letter and word recognition / identification
Page | 87
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Common Core State Standards &
MN K–12 Standards
RF K.1, RF 1.1
RI K.5, RF K.1a
RF K.1a
RF K.1a,c,d; RF 1.1a, L 1.2b,c; L 2.2b,c,
etc.; L 5.2a,b,c,d, RI K.6
RF K.1b,d; L K.1a, L 1.1a
RF K.1b; RF K.1d

L 3.2f

Section 6. FAST as Evidence-Based Practice
Syllabication
Letter and word recognition/identification
Onset-rime
Phonemic categorization
Phonemic isolation
Phonemic identification
Phonemic manipulation
Skills Related to Phonics:
Single consonant identification
Single vowel identification
Combined vowel identification
Consonant blends
Consonant digraphs
R-, L-, -gh controlled vowel identification
Vowel digraphs and diphthongs
Phonograms
Recognize and analyze word roots and affixes
Spelling
Skills Related to Vocabulary:
Single-meaning words
Double-meaning words
Compound words and contractions
Base/root word identification and use
Word relationships

Skills Related to Comprehension:
Identify and locate information
Using inferential processes
Comprehension monitoring
Awareness of text/story structure
Awareness of vocabulary use
Evaluative and analytical processes

aReading literary passages include the following text types:


Fiction / Literature



Literary Nonfiction / Narrative Nonfiction



Poetry

Informational text types include:


Exposition

Page | 88
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

RF K.2b; RF 1.3d
L.3.2f
RF k.2a, c
RF K.3d
RF K.2b,c,d; RF 1.2a,c
RF 1.2b,d
RF K.2e
RF K.2d; RF K.3a; RF 1.2c
RF K.2d; RF K.3b; RF 1.2a,c; RF 2.3a,c
RF 1.3c; RF 2.3b
RF 1.2b
RF 1.3a; L K.2c
RF 1.2b
RF 1.3c; RF 2.3b
RF 3.3b; L 4.4b; L 5.4b
RF 2.3c
L K.4a,b
RF 1.3f; RF 2.3d; L K.4a,b
L 2.4d
RF 3.3b; L K 4b; L1.4b,c; L 3.3a; L 3.4c; L
4.4b; L 5.4b
RL 3.4; RL 4.4; RL 5.4; L K.5a,b; L 1.5a,b; L
3.1a,c; L 3.4; L 2.5, 3.5, 4.5, etc.; L 4.4c; L
5.4a,c
RL K.1, 1.1, 2.1, etc.; RI K.1, 1.1, 2.1, etc.
RL K.7; 1.7, 2.7, etc.; RL K.9, RL 2.2, 2.3; RL
3.2, 3.3; RI K.7, 1.7, 2.7, etc.
RL K.4; RI K.4; RI 1.4
RL k.5, 1.5, 2.5, etc.; RI 4.5, RI 5.5
RL 2.4
RL 2.6, 3.6, 4.6, etc.; RI 3.6

Section 6. FAST as Evidence-Based Practice


Argumentation and persuasive text



Procedural text and documents.

For Reading Comprehension items, both literary and informational text types were used.
For Vocabulary, Orthography and Morphology items, response types include:


Selected Response



Conventional Multiple Choice



Alternate Choice



Matching



Multiple True-False

Targeted standards include:


Identify common misspellings



Recognize and analyze affixes and roots to determine word meaning in context;



Use / recognize appropriate word change patterns that reflect change in meaning or
parts of speech (e.g., derivational suffixes)



Identify word meaning in context



Demonstrate ability to use reference materials to pronounce words, clarify word
meaning



Understand figurative language in context



Use word relations to improve understanding



Distinguish similar-different word connotations and/or denotations

Test items were created based upon the following categories from the Common Core State Standards:


Key Ideas and Details



Craft and Structure



Integration of Knowledge and Ideas

The test question types include the following:


Locate / Recall: Identify textually explicit information and make simple inferences
within and across texts



Integrate / Interpret: Make complex inferences and/or integrate information within
and across texts



Critique / Evaluate: Consider text(s) or author critically

Pilot Test
In 2007, there were seven data collections across four districts, six schools, and 78 classrooms with
1,364 total students. Those data were used to estimate item parameters using student samples from
Kindergarten through Third Grade. See Table 47 below.
Page | 89
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 47. School Data Demographics for aReading Pilot Test
Category
White

School A
94%

School B
60%

School C
10%

School D
71%

School E
3%

School F
19%

Black
Hispanic
Asian/Pacific Islander
American
Indian/Alaska Native
Free/Reduced Lunch

<1%
<1%
5%
<1%

11%
<1%
23%
<1%

64%
21%
3%
2%

16%
4%
8%
1%

83%
2%
12%
<1%

51%
1%
28%
<1%

13%

36%

89%

28%

96%

76%

Twelve to thirteen laptop computers were used in each data collection. The test was administered to
students by class. Each student used a mouse to navigate the computer and headphones to hear the
audio. Keyboards were covered so that students could not touch them, and each computer was
surrounded by a foam board to discourage students from looking on other student's computer
screens.
Data collections were proctored by two to three data collectors who entered each student's name and
identifying information. Proctors attended to the children if they had questions, but did not provide
help on any item. Tests consisted of 50–80 items across six domains of reading; 20 items, the linking
items, were on each test and administered to all students in all sites. The linking items make it possible
to import all items into a data super matrix, enabling each item parameter to be based directly and
inferentially on data from all participants across all tests.
A summarization of the mean, standard deviation, minimum and maximum values for the three IRT
parameters is presented in Table 48 for each reading domain. The number of items developed in each
reading domain for the different ability levels are presented in Table 49. From over 500 items
developed, 366 items were identified as being accurately developed and having residual values that
were less than 1.68.
Table 48. Summarization of K–5 aReading Parameter Estimates by Domain
Domain
Phonics
Comprehension
Vocabulary
Concepts of Print
Decoding &
Fluency
Phonemic
Awareness

Parameter (a)
M
SD
Min

Max

198

1.49

.16

1.06

2.01

.19

1.0

-2.13

2.65

.17

.04

.09

.35

24

1.44

.17

1.17

1.94

.77

.37

.14

1.37

.17

.04

.09

.24

34

1.55

.27

1.15

2.50

.83

.99

-2.55

2.95

.16

.05

.09

.30

38

1.50

.19

1.18

2.20

-1.55

.75

-2.67

.64

.16

.02

.13

.22

65

1.57

.23

1.17

2.23

-.37

.68

-1.78

1.77

.18

.06

.11

.40

6

1.28

.15

1.03

1.43

-.41

1.03

-2.22

.45

.21

.04

.14

.28

N

Parameter (b)
M
SD
Min

Max

Parameter (c)
M
SD
Min

Max

Note. M = Mean.

A summary of item difficulty information is provided in Table 49. Analysis on level of difficulty after
item parameterization for the five domains indicated that aReading items are representing domains as
Page | 90
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
would be expected—with Concepts of Print being administered at the low end of ability and
Comprehension at the high end.
Table 49. Item Difficulty Information for K-5 aReading Items
Number of Items at Each Parameter b-Value of Difficulty
Domain

(b≤-3)

(-3 +2.0) were needed to expand the range of linking items. As a result,
project personnel focused on adding a majority of extremely easy items (in the domains of Concepts
of Print and Phonological Awareness) to the grades K–1 test. The 4–5th grade test included high level
Vocabulary items and Comprehension items considered very difficult due to length of passage,
vocabulary, and question type. The tests at this school consisted of an entirely new item set. Three
hundred and eighty total students participated in this data collection at this school; specifically, 100
students were administered the grades K–1 test, 130 students were administered the grades 2–3 test,
and 150 students were administered the grades 4–5 test.
WINTER 2010 DATA COLLECTIONS: FEBRUARY
School C. Again, this school served as the make-up school for previous fall data collections (i.e., School
B, School D, and School E). Three hundred and sixty two total students participated in this data
collection at this school; specifically, 126 students were administered the grades K–1 test, 122 students
were administered the grades 2–3 test, and 114 students were administered the grades 4–5 test.

Page | 94
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

School D. Items were chosen with consideration for difficulty level and content balance. Tests
administered consisted of an entirely new item set. Seven hundred and ten total students
participated in this data collection at this school; specifically, 234 students were administered the
grades K–1 test, 252 students were administered the grades 2–3 test, and 224 students were
administered the grades 4–5 test.

School G. Items were chosen with consideration for difficulty level and content balance. Five hundred
and seventy one total students participated in this data collection at this school; specifically, 215 were
administered the grades K–1 test, 174 students were administered the grades 2–3 test, and 182
students were administered the grades 4–5 test.
SPRING 2010 DATA COLLECTIONS: MARCH
School B. Items were chosen with consideration for difficulty level and content balance. Tests
administered at School B were the same as those administered during the winter data collection at
School F. Six hundred and fifty three total students participated in this data collection at this school;
specifically, 203 students were administered the grades K–1 test, 238 students were administered the
grades 2–3 test, and 212 students were administered the grades 4–5 test.

School E. School E served as the make-up school for all data collections in need of more
administrations for items. Seven hundred and sixty eight total students participated in this data
collection at this school; specifically, 278 students were administered the grades K–1 test, 258 students
were administered the grades 2–3 test, and 232 students were administered the grades 4–5 test.

School F. Items were chosen with consideration for difficulty level and content balance. Tests
administered consisted of an entirely new item set. Five hundred and seventy six total students
participated in this data collection at this school; specifically, 216 students were administered the
grades K–1 test, 180 students were administered the grades 2–3 test, and 180 students were
administered the grades 4–5 test.
SPRING 2012 DATA COLLECTION
Analysis of the initial field parameterization indicated that the item bank provided the most
information for students with ability levels in the middle range on the theta scale (mostly between -2.0
and 2.0). However, the bank lacked items for students’ ability levels at the extreme ends of the theta
scale. Because the test bank needs more information for students with extremely low ability levels
(less than -2.0 on the theta scale) and those with extremely high ability levels (more than 2.0), the
aReading team developed additional items that targeted Kindergarten through early First Grade level
students and also late fifth through early Sixth Grade students. The new easy items focused on the
Concept of Print domain. The new items created for the higher grade students were all
Comprehension items that contained reading passages that reflect fifth through Sixth Grade reading
content.
The participants were students from two public schools in South St. Paul, Minnesota. The schools
provided a richly diverse group of students. The aReading team administered the Concept of Print
Page | 95
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
items to 411 Kindergarten and First Grade students (N = 19 classrooms) and the Comprehension items
to 391 fifth and Sixth Grade students (N = 18 classrooms).
The team collected data for the Concept of Print items in November and December 2011 and the
Comprehension items in March 2012. A range of two to three group administrations were completed
in a day (during school hours and with transitions and scheduling conflicts). Research assistants and
other project personnel supervised the test administrations, ensuring student safety and the integrity
of collecting the data. Data collectors provided oral directions to the students. Students also received
automated directions and on-screen demonstrations before beginning the test. The students
completed the items on a web-based browser using laptops provided by the team. Each
administration lasted approximately 15 to 30 minutes for each student and all responses were
automatically saved in a secure online database.
FALL 2012 AND SPRING 2013 DATA COLLECTION
Over 9,000 students from six middle and junior high schools and four high schools in the upper
Midwest participated to test items generated for aReading grades 6–12 (N=70 unique classrooms).
Trained project personnel administered the assessments. Each test included linking items and 20
unique aReading items targeted to grade level spans from 6–8, 9–12, and 11–12. Each item was given
to approximately 300 students. Administration lasted 20 to 45 minutes per student. Criterion
measures were collected from each school and used for additional analyses.

Parameterization
All items were parameterized within a 3-PL IRT framework via the computer program Xcalibre (Guyer &
Thompson, 2012). Xcalibre allows for the use of a sparse data-matrix used with linking items
(described above) deriving item parameters for every item, even if all students did not complete each
item. Table 52 below shows descriptive statistics for item parameters for aReading. Item parameters
are used to calculate the level of information for each item at a given ability estimate (discussed in the
previous section). Based on a student’s current ability estimate during a CAT, the aReading algorithm
selects the item that is likely to provide the most information for that student.
Table 52. Descriptive Statistics of K–12 aReading Item Parameters
Domain

N

M

SD

Min

Max

M

SD

Min

Max

M

SD

Min

Max

Overall

1101

1.34

.40

.46

3.78

.11

1.02

-2.87

2.77

.24

.04

.06

.36

COP

98

1.5

.46

.63

2.82

-1.14

.96

-2.64

1.45

.21

.02

.11

.25

Comprehension

521

1.39

.4

.46

3.78

.39

.71

-1.28

2.77

.24

.04

.06

.36

Vocabulary

229

1.34

.37

.61

3.03

.33

1.13

-2.87

2.62

.22

.03

.06

.29

Phonics

128

1.38

.44

.52

2.73

-.34

1.09

-2.58

1.95

.21

.02

.13

.33

PA

66

1.11

.32

.61

1.9

-.61

.87

-2.56

1.24

.21

.01

.16

.23

OMF

57

1.31

.31

.76

2.27

.58

.72

-.76

2.63

.24

.06

.06

.29

Note. COP = Concepts of Print; PA = Phonological Awareness; OMF = orthography, morphology, and figurative
language.

Page | 96
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Criterion-Related Validity
Criterion-related validity of aReading tests was examined using the Gates MacGinitie Reading Tests-4th
Edition (GMRT-4th; MacGinitie, MacGinitie, Maria, & Dreyer, 2000). The GMRT-4th is a norm-referenced,
group administered measure of reading achievement distributed by Riverside Publishing Company. It
is designed to provide guidance in planning instruction and intervention and is typically used as a
diagnostic tool for general reading achievement. The GMRT-4th was normed with students in the prereading stages through high school levels. The GMRT-4th was selected because of its strong criterion
validity. Correlations between the GMRT composite score and comprehension and vocabulary
subtests of the Iowa Test of Basic Skills and GMRT composite scores across grades is high (.76 and .78
respectively; Morsy, Kieffer, & Snow, 2010). A similar pattern of results were observed between the
GMRT and subscales of the California Tests of Basic Skills (.84 and .81 respectively; Morsy et al., 2010).
GMRT scores also correlate highly with Comprehensive Tests of Basic Skills vocabulary,
comprehension, and composite scores (.72, .79, and. 83 respectively; Morsy et al., 2010). Further, the
correlation between GMRT composite scores and reading scores on the Basic Academic Skills Samples
were strong as well (.79; Jenkins & Jewell, 1992).
The measure of interest with the GMRT-4th is the extended scale scores (ESS). The ESS puts the results
from the test on a single continuous scale to allow comparison across time and grades. All materials
were provided to students, including the test booklet and answer booklet.
Five trained aReading project team data collectors administered the GMRT-4th during February of 2011
at two separate schools. Participants included students in first through Fifth Grades. Three classrooms
per grade at School A participated (n = 622); all students in first through Fifth Grades at School B
participated (n = 760). The majority of students at Schools A and B were white (69%). Students were
administered the word decoding/vocabulary and comprehension subtests of the GMRT-4th during two
separate testing sessions. Some students were administered the word decoding/vocabulary section
first while other students were administered the comprehension subtest first. See the Table 53 below
for demographic information, disaggregated by school.
Table 53. Demographics for Criterion-Related Validity Sample for GMRT-4 th and aReading
Category

School A

School B

White

70%

69%

Black

6%

5%

Hispanic

9%

6%

Asian/Pacific Islander

13%

19%

American Indian/Alaskan Native
Free/Reduced Lunch

3%
19%

1%
14%

LEP

14%

14%

Special Education

11%

10%

Note. LEP = Limited English Proficiency. Percentages are rounded to whole numbers and therefore may not add
to precisely 100.

Page | 97
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Descriptive information for the GMRT-4th and correlation coefficients between each scale and
aReading scores are provided in Table 63.
Table 54. Sample-Related Information for aReading Criterion-Related Validity Data
Decoding

Vocabulary

Comprehension

Composite

Grade

N

Mean

SD

N

Mean

SD

N

Mean

SD

N

Mean

SD

1

348

436

47

-

-

-

130

407

44

125

409

43

2

163

449

43

-

-

-

215

459

49

-

-

-

3

-

-

-

170

485

41

168

484

42

165

484

40

4

-

-

-

182

504

39

180

503

42

175

502

36

5

-

-

-

182

513

31

187

518

35

181

514

30

1–5

511

442.5

45

534

501

39

881

477

56

646

483

53

Note. M = Mean; SD = Standard Deviation

Table 55. Correlation Coefficients between GMRT-4 th and aReading Scaled Score
Grade

Decoding

Vocabulary

Comprehension

Composite

1
2
3
4
5

.82 (131)
.68 (163)
-

.79 (170)
.76 (182)
.65 (182)

.73 (130)
.75 (215)
.81 (168)
.72 (180)
.58 (187)

.83 (125)
.84 (165)
.78 (175)
.64 (181)

1–5

.75 (348)

.74 (534)

.82 (881)

.86 (646)

Note. Sample size is denoted by ().

Overall, there appears to be a strong positive correlation between composite scores from the GMRT4th and aReading scaled scores. There is some variability between grades, with coefficient values
between .64 and .83. Subtests showed greater variability. Specifically, comprehension correlation
coefficients ranged from .58 to .81.
Content, construct, and predictive validity of aReading is summarized in.
Table 56.

Page | 98
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 56. Content, Construct, and Predictive Validity of aReading
Type of
Validity

Grade

Test or Criterion

Content

K–5

Reading teachers/experts

Content

K–3

Items administered by Theta Level

Predictive

Construct

Construct

1–5
1
2*
3
4
5
1–5
1
2
3
4
5
1-5
1
2
3
4
5

N (range)

Gates-MacGinitie

Curriculum-Based Measurement of
Oral Reading Fluency

MAP

Coefficient
(if applicable)
Range

Median

287
125–215
125
215
165
175
181
55–171
55
171
108
114
103
55–398
55
302
391
398
376

0.64–0.84
0.83
0.75
0.84
0.78
0.64
0.56–0.83
0.83
0.81
0.74
0.80
0.56
0.69–0.83
0.69
0.83
0.83
0.77
0.73

0.78

0.80

0.77

Note. Grade 2 Predictive validity for the GMRT-4th is based on the Comprehension subtest, whereas all other
grades are based on the overall composite score.

More recently, data collections have produced aReading criterion-related evidence with various other
criterion measures.
Table 57. Criterion Validity of Spring aReading with Spring Minnesota Comprehensive
Assessment III (MCA-III) in Reading: MN LEA 1 (Spring Data Collection)
Grade

N

aReading
MCA
M (SD)
M (SD)
Some Risk (Meets Standards)
6
202 522.01 (15.45)
651.36 (15.07)
7
126 522.81 (14.95)
742.01 (15.26)
8
94
524.95 (14.11)
843.22 (11.88)
High Risk (Does Not Meet Standards)
6
202 522.01 (15.45)
651.36 (15.07)
7
126 522.81 (14.95)
742.01 (15.26)
8
94
524.95 (14.11)
843.22 (11.88)

Page | 99
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

r(x,y)

Cut

AUC

Sens.

Spec.

.72**
.66**
.58**

521.5
528.5
534.5

.86
.83
.85

.78
.77
.81

.77
.76
.79

.72**
.66**
.58**

515.5
523
524.5

.83
.82
.84

.74
.75
.78

.74
.77
.78

Section 6. FAST as Evidence-Based Practice
Table 58. Criterion Validity for Spring aReading with Spring MCA-III in Reading: MN LEA 4
(Spring Data Collection)
Grade

aReading
M (SD)

N

MCA-III
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

Some Risk (Does Not Meet or Partially Meets Standards)
3

629

534 (27)

347 (21)

.82

532.5

.90

.83

.82

4

615

549 (28)

447 (16)

.81

550.5

.89

.82

.82

5

516

564 (32)

553 (16)

.84

556.5

.93

.84

.84

High Risk (Does Not Meet Standards)
3

629

534 (27)

347 (21)

.82

524.5

.90

.82

.81

4

615

549 (28)

447 (16)

.81

536.5

.92

.84

.84

5

516

564 (32)

553 (16)

.84

544.5

.96

.89

.86

Table 59. Criterion Validity for Spring aReading with Spring MCA-III in Reading: MN LEA 3
(Spring Data Collection)
Grade

N

aReading
M (SD)

MCA-III
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

Some Risk (Does Not Meet or Partially Meets Standards)
3

156

504.97 (18.09)

351.56 (19.77)

.82

502.5

.89

.80

.79

4

63

508.17 (20.75)

447.33 (16.11)

.78

509.5

.86

.78

.74

5

148

528.02 (19.58)

557.91 (14.22)

.82

523.5

.91

.84

.83

6

152

529.18 (20.93)

655.09 (18.11)

.76

530.5

.86

.80

.78

Some Risk (Partially Meets Standards)
3

156

504.97 (18.09)

351.56 (19.77)

.82

--

.59

--

--

4

63

508.17 (20.75)

447.33 (16.11)

.78

--

.46

--

--

5

148

528.02 (19.58)

557.91 (14.22)

.82

522.5

.80

.75

.75

6

152

529.18 (20.93)

655.09 (18.11)

.76

530.5

.64

.68

.63

High Risk (Does Not Meet Standards)
3

156

504.97 (18.09)

351.56 (19.77)

.82

498.5

.92

.83

.82

4

63

508.17 (20.75)

447.33 (16.11)

.78

503.00

.95

.90

.85

5

148

528.02 (19.58)

557.91 (14.22)

.82

510.5

.96

.85

.87

6

152

529.18 (20.93)

655.09 (18.11)

.76

518.5

.95

.84

.87

Criterion-related validity evidence for aReading is not limited to the Midwest. See tables below.
Table 60. Criterion Validity of Spring aReading with Spring CRCT in Reading: GA LEA 1
(Spring to Spring Prediction)
Grade

N

aReading
M (SD)
Some Risk (Meets Standards)
3
327
501.56 (20)
4
314
513.53 (20)

CRCT
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

848.62 (28)
848.30 (27)

.75*
.77*

501.50
514.50

.84
.86

.78
.79

.78
.79

Page | 100
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
5
347
519.34 (19)
841.27 (25)
High Risk (Does Not Meet Standards)
3
327
501.56 (20)
848.62 (28)
4
314
513.53 (20)
848.30 (27)
5
347
519.34 (19)
841.27 (25)

.74*

525.50

.83

.79

.78

.75*
.77*
.74*

478.50
493.00
502.50

.95
.93
.90

.91
.83
.85

.85
.84
.84

Table 61.Criterion Validity of Spring aReading with Spring Massachusetts Comprehensive
Assessment (MCA): MA LEA 1 (Spring Data Collection)
Grade

N

aReading
MCA
M (SD)
M (SD)
Some Risk (“Warning” and “Needs Improvement”)
3
93
508.96 (22.16)
241.89 (14.08)
4
93
512.17 (19.09)
238.40 (19.09)
5
74
524.04 (15.41)
243.63 (13.20)
Some Risk (“Needs Improvement”)
3
93
508.96 (22.16)
241.89 (14.08)
4
93
512.17 (19.09)
238.40 (19.09)
5
74
524.04 (15.41)
243.63 (13.20)
High Risk (“Warning”)
3
93
508.96 (22.16)
241.89 (14.08)
4
93
512.17 (19.09)
238.40 (19.09)
5
74
524.04 (15.41)
243.63 (13.20)

r(x,y)

Cut

AUC

Sens.

Spec.

.81**
.78**
.64**

503.5
513.5
523.5

.89
.88
.85

.76
.75
.76

.77
.75
.76

.81**
.78**
.64**

504.5
515
522.5

.77
.73
.83

.76
.71
.75

.78
.63
.76

.81**
.78**
.64**

483.00
496.5
511.5

.97
.95
.84

.88
.89
1.00

.92
.85
.78

Chapter 2.7: Diagnostic Accuracy
earlyReading
earlyReading diagnostic accuracy information was derived from the sample described in Table 31.
earlyReading diagnostic accuracy information is provided for both Kindergarten and First Grade, using
the Group Reading Assessment Diagnostic Evaluation (GRADE™) as a criterion measure. Measures of
diagnostic accuracy were used to determine decision thresholds using criteria related to sensitivity,
specificity, and area under the curve (AUC). Specifically, specificity and sensitivity were computed at
different cut scores in relation to maximum AUC values. Decisions for final benchmark percentiles
were generated based on maximizing each criterion at each cut score (i.e., when the cut score
maximized specificity ≥ .70, and sensitivity was also ≥ .70; see Silberglitt & Hintze, 2005). In the
scenario for which a value of .70 could not be achieved for either specificity or sensitivity, precedence
was given to maximizing specificity.

Page | 101
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 62. Kindergarten Diagnostic Accuracy for earlyReading Measures

7
7

15th Percentile
Sens Spec.
.
.80
.82
.76
.71

Onset Sounds
F to F1
.94
F to S
.88
W to S
.76

8
7
14

.80
.82
.69

.81
.89
.86

.81
.89
.85

.79
.74
.73

12
12
15

.74
.80
.62

.69
.77
.78

.72
.78
.74

Letter Names
F to F1
.73
F to S
.81
W to S
.79
S to S
.78

22
20
35
48

.60
.76
.72
.71

.59
.74
.73
.70

.59
.74
.73
.70

.65
.76
.77
.76

25
25
40
51

.60
.69
.73
.71

.64
.69
.73
.72

.62
.69
.73
.72

Letter Sounds
F to F1
.95
F to S
.81
W to S
.85
S to S
.85

1
6
22
34

.86
.76
.78
.82

.14
.78
.75
.80

.87
.78
.75
.80

.68
.75
.82
.71

7
10
28
42

.63
.67
.75
.73

.60
.71
.73
.59

.61
.70
.74
.63

Rhyming
F to F1
F to S
W to S
S to S

5
6
7
14

.80
.76
.88
.88

.80
.75
.89
.75

.80
.75
.89
.76

.77
.81
.83
.76

9
8
12
15

.72
.79
.75
.71

.69
.72
.76
.69

.71
.74
.76
.70

1
4
9

.88
.82
.88

.53
.80
.72

.56
.80
.73

.69
.74
.75

1
8
9

.77
.73
.64

.60
.55
.79

.65
.60
.75

.82
.94

.81
.87

.81
.87

.78
.76

15
32

.75
.72

.76
.57

.76
.61

25

.78

.72

.72

.74

39

.71

.65

.66

3

.78

.87

.86

.77

9

.72

.68

.69

3
6

.83
.83

.79
.81

.79
.81

.76
.78

5
8

.70
.70

.72
.79

.72
.76

28
33
40
52

.80
.88
.94
.75

.91
.84
.72
.74

.91
.84
.77
.74

.79
.84
.85
.81

42
40
47
60

.74
.80
.84
.75

.67
.77
.72
.74

.71
.78
.75
.74

AUC
Concepts of Print
F to F1
.86
F to S
.82

.89
.80
.92
.86

Word Blending
F to S
.70
W to S
.82
S to S
.85

Cutpoint

Word Segmenting
W to S
.85
4
S to S
.90
28
Sight Words 50
S to S
.82
Decodable Words
S to S
.90
Nonsense Words
W to S
.86
S to S
.90
Composite
F to F1
.96
F to S
.91
W to S
.91
S to S
.95

Classification

AUC

Cutpoint

.82
.71

.80
.74

9
8

40th Percentile
Sens Spec.
.
.74
.69
.75
.65

Classification
.72
.68

Note. F = Fall; W = Winter; S = Spring; 1Base rates below the 15th percentile were low and above the 40th
percentile were high. Note. Fall to Fall was concurrent and used the GRADE Level P as the criterion. All others
used the GRADE Level K.

Based on these analyses, the values at the 40th and 15th percentiles were identified as the primary
and secondary benchmarks for earlyReading, respectively. These values thus correspond with a
Page | 102
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
prediction of performance at the 40th and 15th percentiles on the GRADE, a nationally normed
reading assessment of early reading skills. Performance above the primary benchmark indicates the
student is at low risk for long-term reading difficulties. Performance between the primary and
secondary benchmarks indicates the student is at some risk for long-term reading difficulties.
Performance below the secondary benchmark indicates the student is at high risk for long-term
reading difficulties. These risk levels help teachers accurately monitor student progress using the
FAST™ earlyReading measures. See

Table 62 below for Diagnostic Accuracy results using the GRADE as the criterion measure.
Table 63. First Grade Diagnostic Accuracy for earlyReading Measures
15th Percentile
AUC

40th Percentile

Cutpoint

Sens.

Spec.

Classification

AUC

Cutpoint

Sens.

Spec.

Classification

Word Blending
F to S

.87

6

.86

.80

.80

.82

7

.72

.72

.72

W to S

.82

8

.60

.80

.79

.78

8

.56

.84

.80

S1 to S2

.68

9

.55

.67

.66

.65

9

.50

.69

.66

Word Segmenting
F to S

.78

25

.71

.72

.72

.75

27

.72

.68

.68

W to S

.82

29

.70

.74

.74

.79

30

.67

.67

.67

S1 to S2

.71

30

.64

.71

.70

.67

30

.53

.74

.70

Sight Words 150
F to S

.97

5

.86

.92

.92

.91

14

.84

.81

.82

W to S

.97

21

.90

.95

.94

.95

44

.85

.89

.88

S1 to S2

.97

48

.91

.96

.96

.95

61

.87

.92

.91

Decodable Words
F to S

.90

2

.86

.85

.85

.88

5

.76

.74

.75

W to S

.93

10

.80

.91

.91

.95

15

.85

.86

.86

S1 to S2

.93

22

.82

.82

.82

.96

24

.84

.89

.88

Nonsense Words
F to S

.93

2

.86

.88

.88

.84

5

.76

.79

.79

W to S

.88

12

.80

.75

.75

.92

14

.93

.78

.81

S1 to S2

.87

14

.89

.80

.81

.87

17

.81

.77

.78

Sentence Reading/CBMR1
F to S

.97

10

.86

.93

.93

.93

18

.84

.84

.84

W to S

.98

192

1.0

.95

.95

.98

372

.96

.91

.92

S1 to S2

.98

362

.82

.98

.97

.98

652

.94

.93

.93

F to S

.98

25

1.0

.93

.93

.93

28

.76

.84

.83

W to S

.98

34

1.0

.82

.83

.97

37

1.0

.77

.81

Composite

Page | 103
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
.99

S1 to S2

45

.89

.90

.90

.97

51

.92

.92

.92

Note. F = Fall; W = Winter; S = Spring; Sentence Reading was administered in the fall, CBMR was administered in
the winter and spring. 2Scores shown are equated.
1

More recently, diagnostic accuracy analyses have also been conducted using earlyReading subtest
scores and composite scores to predict aReading. These findings are summarized in the tables below.
Students were recruited from several school districts in Minnesota. Cut score was selected by
optimizing sensitivity at about .70 and balancing sensitivity with specificity (Silberglitt & Hintze, 2005).
In the tables that follow, dashes indicate unacceptable sensitivity and specificity due to low AUC.
Table 64. Diagnostic Accuracy of Fall earlyReading Concepts of Print Subtest with Winter
aReading: MN LEA 3 (Fall to Winter Prediction)
Grade

N

Concepts of Print
M (SD)
Some Risk (< 40th percentile)
KG
5173
8.31 (2.63)
Some Risk (20th to 40th percentile)
KG
5173
8.31 (2.63)
High Risk (< 20th percentile)
KG
5173
8.31 (2.63)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

430.28 (27.43)

.58**

6.5

.80

.77

.71

430.28 (27.43)

.58**

6.5

.81

.74

.75

430.28 (27.43)

.58**

6.5

.81

.79

.70

Table 65. Diagnostic Accuracy of Fall earlyReading Onset Sounds Subtest with Winter aReading:
MN LEA 3 (Fall to Winter Prediction)
Grade

N

Onset Sounds
M (SD)
th
Some Risk (< 40 percentile)
KG
13203
11.59 (4.28)
Some Risk (20 th to 40 th percentile)
KG
13203
11.59 (4.28)
High Risk (< 20 th percentile)
KG
13203
11.59 (4.28)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

430.28 (27.43)

63**

8.50

.84

.79

.77

430.28 (27.43)

.63**

8.50

.84

.79

.77

430.28 (27.43)

.63**

7.5

.83

.79

.77

Table 66. Diagnostic Accuracy of Fall earlyReading Letter Names Subtest with Winter aReading:
MN LEA 3 (Fall to Winter Prediction)
Grade

N

Letter Names
M (SD)
Some Risk (< 40 th percentile)
KG
5173
27.29 (22.38)
th
th
Some Risk (20 to 40 percentile)
KG
5173
27.29 (22.38)
th
High Risk (< 20 percentile)
KG
5173
27.29 (22.38)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

430.28 (27.43)

.69**

11.50

.78

.71

.70

430.28 (27.43)

.69**

14.5

.79

.74

.73

430.28 (27.43)

.69**

9.5

.82

.73

.73

Page | 104
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Table 67. Diagnostic Accuracy of Fall earlyReading Letter Sounds Subtest with Winter aReading:
MN LEA 3 (Fall to Winter Prediction)
Grade

N

Letter Sounds
M (SD)
Some Risk (< 40 th percentile)
KG
5173
11.57 (11.76)
1
842
34.83 (14.89)
Some Risk (20 th to 40 th percentile)
KG
5173
11.57 (11.76)
High Risk (< 20 th percentile)
KG
5173
11.57 (11.76)
1
842
34.83 (14.89)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

430.28 (27.43)
466.04 (27.16)

.62**
.65

2.5
30.5

.80
.73

.71
.66

.73
.65

430.28 (27.43)

.62**

3.5

.77

.72

.67

430.28 (27.43)
466.04 (27.16)

.62**
.65

1.5
17.5

.82
.99

.79
1.00

.78
.89

Table 68. Diagnostic Accuracy of Fall earlyReading Letter Sounds Subtest with Spring aReading:
MN LEA 3 (Fall to Spring Prediction)
Grade

N

Letter Sounds
M (SD)
th
Some Risk (< 40 percentile)
1
842
34.83 (14.89)
High Risk (< 20 th percentile)
1
842
34.83 (14.89)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

457.39 (25.21)

.31**

29.50

.75

.70

.69

457.39 (25.21)

.31**

27.50

.78

.74

.71

Table 69. Diagnostic Accuracy of Winter earlyReading Letter Sounds Subtest with Spring
aReading: MN LEA 3 (Winter to Spring Prediction)
Grade

N

Letter Sounds
M (SD)
th
Some Risk (< 40 percentile)
1
208
43.62 (18.25)
High Risk (< 20 th percentile)
1
208
43.62 (18.25)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

457.39 (25.21)

.58**

40.50

.77

.70

.72

457.39 (25.21)

.58**

40.50

.76

.73

.70

Table 70. Diagnostic Accuracy of Winter earlyReading Rhyming Subtest with Spring aReading:
MN LEA 3 (Winter to Spring Prediction)
Grade

N

Rhyming
M (SD)
Some Risk (< 40 th percentile)
1
106
13.28 (3.34)
High Risk (< 20 th percentile)
1
106
13.28 (3.34)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

457.39 (25.21)

.53**

14.50

.75

.70

.70

457.39 (25.21)

.53**

14.50

.71

.70

.64

Page | 105
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

Table 71. Diagnostic Accuracy of Fall earlyReading Word Segmenting Subtest with Winter
aReading: MN LEA 3 (Fall to Winter Prediction)
Grade

N

Word Segmenting
M (SD)
Some Risk (< 40 th percentile)
1
th
High Risk (< 20 percentile)
1
9843
26.20 (7.41)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

-

-

-

-

-

-

466.04 (27.16)

.51**

22.50

.87

.78

.78

Table 72. Diagnostic Accuracy of Fall earlyReading Nonsense Words Subtest with Winter
aReading: MN LEA 3 (Fall to Winter Prediction)
Grade

N

Nonsense Words
M (SD)
th
Some Risk (< 40 percentile)
1
7997
13.14 (22.13)
th
High Risk (< 20 percentile)
1
7997
13.14 (22.13)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

466.04 (27.16)

.62**

9.50

.66

.60

.61

466.04 (27.16)

.62**

5.50

.89

.83

.78

Table 73. Diagnostic Accuracy of Fall earlyReading Sight Words Subtest with Winter aReading:
MN LEA 3 (Fall to Winter Prediction)
Grade

N

Sight Words
M (SD)
th
Some Risk (< 40 percentile)
1
9685
32.76 (33.40)
th
High Risk (< 20 percentile)
1
9685
32.76 (33.40)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

466.04 (27.16)

.67**

26.50

.69

.71

.58

466.04 (27.16)

.67**

4.5

.92

.85

.87

Table 74. Diagnostic Accuracy of Fall earlyReading Sentence Reading Subtest with Winter
aReading: MN LEA 3 (Fall to Winter Prediction)
Grade

N

Sentence Reading
M (SD)
th
Some Risk (< 40 percentile)
1
9618
34.23 (37.43)
th
High Risk (< 20 percentile)
1
9618
34.23 (37.43)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

466.04 (27.16)

.70**

19.5

.71

.70

.60

466.04 (27.16)

.70**

6.5

.94

.90

.87

Page | 106
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 75. Diagnostic Accuracy of Fall earlyReading Sentence Reading Subtest with Spring
aReading: MN LEA 3 (Fall to Spring Prediction)
Grade

N

Sentence Reading
M (SD)
th
Some Risk (< 40 percentile)
1
th
High Risk (< 20 percentile)
1
9618
34.23 (37.43)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

-

-

-

-

-

-

457.39 (25.21)

.66**

15.50

.72

.68

.66

Table 76. Diagnostic Accuracy of Winter earlyReading Sentence Reading Subtest with Spring
aReading: MN LEA 3 (Winter to Spring Prediction)
Grade

N

Sentence Reading
M (SD)
th
Some Risk (< 40 percentile)
1
433
57.36 (36.69)
High Risk (< 20 th percentile)
1
433
57.36 (36.69)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

457.39 (25.21)

.78**

32.50

.82

.76

.76

457.39 (25.21)

.78**

22.50

.92

.85

.85

Table 77. Diagnostic Accuracy of Winter earlyReading Composite with Winter aReading:
MN LEA 3 (Fall to Winter Prediction)
Grade

N

Composite
M (SD)
Some Risk (< 40 th percentile)
KG
577
34.45 (12.86)
1
589
30.44 (9.99)
th
Some Risk (20 to 40 th percentile)
KG
577
34.45 (12.86)
1
590
30.44 (9.99)
th
High Risk (< 20 percentile)
KG
577
34.45 (12.86)
1
590
30.44 (9.99)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

430.95 (27.353)
466.74 (27.05)

.70**
.73**

27.5
24.5

.84
.85

.77
.85

.78
.73

430.95 (27.353)
466.74 (27.05)

.70**
.73**

28.5
24.5

.79
.73

.72
.75

.70
.66

430.95 (27.353)
466.74 (27.05)

.70**
.73**

25.5
23.5

.82
.89

.79
.86

.76
.77

Table 78. Diagnostic Accuracy of Fall earlyReading Composite with Spring aReading: MN LEA 3
(Fall to Spring Prediction)
Grade

N

Composite
M (SD)
Some Risk (< 40 th percentile)
KG
577
34.45 (12.86)
th
High Risk (< 20 percentile)
KG
577
34.45 (12.86)

aReading
M (SD)

r(x,y)

Cut

415.34 (27.16)

.64**

415.34 (27.16)

.64**

Page | 107
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

AUC

Sens.

Spec.

37.50

.73

.68

.70

34.50

.76

.71

.70

Section 6. FAST as Evidence-Based Practice
Table 79. Diagnostic Accuracy of Winter earlyReading Composite with Spring aReading:
MN LEA 3 (Winter to Spring Prediction)
Grade

N

Composite
M (SD)
th
Some Risk (< 40 percentile)
KG
479
42.92 (15.6)
th
High Risk (< 20 percentile)
KG
479
42.92 (15.6)

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

415.34 (27.16)

.74**

49.50

.84

.76

.76

415.34 (27.16)

.74**

43.50

.84

.73

.77

Table 80. Diagnostic Accuracy of Fall earlyReading Composite (2014–15 Weights) with Spring
aReading: MN LEA 3 (Fall to Spring Prediction)
Composite
M (SD)
th
Some Risk (<40 percentile)
KG
514
35.03 (6.28)
1
170
40.23 (10.78)
High Risk (<20 th percentile)
KG
514
35.03 (6.28)
1
170
40.23 (10.78)
Grade

N

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

415.64 (27.24)
459.89 (26.03)

.69**
.81**

35.5
34.5

.75
.65

.69
.63

.70
.60

415.64 (27.24)
459.89 (26.03)

.69**
.81**

34.5
34.5

.78
.70

.74
.69

.70
.60

Table 81. Diagnostic Accuracy of Winter earlyReading Composite (2014-15 Weights) with Spring
aReading: MN LEA 3 (Winter to Spring Prediction)
Composite
M (SD)
th
Some Risk (<40 percentile)
KG
522
48.02 (9.91)
1
171
49.77 (13.87)
th
High Risk (<20 percentile)
KG
522
48.02 (9.91)
1
171
49.77 (13.87)
Grade

N

aReading
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

415.34 (27.18)
459.25 (27.28)

.75
.77

51.5
--

.85
.65

.79
--

.75
--

415.34 (27.18)
459.25 (27.28)

.75
.77

48.5
43.5

.86
.71

.79
.71

.77
.62

CBMreading
CBMreading diagnostic accuracy information is provided for first through Sixth Grades, using the
TOSREC, and MAP as the criterion measures. Measures of diagnostic accuracy were used to determine
decision thresholds using criteria related to sensitivity, specificity, and area under the curve (AUC).
Specifically, specificity and sensitivity were computed at different cut scores in relation to maximum
AUC values. Decisions for final benchmark percentiles were generated based on maximizing each
criterion at each cut score (i.e., when the cut score maximized specificity ≥ .70, and sensitivity was also
≥ .70; see Silberglitt & Hintze, 2005). In the scenario for which a value of .70 could not be achieved for
either specificity or sensitivity, precedence was given to maximizing specificity.

Page | 108
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
CBMreading diagnostic accuracy was determined based on a sample of 1,153 students in the state of
Minnesota, spanning across three regions. Data was collected during the 2012–13 school year. The
sample consisted of approximately 45% males and 55% females. Approximately 20% of the students
involved were eligible for free and reduced lunch. The majority of students were White (52%). The
remainder of the sample consisted of approximately 30% Hispanic, 12% Black, 4% Asian or Pacific
Islander, and 1% American Indian or Alaska Native. Approximately 15% of students were receiving
special education services. All participants were proficient in English. See Table 82 on the following
page for diagnostic accuracy results.
Table 82. Diagnostic Accuracy by Grade Level for CBMreading Passages
Grade Level
20th Percentile
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
30th Percentile
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1

N

Cut a

AUC

Sens.

Spec.

Classif.

Lag Time

Criterion Measure

171
206
188
181
202
205
171
206
188
181
202
205
171
206
188
181
202
205
171
206
188
181
202
205

16.5
42.5
75.5
108.5
107.5
118.5
17
57
88
113
101
126
21
63
67
104
97
126
16
82
88
114
108
126

0.81
0.93
0.89
0.87
0.90
0.90
0.77
0.82
0.8
0.88
0.89
0.89
0.79
0.82
0.77
0.89
0.89
0.89
0.8
0.9
0.89
0.87
0.89
0.85

0.75
0.88
0.84
0.78
0.84
0.92
0.63
0.63
0.77
0.82
0.7
0.83
0.78
0.83
0.51
0.8
0.74
0.85
0.66
0.87
0.82
0.83
0.8
0.77

0.63
0.87
0.83
0.82
0.79
0.72
0.82
0.85
0.75
0.81
0.91
0.82
0.7
0.68
0.88
0.86
0.92
0.82
0.81
0.84
0.81
0.76
0.85
0.79

0.71
0.88
0.84
0.79
0.83
0.88
0.74
0.76
0.76
0.81
0.86
0.82
0.73
0.72
0.75
0.84
0.88
0.83
0.76
0.86
0.82
0.78
0.84
0.79

2 to 4
2 to 4
2 to 4
2 to 4
2 to 4
2 to 4
4 months
4 mo.
4 mo.
4 mo.
4 mo.
4 mo.
8 mo.
8 mo.
8 mo.
8 mo.
8 mo.
8 mo.
~1 year
~1 year
~1 year
~1 year
~1 year
~1 year

TOSREC

171
206
188
181
202
205
171
206
188
181
202
205
171
206
188
181
202
205
171

16.5
44.5
79.5
117.5
115.5
135.5
31
82
85
128
125
144
24
82
98
125
128
144
22

0.81
0.94
0.88
0.83
0.87
0.88
0.78
0.83
0.77
0.82
0.82
0.82
0.78
0.78
0.8
0.84
0.86
0.79
0.83

0.75
0.91
0.83
0.73
0.79
0.76
0.84
0.81
0.57
0.84
0.65
0.76
0.74
0.77
0.82
0.66
0.8
0.74
0.76

0.63
0.83
0.83
0.79
0.77
0.82
0.57
0.73
0.86
0.59
0.75
0.75
0.7
0.54
0.65
0.85
0.73
0.7
0.77

0.71
0.89
0.83
0.75
0.79
0.78
0.74
0.78
0.66
0.74
0.69
0.75
0.73
0.67
0.76
0.75
0.76
0.71
0.76

2 to 4
2 to 4
2 to 4
2 to 4
2 to 4
2 to 4
4 mo.
4 mo.
4 mo.
4 mo.
4 mo.
4 mo.
8 mo.
8 mo.
8 mo.
8 mo.
8 mo.
8 mo.
~1 year

TOSREC

Page | 109
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

MAP

MAP

Section 6. FAST as Evidence-Based Practice
2
3
4
5
6

206
188
181
202
205

82
104
122
135
144

0.91
0.85
0.85
0.82
0.8

0.72
0.63
0.7
0.77
0.76

0.89
0.92
0.86
0.68
0.75

0.76
0.71
0.78
0.73
0.75

~1 year
~1 year
~1 year
~1 year
~1 year

Note. aCut score was selected by optimizing sensitivity and then balancing it with specificity using methods
presented by Silberglitt and Hintze (2005)

Further diagnostic accuracy analyses were conducted using the Minnesota Comprehensive
Assessment III. Students were administered the Minnesota Comprehensive Assessment III (MCA-III) in
Reading in grades 3, 4, and 5. The MCAs are state tests that help school districts measure student
progress toward Minnesota’s academic standards and meet the requirements of the Elementary and
Secondary Education Act (ESEA). Additionally, students completed three FastBridge Learning
CBMreading probes during the spring. The median score was computed. Only those students
providing complete data were utilized in the diagnostic accuracy analyses. More specifically, students
with incomplete data regarding CBM-R Words Read Correctly (WRC) per minute, or those students
with incomplete MCA-III Achievement Level Scores were excluded from analyses. ROC Analysis was
used to determine diagnostic accuracy of FastBridge Learning CBMreading probes with Spring MCA-III
scale scores serving as the criterion measure. Students were disaggregated by grade level. Diagnostic
accuracy was computed for students identified as being at “High Risk” and those identified as
“Somewhat at Risk” for reading difficulties using MCA-III Achievement Level Criteria (See Table 83).
Data collection is ongoing.
Table 83. Diagnostic Accuracy for CBMreading and MCA III
CBM-R
MCA-III
r(x,y)
Cut a
AUC
M (SD)
M (SD)
High Risk (Does Not Meet Standards)
K
--NA
---1
--NA
---2
--NA
---3
852 139 (40)
348 (20)
.76
131.5
.88
4
818 165 (39)
447 (15)
.71
153.5
.87
5
771 165 (40)
552 (16)
.70
151.5
.89
6 to 12
--Pending
--Somewhat High Risk (Does Not Meet or Partially Meets Standards)
K
--NA
---1
--NA
---2
--NA
---3
852 139 (40)
348 (20)
.76
141.5
.86
4
818 165 (39)
447 (15)
.71
164.5
.83
5
771 165 (40)
552 (16)
.70
163.5
.84
6 to 12
--Pending
--Grade

N

Sensitivity

Specificity

---.80
.80
.80
--

---.79
.78
.79
--

---.78
.75
.77
--

---.76
.71
.76
--

Note. M = Mean. SD = Standard Deviation. aCut score was selected to balance sensitivity and specificity using
methods modified from Silberglitt and Hintze (2005)

Page | 110
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Additional diagnostic accuracy analyses for various grade levels across multiple states is summarized
in the tables on the following page.
Table 84. Diagnostic Accuracy on Fall CBMreading with Spring CRCT in Reading: GA LEA 1 (Fall
to Spring Prediction)
Grade

N

CBMreading
CRCT
M (SD)
M (SD)
Some Risk (Meets Standards)
3
329
115.96 (42)
848.65 (28)
4
320
137.64 (41)
848.18 (27)
5
353
149.27 (40)
841.22 (25)
High Risk (Does Not Meet Standards)
3
329
115.96 (42)
848.65 (28)
4
320
137.64 (41)
848.18 (27)
5
353
149.27 (40)
841.22 (25)

r(x,y)

Cut

AUC

Sens.

Spec.

.66*
.65*
.57*

116.50
130.50
150.50

.79
.81
.71

.72
.72
.66

.71
.73
.66

.66*
.65*
.57*

80.50
100.50
128.50

.89
.85
.82

.82
.83
.79

.83
.85
.71

Table 85. Diagnostic Accuracy on Winter CBMreading on Spring CRCT in Reading: GA LEA 1
(Winter to Spring Prediction)
Grade

N

CBMreading
CRCT
M (SD)
M (SD)
Some Risk (Meets Standards)
3
327
117.22 (43)
848.64 (28)
4
318
136.52 (41)
848.33 (27)
5
350
149.75 (40)
841.14 (25)
High Risk (Does Not Meet Standards)
3
327
117.22 (43)
848.64 (28)
4
318
136.52 (41)
848.33 (27)
5
350
149.75 (40)
841.14 (25)

Page | 111
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

r(x,y)

Cut

AUC

Sens.

Spec.

.64*
.61*
.57*

119.50
133.50
152.50

.76
.78
.71

.69
.70
.65

.69
.71
.66

.64*
.61*
.57*

72.00
101.00
133.50

.92
.90
.80

.83
.83
.64

.87
.84
.68

Section 6. FAST as Evidence-Based Practice
Table 86. Diagnostic Accuracy of Fall CBMreading with Spring MCA-III in Reading: MN LEA 3 (Fall
to Spring Prediction)
CBMreading
MCA-III
Grade
N M (SD)
M (SD)
r(x,y)
Some Risk (Does Not Meet or Partially Meets Standards)
3
488 98.94 (42.72)
353.22 (22)
.71
4
486 126.67 (46.38)
451.25 (22)
.70
5
492 139.96 (40.73)
554.19 (16)
.73
6
463 145.71 (39.80)
654.68 (19)
.68
Some Risk (Partially Meets Standards)
3
488 98.94 (42.72)
353.22 (22)
.71
4
486 126.67 (46.38)
451.25 (22)
.70
5
492 139.96 (40.73)
554.19 (16)
.73
6
463 145.71 (39.80)
654.68 (19)
.68
High Risk (Does Not Meet Standards)
3
488 98.94 (42.72)
353.22 (22)
.71
4
486 126.67 (46.38)
451.25 (22)
.70
5
492 139.96 (40.73)
554.19 (16)
.73
6
463 145.71 (39.80)
654.68 (19)
.68

Cut

AUC

Sens.

Spec.

89.5
126.5
131.5
143.5

.85
.87
.87
.83

.74
.79
.78
.76

.75
.78
.77
.74

-127.5
137.5
151.5

.57
.64
.68
.62

-.70
.72
.70

-.59
.58
.53

77.5
109.5
116.5
132.5

.89
.88
.91
.88

.78
.80
.82
.77

.81
.79
.81
.78

Table 87. Diagnostic Accuracy for Fall CBMreading with Spring Minnesota Comprehensive
Assessment III (MCA-III) in Reading: MN LEA 2 (Fall to Spring Prediction)
CBMreading
MCA
Grade
N M (SD)
M (SD)
r(x,y)
Some Risk (Does Not Meet or Partially Meets Standards)
3
249 107.97 (42)
351.53 (28)
.71**
4
236 134.75 (40)
452.51 (16)
.70**
5
229 154.78 (36)
559.07 (15)
.65**
Some Risk (Partially Meets Standards)
High
3
Risk249
(Does 107.97
not meet
(42)
351.53 (28)
(28)
3
249
107.97
(42)Standards)
351.53
44
236
134.75
(40)
452.51
(16)
236 134.75 (40)
452.51 (16)
5
229 154.78 (36)
559.07 (15)
5
229 154.78 (36)
559.07 (15)

Page | 112
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

.71**
.71**
.70**
.70**
.65**
.65**

Cut

AUC

Sens.

Spec.

106
131.5
146.5

.82
.85
.85

.77
.76
.74

.77
.77
.73

97.5
107.5
117.5
132.5
129.5
150.5

.82
.62
.89
.65
.91
.71

.77
.68
.81
.69
.88
.70

.76
.58
.80
.61
.85
.62

Section 6. FAST as Evidence-Based Practice
Table 88. Diagnostic Accuracy for Winter CBMreading with Spring Minnesota Comprehensive
Assessment III (MCA-III) in Reading: MN LEA 2 (Winter to Spring Prediction)
CBMreading
MCA
N
Grade
M (SD)
M (SD)
Some Risk (Does Not Meet or Partially Meets Standards)
3
250 125.46 (41)
352.28 (26)
4
240 149.38 (40)
451.76 (16)
5
229 164.52 (34)
558.99 (15)
Some Risk (Partially Meets Standards)
3
250 125.46 (41)
352.28 (26)
4
240 149.38 (40)
451.76 (16)
5
229 164.52 (34)
558.99 (15)
High Risk (Does Not Meet Standards)
3
250 125.46 (41)
352.28 (26)
4
240 149.38 (40)
451.76 (16)
5
229 164.52 (34)
558.99 (15)

r(x,y)

Cut

AUC

Sens.

Spec.

.73**
.69**
.65**

125.5
148.5
153.5

.83
.83
.87

.75
.76
.79

.76
.76
.77

.73**
.69**
.65**

--152.5

.59
.60
.74

--.70

--.69

.73**
.69**
.65**

113.5
135.5
138.5

.85
.89
.91

.78
.80
.88

.79
.80
.85

Table 89. Diagnostic Accuracy for Winter CBMreading with MCA-III in Reading: MN LEA 3 (Winter
to Spring Prediction)
CBMreading
MCA-III
Grade
N M (SD)
M (SD)
Some Risk (Does Not Meet or Partially Meets Standards)
3
496 123.11 (44)
353.06 (22)
4
497 145.20 (48)
450.97 (22)
5
497 159.21 (42)
554.22 (15)
6
466 162.05 (41)
654.55 (19)
Some Risk (Partially Meets Standards)
3
496
123.11 (44)
353.06 (22)
4
497
145.20 (48)
450.97 (22)
5
497
159.21 (42)
554.22 (15)
6
466
162.05 (41)
654.55 (19)
High Risk (Does Not Meet Standards)
3
496
123.11 (44)
353.06 (22)
4
497
145.20 (48)
450.97 (22)
5
497
159.21 (42)
554.22 (15)
6
466
162.05 (41)
654.55 (19)

Page | 113
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

r(x,y)

Cut

AUC

Sens.

Spec.

.73
.69
.72
.67

117.5
144.5
150.5
162.5

.86
.87
.86
.81

.77
.78
.77
.73

.79
.76
.79
.73

.73
.69
.72
.67

-147.5
156.5
167.5

.56
.63
.65
.58

-.69
.67
.68

-.56
.54
.50

.73
.69
.72
.67

105.5
129.0
134.5
148.5

.90
.88
.93
.88

.81
.80
.85
.81

.82
.79
.85
.83

Section 6. FAST as Evidence-Based Practice
Table 90. Diagnostic Accuracy of Winter CBMreading with Spring Massachusetts Comprehensive
Assessment (MCA): MA LEA 1 (Winter to Spring Prediction)
CBMreading
MCA
Grade
N
M (SD)
M (SD)
Some Risk (“Warning” and “Needs Improvement”)
3
92
123.82 (38)
241.89 (14)
4
93
134.33 (41)
238.40 (15)
5
75
153.03 (26)
243.63 (13)
Some Risk (“Needs Improvement”)
3
92
123.82 (38)
241.89 (14)
4
93
134.33 (41)
238.40 (15)
5
75
153.03 (26)
243.63 (13)
High Risk (“Warning”)
3
92
123.82 (38)
241.89 (14)
4
93
134.33 (41)
238.40 (15)
5
75
153.03 (26)
243.63 (13)

r(x,y)

Cut

AUC

Sens.

Spec.

.71**
.68**
.66**

114
132.5
155.5

.82
.81
.90

.76
.73
.76

.75
.73
.74

.71**
.68**
.66**

116.5
134
154.5

.68
.69
.88

.72
.72
.75

.64
.60
.77

.71**
.68**
.66**

90.5
109.5
130.5

.97
.89
.82

.88
.78
1.00

.89
.79
.80

aReading
aReading diagnostic accuracy was derived from a sample of 777 students in first through Fifth Grades
from two suburban schools in the Midwest. In the sample, 116 students were in First Grade, 188 in
second, 159 in third, 156 in fourth, and 158 in Fifth Grade. Gender of the sample was approximately
49% female and 51% male. Approximately 67% of students in the sample were White, 5% Black, 19%
Asian/Pacific Islander, 5% Hispanic, 2% American Indian, and 2% unspecified. In addition, 10% of
students were receiving special education services, and 10% of students were classified as having
limited English language proficiency. Socioeconomic status information was not available for the
sample, but the schools the students were drawn from had rates of free and reduced lunch of 13% and
23% in 2009–10. Cut scores for aReading to predict students “At Risk” and “Somewhat At Risk” for
reading difficulties were developed for the Gates-MacGinitie Reading Tests–Fourth Edition (GMRT-4th;
MacGinitie, MacGinitie, Maria, & Dreyer, 2000) and the Measures of Academic Progress (MAP).
Categories for the former were defined as students scoring below the 40th and 20th percentiles of the
local sample and cut scores for each category developed by an adjacent school district for MAP were
used on this sample. An additional analysis regarding diagnostic accuracy of aReading using The
Minnesota Comprehensive Assessments (MCAs) as the criterion measure is briefly discussed. At the
beginning of the school year (October, 2010) students completed an aReading assessment. The
measure was group administered via a mobile computer lab. Scaled scores were calculated for each
student. In February 2011, the same students completed the GMRT-4th. Composite scores were
available for all grades except Second Grade. Due to time constraints, one subtest could not be
administered to Second Grade students (the only grade that requires three subtests to yield a
composite score). As a result, comprehension subtest scores were used for analysis. The GMRT-4th was
group administered by a team of graduate students. Administrators completed advanced coursework
in psychological assessment and completed an in-service training to administer the test. Test booklets
Page | 114
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
were hand scored and inter-rater reliability was 100% across all subtest and composite scores. MAP
scores for spring testing were provided to the aReading team from school administrators.
Table 91 below presents the ROC curve analysis results for each grade for students at high risk and
somewhat at risk using the Youden Index with the GMRT-4th. In addition, sensitivity, specificity, PPP
and NPP are displayed for the each grade for each cut score. ROC curves across grades and risk levels
were far from the diagonal line indicating that aReading predicts reading difficulties with much more
accuracy than chance. Evaluation of the table below indicated that across grades, AUC statistics were
extremely high, especially for students at high risk (Mdn = .92) and values were still high for students
somewhat at risk (Mdn = .87). In addition, Sensitivity was higher for each grade when determining
students at high risk compared to somewhat risk. Positive Predictive Power was higher across grades
when predicting students at somewhat risk (Mdn = .72 versus .56) while the opposite was true for
Negative Predictive Power (Mdn = .82 versus .96) respectively.
Table 91. Diagnostic Accuracy statistics for aReading and GMRT-4 th
Grade
N
aReading Cut Score
High Risk – Below 20th Percentile a
1
116
430
2
188
461
3
159
490
4
156
495
5
159
506
Somewhat at Risk – Below 40th Percentile
1
116
436
2
188
477
3
159
490
4
156
506
5
159
522

Sensitivity

Specificity

PPP

NPP

AUC

.88
.70
.97
.85
.85

.87
.93
.77
.92
.84

.66
.74
.50
.72
.59

.96
.92
.99
.96
.95

.94
.88
.92
.94
.87

.76
.86
.82
.72
.83

.86
.71
.87
.82
.76

.81
.71
.78
.73
.71

.82
.86
.90
.81
.86

.91
.87
.89
.82
.85

Note. aThe 20th percentile was used for this sample, which should approximate the 15th percentile.

A similar pattern of results emerged when predicting performance on the MAP (See Table 92).
Compared to the GMRT-4th as a criterion, NPP was much higher when predicting MAP scores. This
could be attributed to the fact that the base rate of students at risk was much lower for MAP scores.
Data collection for Kindergarten and for grades 6–12 are ongoing, and results are pending.

Page | 115
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 92. Diagnostic Accuracy Statistics for aReading and MAP
Grade
N
aReading Cut Score
High Risk - 20th Percentile a
2
188
497
3
159
517
4
156
537
5
159
537
Somewhat at Risk - 40th Percentile
2
188
490
3
159
527
4
156
537

Sensitivity

Specificity

PPP

NPP

AUC

1
.95
.96
1

.73
.76
.78
.82

.14
.21
.30
.20

1
1
1
1

.89
.95
.94
.93

.77
.89
.82

.84
.77
.87

.41
.46
.65

.96
.97
.94

.89
.89
.92

5

.93

.77

.39

.99

.88

159

547

Note. The 20 percentile was used for this sample, which should approximate the 15 percentile.
a

th

th

Finally, diagnostic accuracy analyses were conducted with aReading and the Minnesota
Comprehensive Assessment (MCA) to determine if aReading predicted state reading assessments. The
MCA is a state test that helps school districts measure student progress toward academic standards
and meets the requirements of the Elementary and Secondary Education Act (ESEA).
The sample consisted of 1,786 students in third, fourth, and Fifth Grades from eight schools in the
upper Midwest (MCAs are not administered to students in grades K–1). In the sample, 631 students
were in third, 618 students were in fourth, and 537 students were in Fifth Grade. Gender of the sample
was approximately 50% female and 50% male. Ethnic breakdown was approximately 45% White, 23%
Black, 8% Asian/Pacific Islander, 15% Hispanic, 0.8% American Indian or Alaska Native, and 9%
multiracial. (These percentages were rounded and may not add to 100). In addition, 12% of students
were receiving special education services. Socioeconomic status information was not available for the
sample, but the schools the students were drawn from had rates of free and reduced lunch ranging
from 16% to 83% in 2013.
Students completed the Adaptive Reading (aReading) assessment and MCAs during the spring of
2013. Students with incomplete data in aReading, or those students with incomplete MCA
Achievement Level Scores were excluded from analyses.
ROC Analysis was used to determine diagnostic accuracy of FAST™ aReading with Spring MCA scale
scores serving as the criterion measure. Students were disaggregated by grade level. Diagnostic
accuracy was computed for students at “High Risk” and “Somewhat At Risk” on MCA Scale Scores.
“High Risk” includes those students that did not meet standards. “Somewhat At Risk” includes those
students who did not meet or only partially met standards. Diagnostic accuracy statistics are provided
in Table 93. Data collection is ongoing for all grade levels.
Table 93. Diagnostic Accuracy for aReading and MCA-III
Grade
aReading
MCA-III
N
Level
M (SD)
M (SD)
High Risk (Does Not Meet Standards)
Page | 116
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

r(x,y)

Cut a

AUC

Sens.

Spec.

Section 6. FAST as Evidence-Based Practice
3
629
534 (27)
347 (21)
.82
524.5
4
615
549 (28)
447 (16)
.81
536.5
5
516
564 (32)
553 (16)
.84
544.5
Somewhat High Risk (Does Not Meet or Partially Meets Standards)
3
629
534 (27)
347 (21)
.82
532.5
4
615
549 (28)
447 (16)
.81
550.5
5
516
564 (32)
553 (16)
.84
556.5

.90
.92
.96

.82
.84
.89

.81
.84
.86

.90
.89
.93

.83
.82
.84

.82
.82
.84

More recently, the following diagnostic accuracy statistics were derived from samples of students
across various states, using various criterion measures.
Table 94. Diagnostic Accuracy of Spring aReading with Spring MAP in Reading: WI LEA 1 (Spring
Data Collection)
Grade

N

aReading
M (SD)
Some Risk (≤ 40 th percentile)
2
33
477.61 (22.09)
3
26
496.19 (26.13)
4
31
509.35 (13.51)
5
28
514.29 (13.49)
6
25
521.48 (19.44)

MAP
M (SD)

r(x,y)

Cut

AUC

Sens.

Spec.

181.61 (19.34)
195.65 (17.25)
208.23 (13.25)
211.11 (12.90)
215.08 (11.56)

.90**
.95**
.83**
.79**
.85**

474.5
499.00
504.00
514.5
520.5

.96
1.00
.94
.90
.92

.91
1.00
.86
.89
.86

.86
.89
.87
.79
.83

Some Risk (20 th to 40 th percentile)
2
33
477.61 (22.09)
3
26
496.19 (26.13)
4
31
509.35 (13.51)
5
28
514.29 (13.49)
6
25
521.48 (19.44)

181.61 (19.34)
195.65 (17.25)
208.23 (13.25)
211.11 (12.90)
215.08 (11.56)

.90**
.95**
.83**
.79**
.85**

473
499
504
514.5
520.5

.79
.86
.79
.72
.74

1.00
1.00
.75
.80
.75

.70
.73
.78
.65
.71

High Risk (≤ 20 th percentile)
2
33
477.61 (22.09)
3
26
496.19 (26.13)
4
31
509.35 (13.51)
5
28
514.29 (13.49)
6
25
521.48 (19.44)

181.61 (19.34)
195.65 (17.25)
208.23 (13.25)
211.11 (12.90)
215.08 (11.56)

.90**
.95**
.83**
.79**
.85**

466.00
475.00
501.00
503.5
514.5

.97
1.00
1.00
.95
1.00

.86
1.00
1.00
.75
1.00

.85
.92
.89
.87
.86

Page | 117
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 95. Diagnostic Accuracy of Fall aReading with Spring MCA-III in Reading: MN LEA 3 (Fall to
Spring Prediction)
aReading
MCA-III
M (SD)
M (SD)
Some Risk (Does not Meet or Partially Meets Standards)
3
482 484.45 (19.73)
353.43 (21.78)
4
485 496.62 (22.72)
452.05 (17.67)
5
495 504.87 (19.72)
554.17 (15.46)
6
459 511.37 (22.43)
654.81 (18.75)
Some Risk (Partially Meets Standards)
3
482 484.45 (19.73)
353.43 (21.78)
4
485 496.62 (22.72)
452.05 (17.67)
5
495 504.87 (19.72)
554.17 (15.46)
6
459 511.37 (22.43)
654.81 (18.75)
High Risk (Does Not Meet Standards)
3
482 484.45 (19.73)
353.43 (21.78)
4
485 496.62 (22.72)
452.05 (17.67)
5
495 504.87 (19.72)
554.17 (15.46)
6
459 511.37 (22.43)
654.81 (18.75)
Grade

N

r(x,y)

Cut

AUC

Sens.

Spec.

.77
.77
.80
.73

478.5
499.5
503.5
508.5

.89
.89
.91
.85

.82
.80
.82
.75

.82
.82
.81
.76

.77
.77
.80
.73

474.5
487.5
499.5
504.5

.89
.94
.93
.96

.86
1.00
.94
1.00

.71
.71
.71
.70

.77
.77
.80
.73

475.5
490.5
497.5
504.5

.90
.92
.93
.89

.79
.82
.84
.83

.83
.82
.85
.81

Table 96. Diagnostic Accuracy of Winter aReading with Spring MCA-III in Reading: MN LEA 3
(Winter to Spring Prediction)
aReading
MCA-III
M (SD)
M (SD)
Some Risk (Does Not Meet or Partially Meets Standards)
5
134 519.97 (15.30)
560.84 (13.80)
6
160 524.56 (18.42)
661.15 (16.55)
Some Risk (Partially Meets Standards)
5
134 519.97 (15.30)
560.84 (13.80)
6
160 524.56 (18.42)
661.15 (16.55)
High Risk (Does Not Meet Standards)
5
134 519.97 (15.30)
560.84 (13.80)
6
160 524.56 (18.42)
661.15 (16.55)
Grade

N

r(x,y)

Cut

AUC

Sens.

Spec.

.77
.70

514.5
520.5

.89
.80

.83
.70

.84
.74

.77
.70

514.5
522.5

.79
.68

.71
.74

.77
.67

.77
.70

510.5
514.5

.93
.92

.80
.83

.82
.84

aReading evidence of diagnostic accuracy is not limited to the Midwest. The following diagnostic
accuracy information was obtained from samples of students in other regions of the US.

Page | 118
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 97. Diagnostic Accuracy Fall aReading with Spring Massachusetts Comprehensive
Assessment (MCA): Cambridge, MA (Fall to Spring Prediction)
Grade

N

aReading
MCA
M (SD)
M (SD)
Some Risk (“Warning” and “Needs Improvement”)
3
93
485.91 (17.59)
241.89 (14.08)
4
93
492.68 (18.52)
238.40 (15.14)
5
72
506.93 (18.58)
243.63 (13.20)
Some Risk (“Needs Improvement”)
3
93
485.91 (17.59)
241.89 (14.08)
4
93
492.68 (18.52)
238.40 (15.14)
5
72
506.93 (18.58)
243.63 (13.20)
High Risk (“Warning”)
3
93
485.91 (17.59)
241.89 (14.08)
4
93
492.68 (18.52)
238.40 (15.14)
5
72
506.93 (18.58)
243.63 (13.20)

r(x,y)

Cut

AUC

Sens.

Spec.

.63**
.69**
.69**

478.5
492.5
507.5

.79
.85
.90

.73
.75
.85

.73
.78
.79

.63**
.69**
.69**

480.5
494.5
506.5

.71
.72
.88

.72
.71
.78

.63
.60
.87

.63**
.69**
.69**

475.5
476.5
495.5

.80
.92
.85

.75
.89
1.00

.72
.84
.79

Table 98. Diagnostic Accuracy of Winter aReading with Spring Massachusetts Comprehensive
Assessment (MCA): MA LEA 1 (Winter to Spring Prediction)
Grade

N

aReading
MCA
M (SD)
M (SD)
Some Risk (“Warning” and “Needs Improvement”)
3
91
498.99 (18.66)
241.89 (14.08)
4
94
504.33 (16.37)
238.40 (16.37)
5
74
515.31 (14.42)
243.63 (14.42)
Some Risk (“Needs Improvement”)
3
91
498.99 (18.66)
241.89 (14.08)
4
94
504.33 (16.37)
238.40 (16.37)
5
74
515.31 (14.42)
243.63 (14.42)
High Risk (“Warning”)
3
91
498.99 (18.66)
241.89 (14.08)
4
94
504.33 (16.37)
238.40 (16.37)
5
74
515.31 (14.42)
243.63 (14.42)

Page | 119
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

r(x,y)

Cut

AUC

Sens.

Spec.

.76**
.69**
.61**

499.5
505.5
516.5

.85
.85
.83

.76
.76
.82

.74
.78
.78

.76**
.69**
.61**

501.5
506.5
516.5

.71
.71
.81

.72
.78
.81

.64
.64
.76

.76**
.69**
.61**

476.5
485.5
506.00

.97
.94
.86

.88
.78
1.00

.95
.88
.80

Section 6. FAST as Evidence-Based Practice
Table 99. Diagnostic Accuracy of Fall aReading with Spring CRCT in Reading: GA LEA 1
(Fall to Spring Prediction)
Grade

N

aReading
CRCT
M (SD)
M (SD)
Some Risk (Meets Standards)
3
329
483.81 (18)
848.65 (28)
4
320
491.37 (16)
848.18 (27)
5
353
497.81 (16)
841.22 (25)
High Risk (Does Not Meet Standards)
3
329
483.81 (18)
848.65 (28)
4
320
491.37 (16)
848.18 (27)
5
353
497.81 (16)
841.22 (25)

r(x,y)

Cut

AUC

Sens.

Spec.

.73*
.64*
.64*

481.50
490.50
499.50

.83
.80
.75

.76
.73
.70

.76
.75
.68

.73*
.64*
.64*

466.50
478.50
485.00

.94
.89
.89

.82
.83
.79

.86
.76
.79

Table 100. Diagnostic Accuracy of Winter aReading with Spring CRCT in Reading: GA LEA 1
(Winter to Spring Prediction)
Grade

N

aReading
CRCT
M (SD)
M (SD)
Some Risk (Meets Standards)
3
327
495.67 (18)
848.64 (28)
4
318
505.31 (16)
848.33 (27)
5
351
512.19 (15)
841.14 (25)
6
283
518.78 (13)
850.14 (23)
High Risk (Does Not Meet Standards)
3
327
495.67 (18)
848.64 (28)
4
318
505.31 (16)
848.33 (27)
5
347
512.19 (15)
841.14 (25)
6
283
518.78 (13)
850.14 (23)

r(x,y)

Cut

AUC

Sens.

Spec.

.75*
.71*
.66*
.67*

498.50
505.50
516.50
519.50

.83
.82
.78
.87

.76
.77
.71
.77

.76
.78
.72
.80

.75*
.71*
.66*
.67*

477.50
487.50
500.50
NA

.95
.94
.92
NA

.83
.83
.86
NA

.86
.76
.85
NA

These findings have also been demonstrated in higher grade levels. The following diagnostic accuracy
information was obtained from a sample of approximately 322 7th grade students (50.0% female) and
311 8th grade students (50.3%) in a Georgia Local Education Agency (LEA). Approximately 74.7% of 7th
grade students were White, 8.6% were Hispanic, 8.3% were African American, 5.6% were Multiracial,
2.5% were Asian, and .3% identified themselves as “Other.” Approximately 81.2% of 8th grade students
were White, 7.3% were African American, 5.7% were Hispanic, 4.1% were Multiracial, 1.3% were Asian,
and .3% identified themselves as “Other.” See Table 101 for diagnostic accuracy statistics.

Page | 120
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Table 101. Diagnostic Accuracy of Winter aReading with Spring Criterion-Referenced
Competency Tests (CRCT) in Reading: Georgia LEA 1 (Winter to Spring Prediction)
Grade

N

aReading
CRCT
M (SD)
M (SD)
Some Risk (Meets Standards)
7
322 521.77 (14.86)
842.50 (24.09)
8
311 524.69 (21.76)
850.36 (54.42)
High Risk (Does Not Meet Standards)
7
322 521.77 (14.86)
842.50 (24.09)
8
311 524.69 (21.76)
850.36 (54.42)

r(x,y)

Cut

AUC

Sens.

Spec.

.64**
.33**

512.5
517.5

.91
.95

.82
1.00

.81
.73

.64**
.33**

509.5
511.5

.92
.92

.86
.86

.86
.84

Table 102. Diagnostic Accuracy of Fall aReading with Spring Minnesota Comprehensive
Assessment III (MCA-III) in Reading: MN LEA 2 (Fall to Spring Prediction)
Grade

N

aReading
MCA
M (SD)
M (SD)
Some Risk (Meets Standards)
10
66
527.91 (18.59)
1052.29 (12.99)
High Risk (Does Not Meet Standards)
10
66
527.91 (18.59)
1052.29 (12.99)

r(x,y)

Cut

AUC

Sens.

Spec.

.55**

523.5

.82

.77

.75

.55**

521.5

.77

.75

.77

For additional diagnostic accuracy information for FastBridge Learning Reading Assessments, please
see Appendix B: FastBridge Learning Reading Diagnostic Accuracy.

Section 6. FAST™ as Evidence-Based Practice
This section provides a summary of some evidence for FAST™ as an evidence-based intervention. The
use of well-developed training and support with technology to deliver and automate the use of
formative assessments and data improve student outcomes. FAST™ improves teacher knowledge,
skills, and appreciation for data. The use of FAST™ changes teaching.
The graphic below is our theory of change, which relates to our hypothesis for systems change and
improved student outcomes caused by the adoption and use of FAST™.

Page | 121
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice

6.1: Theory of Change
FastBridge Learning and our
researchers and developers have a
theory of change (right). Our theory is
that student outcomes improve when
teachers use evidence-based
formative assessments in conjunction
with technology-based training and
supports for data-based decision
making. We provide evidence for this
theory of change below.
FastBridge Learning’s researchers and
developers work continuously to
refine and improve the tools. They
work with teachers and educators to
evaluate their needs and satisfaction.
They collect and evaluate student data Figure 3 Theory of Change
to ensure alignment with standards
and improved student outcomes.

6.2: Formative Assessment as Evidence-Based Practice
Effective teachers use formative assessment to guide their practices. This is recommended and
supported in the empirical and professional literature. Teachers and students benefit less from
summative assessments, which occur infrequently, have little instructional relevance, and yield results
that are often delayed for days, weeks, or months (e.g., many state testing programs).
Teachers need an effective formative assessment system. Effective systems provide assessments,
reporting, and guidance for data interpretation and use. Moreover, they support the multi-method
and multi-source approach, which requires multiple types of assessments with varied methods across
the most relevant content areas.
US Department of Education
The US Department of Education’s Practice Guides summarize the evidence and recommendations for
effective practice (http://ies.ed.gov/pubsearch). They recommend formative assessment and
evaluation.
Those guides were developed by panels of national experts who “relied on the WWC Evidence
Standards” (Gersten, 2008, p. 2). After a review of the evidence, the expert panel for Using Student
Achievement Data to Support Instructional Decision Making recommended that educators: (1) make
data part of an ongoing cycle of instructional improvement, (2) have a clear vision for schoolwide data
Page | 122
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
use, (3) foster a data-driven culture, and (4) maintain a districtwide data system (Hamilton et al., 2009,
p. 8). The RTI reading panel found moderate evidence to recommend universal screening and
progress monitoring for students with deficits (Gersten et al., 2008, p. 13). The RTI math panel made
substantially similar recommendations (Gersten et al., 2009, p. 11).
All of the expert panels recommend the use of high-quality and highly efficient formative
assessments. They also recommend the use of assessment data to plan instruction, differentiate
instruction, titrate support (tiered services), devise instructional groups, monitor progress, and
evaluate instructional effects. Some recommendations had lower levels of evidentiary support, but
FAST™ is designed to facilitate the implementation of these recommended practices.
Historical Evidence on Formative Assessment
Effective teachers systematically collect and share student assessment data to help them make
instructional decisions that improve student performance (Lipson, Mosenthal, Mekkelsen, & Russ,
2004; Taylor et al., 2000) by 0.4 to 0.7 standard deviations (Black & Wiliam, 1998).
Teachers should be able to use data to inform their practice (Martone & Sireci, 2009). After a thorough
review of the pre-1990 research literature on effective instruction, Hoffman (1991) concluded that
there is persistent and consistent evidence that the use of instructionally relevant assessments
improved instructional effects and student achievement. Those findings converge with those of
contemporary research. For example, Pressley et al. (2001) identified 103 behaviors that distinguished
teachers who were either highly effective or moderately effective. Classroom assessment practices—
as opposed to external assessments—were a major distinguishing factor in instructionally relevant
assessments. That is, the use of formative assessment data to guide instruction contributes to more
effective instruction and higher student achievement (Jenkins, 2001; Taylor, Pearson, Clark, & Walpole,
2000a, 2000b; Taylor, Pearson, Peterson, & Rodriguez, 2003, 2005).
Formative assessments guide both instruction and student learning. They provide feedback that is
linked to explicit performance standards and provide guidance to achieve those standards (Sadler,
1989). Formative assessment helps establish a learning-assessment process, which is encapsulated by
three key questions (Atkin, Black, & Coffey, 2001; Deno, 2002, 2005; Sadler, 1989):


“What is the expected level of performance?”



“What is the present level of performance?” and



“What must be done to reduce that discrepancy?”

FAST™ provides useful data for educators to address each of these questions through screening, skills
analysis, and progress monitoring. It also provides training, supports, analysis, and reporting to
facilitate the use of data by teachers.
Evidence Based: Contemporary Evidence on Formative Assessment
Data-based decision making and formative assessment are evidence-based. Black and Wiliam (1998)
and other common sources of evidence were cited above; however, the most recent and rigorous
Page | 123
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
meta-analysis on formative assessment was done by Kingston and Nash (2011). The inclusion criteria
for the meta-analysis were: (a) formative assessment was the intervention, (b) K–12 student samples,
(c) control or comparison group, (d) appropriate statistics to estimate effect size, and (e) published
1988 or later. They identified 13 studies with 42 independent effect sizes, which allowed them to
analyze the effects of professional development and computer-based formative assessment—among
other things.
In brief, the weighed mean effect of formative assessment was .20 (Mdn = .25). The largest effects
were observed in the content area of English language arts (.32) with less robust effects in
mathematics (.17); however, there were statistically significant (moderating) effects for both
professional development/training (.30) and technology-based formative assessments (.28). Results
improved when teachers received training and technology to support the implementation and use of
formative assessments. Preliminary evidence indicates that FAST™ is likely to confer even more
substantial effects.
Although recent research (Kingston & Nash, 2011) indicates more modest effects for formative
assessment, these are substantial and meaningful differences in student achievement.
As summarized in Table 103, if 80% of students are proficient (i.e., not in need of supplemental,
intensive, or special education services) and formative assessment is implemented as an intervention,
the proficiency rate is likely improve to 87% (assuming a .30 effect size). That is, the percentage of
students with deficits or disabilities who need RTI or special education is reduced from 20% to 13%.
That is a 35% reduction. Larger effects are observed with the implementation of FAST™. This is
described as follows.
Table 103. Estimates of the Increase in the Percentage of Students who are Proficient or above
with the Implementation of Formative Assessment (Kingston & Nash, 2011, p. 35)
Proficient

Note. “While the weighted mean effect sizes found in this study are smaller than commonly reported in the
literature, they have great practical significance in today’s accountability climate. These improvements can be
restated in terms of the percentage of students achieving summative assessment results at or above the proficient
level. Table 350 shows the improvement that would result from several different effect sizes based on how many
students are currently proficient or above (assuming scores are distributed normally). If currently 20% of students
are proficient or above, then the weighted mean effect size of .20 would lead to an additional 6% of students
achieving proficient status. If currently 50% of students are proficient or above, then the increase would be 8%.
The .30 effect size associated with formative assessment based on professional development would lead to 9% and
12% of all students moving into the proficient category under these two scenarios.” (Kingston & Nash, pp. 34-35).

Page | 124
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
Formative assessment is an evidence-based practice with the most robust effects when it is delivered
with technology, training, and support. The evidence summarized by Kingston & Nash (2011; N = 13
studies, 42 effect sizes) meets the What Works Clearinghouse criteria for “Positive effects: Strong
evidence of positive effect with no overriding contrary evidence” (WWC, 2014, p. 29).

6.3: Evidence-Based: Formative Assessment System for Teachers
FAST™ combines all aspects of those evidence-based practices described above: technology tools for
assessment, training, and support. Two sources of evidence are presented to illustrate that the
implementation and use of FAST™ is an evidence-based practice.
FAST™ Improves Student Achievement
We analyzed grade-level performance of suburban Midwestern local educational agencies (Schools =
8) with aReading data (broad measure of reading achievement). Performance data from the fall of year
1 implementation (i.e., pre-test) were compared to fall of year 2 implementation (e.g., post-test) to
evaluate the effect of FAST™. That is, the difference in performance between second graders in 2013
(control, M = 445) and second graders in 2014 (after teachers implemented FAST™ for one year; M =
451) might be attributed, in part, to the FAST™ implementation.
There were statistically significant differences with meaningful effect sizes in both general and special
education samples (Table 104, p. 125). This was observed at the district and school levels. Although
not all differences were statistically significant (viz., special education) that is attributed to statistical
power—as effect sizes were still robust. The observed effect sizes converge with the findings of
Kingston and Nash (2011). These are meaningful and important improvements that replicated across
all grades and populations except 6th grade special education.
If combined with the estimates in Table 103, FAST™ is likely to reduce the proportion of students at
risk and increase the proportion who achieve proficiency by 7 to 13%--or more.
Table 104. FAST™ Statistical Significance and Effect Sizes
2013

2014

Grade

Group

M

SD

N

M

SD

N

t

df

p

d

2nd

GenEd

463.3

25.8

578

469.5

27.7

523

3.80

1099

.00*

.23

SpEd

445.6

25.1

51

451.6

33.9

46

.10

95

.32

20

GenEd

484.5

19.3

507

492.4

24.3

559

5.83

1064

.00*

36

SpEd

470.2

25.7

46

474.6

26.6

57

.86

101

.40

17

GenEd

498.3

22.0

507

504.6

21.7

513

4.61

1018

.00*

29

SpEd

476.7

26.1

62

487.9

28.8

52

2.17

112

.03*

41

GenEd

504.3

19.4

483

514.7

23.2

505

7.54

986

.00*

48

SpEd

490.8

26.1

76

498.5

24.9

64

1.77

136

.08

30

GenEd

508.8

25.1

431

522.3

24.1

475

8.23

904

.00*

.55

3rd
4th
5th
6th
Page | 125
www.fastbridge.org

© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Section 6. FAST as Evidence-Based Practice
2013
Grade

2014

Group

M

SD

N

M

SD

N

t

df

p

d

SpEd

506.3

20.0

71

506.1

29.5

80

-.04

149

.97

.01

Note. Adaptive Reading (aReading) is a computer-adaptive test of broad reading on a score scale range of 350 to
650, which spans K to 12th grade achievement. Aggregate data are presented for 8 schools. Grade level
performance was compared across the first two years of implementation.

FAST™ Improves the Practice of Teachers
An independent survey of teachers (N=2689 responses) by the Iowa Department of Education
indicates that 86% of educators believe FAST™ “may” or “will definitely” support increased student
achievement (Greg Felderman, Iowa Department of Education, personal communication, January
2015). It also indicated that teachers use data to guide instruction and improve student achievement.


82% of teachers used FAST™ assessment data to form instructional groups (N = 401 teachers
responded),



82% adjusted interventions for students with deficits or disabilities (N = 369); 66% used data at
least once per month (66%), and



25% used FAST™ data at least weekly (N = 376).

These results provide insight as to the effects of FAST™ implementation.
FAST™ Provides High Quality Formative Assessments
As summarized above, the IES practitioner guides recommend the use of high quality assessments.
The FAST™ assessments meet the IES recommendations for quality: reliability, validity, usability,
efficiency, diagnostic accuracy, specificity, sensitivity, efficiency, and instructional relevance.
The FAST™ assessments are evidence-based. Numerous studies were completed with diverse samples
of students across many geographic locations and LEAs (e.g., NY, GA, MN, IA, and WI). Consistent with
the definitions of “evidence-based,” there are many large, multi-site studies with student samples
from the populations and settings of interest (i.e., K–12 students). The samples size for almost all
studies well-exceeded the requirement of 50 students per condition (e.g., assessment, grade, LEA,
instructional condition). On aggregate, more than 15,000 students participated in well-controlled
psychometric research. In addition, norms were developed from samples of approximately 8,000
students per grade (K to 8th) per assessment, which aggregates to 72,000 student participants.
Consistent with the requirements for evidence, the psychometric qualities for reliability and validity
were statistically significant, and the various assessments are meaningful and statistically robust
indicators of relevant outcomes, such as state tests and future performance in school.

Page | 126
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References

References
Adams, M. J. (1990). Beginning to read. Cambridge, MA: MIT Press.
AERA, APA, NCME. (1999). Standards for educational and psychological testing. Washington DC:
American Educational Research Association.
AIMSWeb Benchmark and Progress Monitoring System for Grades K-8. Pearson Education Inc.
Ardoin, S. P., & Christ, T. J. (2009). Curriculum based measurement of oral reading: Estimates of
standard error when monitoring progress using alternate passage sets, School Psychology Review, 38,
266-283.
Ardoin, S. P., Carfolite, J., Christ, T. J., Roof, C. M., & Klubnick, C. (2010). Examining readability estimates’
predictions of students’ oral reading rate: Spache, Lexile, and Forcast. School Psychology Review,
39(2), 277-285.
Ardoin, S. P., Suldo, S. M., Witt, J. C., Aldrich, S., & McDonald, E. (2005). Accuracy of readability estimates'
predictions of CBM performance. School Psychology Quarterly, 20(1), 1-22.
Armbruster, B. (2002). Put reading first DIANE Publishing. Aud, S., Wilkinson-Flicker, S., Kristapovich, P.,
Rathbun, A., Wang, X., & Zhang, J. (2013). The Condition of Education 2013 (NCES 2013-037). U.S.
Department of Education, National Center for Education Statistics. Washington, DC. Retrieved
[November 21st, 2013] from http://nces.ed.gov/pubsearch.
Atkin, J. M., Black, P., & Coffey, J. (2001). Classroom Assessment and the National Science Education
Standards. Washington, DC: National Academy Press.
Baroody, A. J. (2006). Why children have difficulties mastering the basic number combinations and
how to help them. Teaching Children Mathematics, 13, 22 – 31.
Bear, D.R., Invernizzi, M.A., Templeton S.R., & Johnston, F.R. (2012). Words Their Way: Word Study for
Phonics, Vocabulary, and Spelling Instruction, 5th ed. Upper Saddle, NJ: Pearson Education Inc.
Beitchman J, Wilson B, Johnson CJ, Atkinson L, Young A, Adlaf E, et al. (2001). Fourteen-year follow-up
of speech/language-impaired and control children: psychiatric outcome. Journal of the American
Academy of Child and Adolescent Psychiatry, 40, 75–82.
Berch, D. B., & Mazzocco, M. M. (Eds) (2008). Why is math so hard for some children? The nature and
origins of mathematical learning difficulties and disabilities. Baltimore, MD: Paul H. Brookes.
Betts, J., Pickart, M., & Heistad, D. (2009). An investigation of the psychometric evidence of CBM-R
passage equivalence: Utility of readability statistics and equating for alternate forms Journal of School
Psychology, 47, pp. 1–17.

Page | 127
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Black, P., & William, D. (1998). Assessment and classroom learning. Assessment in Education: Principles,
Policy & Practice, 5(1), 7-74.
Bowers, P. N., Kirby, J. R., & Deacon, S. H. (2010). The Effects of Morphological Instruction on Literacy
Skills A Systematic Review of the Literature. Review of Educational Research, 80(2), 144-179.
Bowman, B., Donovan, S., & Burns, M.S. (Eds.). (2000). Eager to learn: Educating our preschoolers.
National Research Council Commission on Behavioral and Social Sciences and Education. Washington,
D.C.: National Academy Press.
Brennan, R. L. (Ed.). (2006). Educational Measurement (4th ed.). Westport, CT: American Council on
Education and Praeger Publishers.
Brown, E.D., Sax, K.L. (2013). Arts enrichment and preschool emotions for low-income children at risk.
Early Childhood Research Quarterly, 28, 337-346.
Burt, J. S. (2006). What is orthographic processing skill and how does it relate to word identification in
reading? Journal of Research in Reading, 29(4), 400-417.
Cain, K., Oakhill, J., & Bryant, P. E. (2004). Children's reading comprehension ability: Concurrent
prediction by working memory, verbal ability, and component skills. Journal of Educational
Psychology, 96, 31-42.
Campbell, F. A., & Ramey, C. T. (1994). Effects of early intervention on intellectual and academic
achievement: A follow‐up study of children from low‐income families. Child Development, 65, 684698.
Carnine, D.W., Silbert, J., Kem’ennui, E.J., & Tarver, S.G. (2009). Direct instruction reading (5th ed.). , OR:
Pearson Education Ltd.
Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. New York:
Cambridge University Press.
Carlisle, J.F. (1995). Morphological awareness and early reading achievement, L.B. Feldment, (ed.),
Morphological aspects of language processing, (pp. 189-209). Hillsdale, NJ: Lawrence Erlbaum
Associates, Publishers.
Carlisle, J. F. (2011). Effects of instruction in morphological awareness on literacy achievement: An
integrative review. Reading Research Quarterly, 45 (4), 464-487.
Case, S.M., & Swanson, D.B. (2002). Constructing written test questions for the basic and clinical
sciences, 3rd edition. Philadelphia, PA: National Board of Medical Examiners.

Page | 128
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Cawley, J. F., Parmar, R. S., Lucas-Fusco, L. M., Kilian, J. D., & Foley, T. E. (2007). Place value and
mathematics for students with mild disabilities: Data and suggested practices. Learning Disabilities: A
Contemporary Journal, 5, 21-39.
Chall, J. (1987). Two vocabularies for reading: Recognition and meaning. In M. McKeown & M. Curtis
(Eds.). The nature of vocabulary acquisition (pp. 7-17). Hillsdale, NJ: Lawrence Erlbaum.
Christ, T. J. (2006). Short term estimates of growth using curriculum-based measurement of oral
reading fluency: Estimates of standard error of the slope to construct confidence intervals. School
Psychology Review, 35, 128-133.
Christ, T. J., & Ardoin, S. P. (2009). Curriculum-based measurement of oral reading: Passage equivalence
and probe-set development. Journal of School Psychology, 47, 55-75.
Christ, T. J., & Boice, C. H. (2009). Rating scale items: A brief review of nomenclature, components and
formatting. Assessment for Effective Intervention, 34(4), 242-250.
Christensen, C. A., & Bowey, J. A. (2005). The efficacy of orthographic rime, grapheme–phoneme
correspondence, and implicit phonics approaches to teaching decoding skills. Scientific Studies of
Reading, 9(4), 327-349.
Clements, D., Sarama, J., & DiBiase, A. M. (Eds.). (2004). Engaging young children in mathematics:
Findings of the 2000 national conference on standards for preschool and Kindergarten mathematics
education. Mahwah, NJ: Erlbaum.
Compton, D. L., Appleton, A. C., & Hosp, M. K. (2004). Exploring the relationship between text-leveling
systems and reading accuracy and fluency in second grade students who are average and poor
decoders. Learning Disabilities Research & Practice, 19(3), 176-184.
Clay, M. (1972). Reading, the Patterning of Complex Behavior. Auckland, New Zealand: Heinemann.
Crocker, L. M., & Algina, J. (1986). Introduction to classical and modern test theory (Vol. 6277). New
York: Holt, Rinehart and Winston.
Deacon, S. H., Parrila, R., & Kirby, J. R. (2008). A review of the evidence on morphological processing in
dyslexics and poor readers: A strength or weakness. The Sage handbook of dyslexia, 212-237.
Deno, S. L. (1985). Curriculum-Based measurement: The emerging alternative. Exceptional Children,
52, 219-232.
Deno, S. L. (1986). Formative evaluation of individual student programs: A new role for school
psychologists. School Psychology Review, 1(3), 358-374.
Deno, S. L. (2002). Problem solving as best practices. In A. Thomas & J. Grimes (Eds.), Best practices in
school psychology IV. (pp. 37-56). Bethesda, MD: National Association of School Psychologists.
Page | 129
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Deno, S. L. (2003). Developments in Curriculum-Based measurement. The Journal of Special Education,
37(3), 184-192.
Deno, S. L. (2005). Problem solving assessment. In R. Brown-Chidsey (Ed.), Assessment for intervention:
A problem-solving approach (pp. 10-42). New York: Guilford Press.
Deno, S. L. & Mirkin, P. K. (1977) Data-based Program Modification: A Manual.
Reston VA: Council for Exceptional Children
De Ayala, R. J. (2009). The theory and practice of item response theory. Guilford Press.
Dickinson, D., McCabe, A., & Essex, M. (2006). A window of opportunity we must open to all: The case
for high-quality support for language and literacy. In D. K. Dickinson & S. B. Neuman (Eds.), Handbook
of early literacy research (pp. 11-28). New York: Guilford Press.
Dominguez, X., Vitiello, V.E., Maier, M.F., & Greefield, D.B. (2010). Longitudinal examination of young
children’s behavior: Child-level and classroom-level predictors of change throughout the preschool
year. School Psychology Review, 39, 29-47.
Downing, J., Ollila, L., & Oliver, P. (1975). Cultural differences in children's concepts of reading and
writing. British Journal of Educational Psychology, 45, 312-316.
Dunn, J. & Brown, J. (1994). Affect expression in the family, children's understanding of emotions, and
their interactions with others. Merrill-Palmer Quarterly, 40, pp. 120–137
Durkin, D. (1993). Teaching them to read (6th ed.). Boston, MA: Allyn and Bacon.
Dynamic Indicators of Basic Early Literacy Skills (DIBELS Next). Eugene, OR: Institute for the
Development of Educational Achievement. Available: https://dibels.org/next/index.php.
Eckoff, A. & Urback, J. (2008). Understanding imaginative thinking during childhood: Sociocultural
conceptions of creativity and imaginative thought. Early Childhood Education Journal, 36, 179-185.
Eisenberg, N., Shepard, S. A., Fabes, R. A., Murphy, B. C, & Guthrie, I. K. (1998). Shyness and children's
emotionality, regulation, and coping: Contemporaneous, longitudinal, and across-context relations.
Child Development, 69, 767-790.
Eisner, E. W. (2002). The arts and the creation of mind. New Haven, CT: Yale University Press.
Ekeland, E., Heian, F. & Hagen, K.B. (2004). Can exercise improve self-esteem in children and young
people? A systematic review of randomized control trials. British Journal of Sports Medicine, 39, 792798.
Flesch, R. F. (1949). A new readability yardstick. Journal of Applied Psychology, 32(3), 221-233.

Page | 130
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Francis, D. J., Santi, K. L., Barr, C., Fletcher, J. M., Varisco, A., & Foorman, B. R. (2008). Form effects on the
estimation of students’ oral reading fluency using DIBELS. Journal of School Psychology, 46, 315-342.
Fry, E. B., & Kress, J. E. (2006). The reading teacher's book of lists. San Francisco: Jossey-Bass.
Fuchs, D., & Fuchs, L. S. (2006). Introduction to Response to Intervention: What, why, and how valid is
it? Reading Research Quarterly, 41, 93-99.
Fuchs, L. S., & Fuchs, D. (2002) Curriculum-Based Measurement: Describing competence, enhancing
outcomes, evaluating treatment effects, and identifying treatment non-responders. Peabody Journal
of Education, 77:2, 64-84, doi: 10.1207/S15327930PJE7702_6
Fuchs, D., Fuchs, L. S., & Compton, D. L. (2004). Identifying reading disabilities by responsiveness-toinstruction: Specifying measures and criteria. Learning Disability Quarterly, 27(4), 216-227.
Fuchs, L. S., Fuchs, D., & Maxwell, L. (1988). The validity of informal reading comprehension
measures. Remedial and Special Education, 9(2), 20-28.
Gajdamaschko, N. (2005). Vygotsky on imagination: Why an understanding of the imagination is an
important issue for schoolteachers. Teaching Education, 16, 13–22
Geary, D.C. (1999). Mathematical Disabilities: What we know and don’t know. Retrieved
August 23, 2011 from the LD Online Web site: http://www.ldonline.org/article/5881.
Geary, D.C. (2004). Mathematics and learning disabilities. Journal of Learning Disabilities, 37,
(1), 4-15.
Gendron, M., Royer, E., Bertrand, R. & Potrin, P. (2004). Behavior disorders, social competence, and the
practice of physical activities among adolescents. Emotional and Behavioral Difficulties, 9, 249-259.
Gersten, R., Clarke, B., Jordan, N. C., Newman-Gonchar, R., Haymond, K., & Wilkins, C. (2012). Universal
screening in mathematics for the primary grades: Beginnings of a research base. Exceptional Children,
78, 423-445.
Gersten, R., Jordan, N. C., & Flojo, J. R. (2005). Early identification and interventions for students with
mathematics difficulties. Journal of learning disabilities, 38(4), 293-304.
Goswami, U. (2000). Phonological and lexical processes. In M. J. Kamil, P. B. Mosenthal, P. D. Pearson, &
R. Barr (Eds.). Handbook of reading research, Volume III (pp. 251-267) Mahwah, NJ: Lawrence Earlbaum.
Goswami, U. (2001). Early phonological development and the acquisition of literacy. In S. B. Neuman &
D. Dickinson (Eds.), Handbook of Early Literacy Research (pp. 111–125). New York: Guilford Press.

Page | 131
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading, and reading disability. Remedial and special
education, 7(1), 6-10.
Goldman, S. R., & Varnhagen, C. K. (1986). Memory for embedded and sequential episodes in stories.
Journal of Memory and Language, 25, 401-418.
Group Reading Assessment and Diagnostic Evaluation (GRADE). Pearson Education Inc.
Graesser, A. C., Leon, J. A., & Otero, J. (2002). Introduction to the psychology of science text
comprehension. In J. Otero, J. A. Leon, & A. C. Graesser (Eds.), The Psychology of Science Text
Comprehension (pp. 1-15). Mahwah, NJ: Erlbaum.
Graesser, A. C., McNamara, D. S., & Louwerse, M. M (2003). What do readers need to learn in order to
process coherence relations in narrative and expository text? In A.P. Sweet and C.E. Snow (Eds.),
Rethinking reading comprehension (pp. 82–98). New York: Guilford Publications.
Graesser, A. C., McNamara, D. S., Lowerse, M. M., & Cai, Z. (2004). Coh-Metrix: Analysis of text on
cohesion and language. Behavior Research Methods, Instruments, & Computers, 36(2), 193-202
Graesser, A.C., Olde, B. A., & Klettke, B. (2002). How does the mind construct and represent stories? In
M. Green, J. Strange, and T. Brock (Eds.), Narrative Impact: Social and Cognitive Foundations. Mahwah:
NJ: Erlbaum.
Graesser, A. C., Singer, M., & Trabasso, T. (1994). Constructing inferences during narrative text
comprehension. Psychological Review, 101(3), 371-395.
Grzybowski, M., & Younger, J. G. (1997). Statistical methodology: III. Receiver operating characteristic
(ROC) curves. Academic Emergency Medicine, 4, 818-826.
Griffin, S. (2008). Early intervention for children at risk of developing mathematical learning difficulties.
In D. Berch & M. Mazzocco (Eds.) Why is math so hard for some children? The nature and origins of
mathematical learning difficulties and disabilities. (pp. 373 – 395) Baltimore, MD: Paul H. Brookes.
Guyer, R., & Thompson, N.A. (2012). User’s Manual for Xcalibre item response theory calibration
software, version 4.1.6. St. Paul MN: Assessment Systems Corporation. Available from
http://www.assess.com/
Haladyna, T. M., Downing, S. M., & Rodriguez, M. C. (2002). A review of multiple-choice item-writing
guidelines for classroom assessment. Applied measurement in education, 15, 309-333.
Haladyna, T.M. (2007). Roles and importance of validity studies in test development (pp. 739-760). In
S.M. Downing and T.M. Haladyna (Eds.) Handbook of test development. Mahwah, NJ: Lawrence
Erlbaum Associates.

Page | 132
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Hardy, M., Stennett, R., & Smythe, P. (1974). Development of auditory and visual language concepts
and relationship to instructional strategies in Kindergarten. Elementary English Journal, 51, 525-532.
Harlin, R., & Lipa, S. (1990). Emergent literacy: A comparison of formal and informal assessment
methods. Reading Horizons, 20, 209-223.
Henard, D.H. (2000). Item response theory. In L. Grimm & P. Yarnold (Eds.), Reading and understanding
more multivariate statistics (pp. 67-97). Washington, DC: American Psychological Association.
Hiebert, E. H., & Taylor, B.M. (2000). Beginning reading instruction: Research on early interventions. In
M. J. Kamil, P. B. Mosenthal, P. D. Pearson, & R. Barr (Eds.). Handbook of reading research, Volume III
(pp. 455-482) Mahwah, NJ: Lawrence Earlbaum.
Hiebert, E. H., & Fisher, C.W. (2007). The critical word factor in texts for beginning readers. Journal of
Educational Research. 101(1), 3-11.
Hintze, J. M., & Christ, T. J. (2004). An examination of variability as a function of passage variance in
CBM progress monitoring. School Psychology Review, 33, 204-217.
Hintze, J. M., & Silberglitt, B. (2005). A longitudinal examination of the diagnostic accuracy and
predictive validity of R-CBM and high-stakes testing. School Psychology Review, 34, 372-386.
Hoffman, K. I. (1993). The USMLE, the NBME subject examinations, and assessment of individual
academic achievement. Academic Medicine, 68(10), 740-7.
Hosp, M. K., Hosp, J. L., & Howell, K. W. (2007). The ABC‟s of CBM: A practical guide to Curriculum-Based
measurement. New York, NY: The Guilford Press.
Hu, L.T., & Bentler, P.M. (1999). Cutoff criteria for fit indices in covariance structure analysis:
Conventional criteria versus new alternatives. STRUCTURAL EQUATION MODELING. 6, 1-55.
Jenkins, D. (2001). Impact of the implementation of the teaching/learning cycle on teacher decisionmaking and emergent readers. Reading Psychology, 22(4), 267-288.
Jenkins, J. R., Hudson, R. F., & Johnson, E. S. (2007). Screening for at-risk readers in a Response-toIntervention (RTI) framework. School Psychology Review. 36(4).
Jenkins, J. R., Zumeta, R., Dupree, O., & Johnson, K. (2005). Measuring gains in reading ability with
passage reading fluency. Learning Disabilities Research & Practice, 20(4), 245-253.
Jenkins, J. R., Hudson, R. F., & Johnson, E. S. (2007). Screening for at-risk readers in a response to
intervention framework. School Psychology Review, 36, 582-600.
Jenkins, J. R., & Jewell, M. (1992). An examination of the concurrent validity of the Basic Academic Skills
Samples (BASS). Assessment for Effective Intervention, 17(4), 273-288.
Page | 133
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Johns, J. (1972). Children’s concepts of reading and their reading achievement. Journal of Reading
Behavior, 4, 56-57.
Johns, J. (1980). First Graders’ concepts about print. Reading Research Quarterly, 15, 529- 549.
Jordan, N. C., Kaplan, D., Ramineni, C., & Locuniak, M. N. (2009). Early math matters: Kindergarten
number competence and later mathematics outcomes. Developmental Psychology, 45, 850-867.
Juel, C. (1988). Learning to read and write: A longitudinal study of 54 children from first through fourth
grades. Journal of educational Psychology, 80, 437-447.
Juel, C. (2006). The impact of early school experience on initial reading. In D. K. Dickinson, & S. B.
Neuman (Eds.). The handbook of early literacy research (v.2), (pp. 410-426). NY: Guilford Press.
Juel, C., & Minden‐Cupp, C. (2000). Learning to read words: Linguistic units and instructional
strategies. Reading Research Quarterly, 35(4), 458-492.
Kane, M. T. (2013). Validating the interpretation and uses of test scores. Journal of Educational
Measurement, 50(1), 1-73.
Katz, L., & Frost, R. (1992). The reading process is different for different orthographies: The
orthographic depth hypothesis. Advances in Psychology-Amsterdam, 94, 67.
Kendeou, P., van den Broek, P., White, M., & Lynch, J. (2007). Preschool and early elementary
comprehension: Skill development and strategy interventions. In D. S. McNamara (Ed.) Reading
comprehension strategies: Theories, interventions, and technologies, (pp. 27–45). Mahwah, NJ:
Erlbaum.

Kilgus, S.P., Chafouleeas, S.M., & Riley-Tillman, T.C. (2013). Development and initial validation of the
social and academic behavior risk screener for elementary grades. School Psychology Quarterly, 28,
210-226.
Kilgus, S. P., Sims, W., von der Embse, N. P., & Riley-Tillman, T. C. (under review). Confirmation of
models for interpretation and use of the Social and Academic Behavior Risk Screener (SABRS). School
Psychology Quarterly.
Kilgus, S. P., Eklund, K., von der Embse, N. P., & Taylor, C. (2014). Diagnostic accuracy of the Social,
Academic, and Emotional Behavior Risk Screener (SAEBRS) in elementary and middle grades.
Manuscript in preparation.
Kilpatrick, J., Swafford, J., & Findell, B. (Eds.). (2001). Adding it up. Washington, DC: National Academy
Press.

Page | 134
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Kim-Kang, G., & Weiss, D. J. (2007). Comparison of computerized adaptive testing and classical
methods for measuring individual change. In Proceedings of the 2007 GMAC Conference on
Computerized Adaptive Testing. Available from www. psych. umn.edu/psylabs/CATCentral.
Kim-Kang, G., & Weiss, D. J. (2008). Adaptive measurement of individual change. Zeitschrift für
Psychologie/Journal of Psychology, 216, 49-58.
Kingsbury, G. G., & Houser, R. L. (1999). Developing computerized adaptive tests for school children. In
F. Drasgow & J. B. Olson Buchanan (Eds.), Innovations in computerized assessment (pp. 93–115).
Mahwah, NJ: Erlbaum.
Kingston, N., & Nash, B. (2011). Formative assessment: A meta‐analysis and a call for research.
Educational Measurement: Issues and Practice, 30(4), 28-37.
Kintsch, W. (1998). Comprehension: A Paradigm for Cognition. Boulder, CO: Cambridge University
Press.
Kintsch, W. & van Dijk, T. A. (1978). Toward a model of text comprehension and production.
Psychological Review, 35(5), 363-394.
Kirkcaldy, B.D., Shephard, R.J. & Siefen, R.G. (2002). The relationship between physical activity and selfimage and problem behavior among adolescents. Social Psychiatry and Psychiatric Epidemiology, 37,
544-550.
Klare, G. R. (1974-1975). Assessing readability. Reading Research Quarterly, 10(1), 61-102.
Kolen, M. J., & Brennan, R. L. (2004). Test equating, scaling, and linking: Methods and practices (2nd
ed.). New York: Springer-Verlag.
Kranzler, J.H., Brownell, M.T., & Miller, M.D. (1998). The construct validity of curriculum-based
measurement of reading: An empirical test of a plausible rival hypothesis. Journal of School
Psychology, 36 (4), 399-415.
Kuhl, J. & Kraska, K. (1989). Self-regulation and metamotivation: Computational mechanisms,
development, and assessment. In R. Kanfer, P. Ackerman, R. Cudeck (Eds.), Abilities, motivation, and
methodology: The Minnesota Symposium on learning and individual differences, Lawrence Erlbaum
Associates, Hillsdale, NJ, pp. 373–374.
LaBerge, D., & Samuels, S. J. (1974). Toward a theory of automatic information processing in reading.
Cognitive psychology, 6, 293-323.
Lee, J., Grigg, W., 8c Donahue, P. (2007). The nation's report card: Reading 2007 (NCES 2007-496).
Washington, DC: National Center for Education Statistics, U.S. Department of Education.

Page | 135
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Levy, B. A., Gong, Z., Hessels, S., Evans, M. A., & Jared, D. (2006). Understanding print: Early reading
development and the contributions of home literacy experiences. Journal of Experimental Child
Psychology, 93(1), 63-93.
Liberman, A. M., Cooper, F. S., Shankweiler, D. P., & Studdert-Kennedy, M. (1967). Perception of the
speech code. Psychological review, 74(6), 431.
Lipson, M.L., Mosenthal, J. H., Mekkelsen, J., & Russ, B. (2004). Building knowledge and fashioning
success one school at a time. The Reading Teacher, 57 (6) 534-542.
Logan, G. D. (1997). Automaticity and reading: Perspectives from the instance theory of
automatization. Reading and Writing Quarterly, 13, 123–146.
Lomax, R. G., & McGee, L.M. (1987). Young children’s concepts about print and reading: Toward a
model of word reading acquisition. Reading Research Quarterly, 22, 237-256.
MacGinitie, W., MacGinitie, R., Maria, K., & Dreyer, L.G. (2000). Gates-MacGinitie Reading Tests. 4th ed.
Itasca, IL: Riverside Publishing Company.
Mandler, J. M. & Johnson, N. S. (1977). Remembrance of things parsed: Story structure and recall.
Cognitive Psychology, 9(1), 111-151.
Markell, M. A., & Deno, S. L. (1997). Effects of increasing oral reading generalization across reading
tasks. The Journal of Special Education, 31(2), 233-250.
Martone, A., & Sireci, S. G. (2009). Evaluating alignment between curriculum, assessment and
instruction. Review of Educational Research, 79(3), 1-76.
Mathes, P. G., Denton, C. A., Fletcher, J. M., Anthony, J. L., Francis, D. J., & Schatschneider, C. (2005). The
Effects of Theoretically Different Instruction and Student Characteristics on the Skills of Struggling
Readers. Reading Research Quarterly, 148-182.
Mazzeo D, Arens S, Germeroth C, Hein H. (2012). Stopping childhood obesity before it begins. Phi
Delta Kappan, 93(7):10-15.
Mazzocco, M. M. M., & Thompson, R. E. (2005). Kindergarten predictors of math learning disability.
Learning Disabilities Research and Practice, 20, 142 – 155.
McGregor, K. K. (2004). Developmental dependencies between lexical semantics and
reading. Handbook of language and literacy, 302-317.
Measures of Academic Progress (MAP). Northwest Evaluation Association.
Messick, S. (1993). Validity. (In R. L. Linn (Ed), Educational measurement (2nd ed. pp. 13—104).
Phoenix: American Council on Education and Oryx Press.)
Page | 136
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Methe, S. A., Hojnoski, R., Clarke, B., Owens, B. B., Lilley, P. K., Politylo, B. C., White, K. M., & Marcotte, A.
M. (2011). Innovations and future directions for early numeracy curriculum-based measurement:
Commentary on the special series. Assessment for Effective Intervention, 36, 367 - 385.
Miller, K. (2004, October). Developing number names: A cross cultural analysis. Presentation at the
Early Childhood Academy, University of Michigan, Ann Arbor.
Morsy, L., Kieffer, M., & Snow, C. (2010). Measure for Measure: A Critical Consumers' Guide to Reading
Comprehension Assessments for Adolescents. Final Report from Carnegie Corporation of New York's
Council on Advancing Adolescent Literacy. Carnegie Corporation of New York.
Morgan, P. L., Farkas, G., & Wu, Q. (2009). Five-year growth trajectories of Kindergarten children with
learning difficulties in mathematics. Journal of Learning Disabilities, 42, 306 321.
Muter, V., Hulme, C., Snowling, M. J., & Stevenson, J. (2004). Phonemes, rimes, vocabulary, and
grammatical skills as foundations of early reading development: evidence from a longitudinal
study. Developmental psychology, 40(5), 665.
Nagy, W. E. (1988). Teaching Vocabulary to improve reading comprehension. National Council of
Teachers of English, 1111 Kenyon Rd., Urbana, IL 61801 (Stock No. 52384-015, $4.95 member, $7.50
nonmember--ISBN-0-8141-5238-4); International Reading Association, PO Box 8139, 800 Barksdale Rd.,
Newark, DE 19714-8139 (No. 151, $4.95 member, $7.50 nonmember--ISBN-0-87207-151-0).
National Center for Education Statistics (2013). The Nation’s Report Card: A First Look: 2013
Mathematics and Reading (NCES 2014-451).Institute of Education Sciences, U.S. Department of
Education, Washington, D.C.
National Center for Education Statistics (2013). The Nation’s Report Card: Trends in Academic Progress
2012 (NCES 2013 456). Institute of Education Sciences, U.S. Department of Education, Washington, D.C.
National Governors Association Center for Best Practices & Council of Chief State School Officers.
(2010). Common Core State Standards for English Language Arts & Literacy in History/Social Studies,
Science, & Technical Subjects. Washington, DC: National Governors Association Center for Best
Practices, Council of Chief State Officers.
National Center for Response to Intervention (2010). Screening Tools Chart. U.S. Office of Special
Education Programs. Retrieved from
http://www.rti4success.org/sites/default/files/Screening%20Tools%20Chart.pdf
National Governors Association Center for Best Practices & Council of Chief State School Officers.
(2010). Common Core State Standards for Mathematics. Washington, DC: Authors.
National Research Council. (2009). Mathematics learning in early childhood: Paths toward excellence
and equity. Washington, DC: Author.

Page | 137
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Nation, K., Clarke, P., Marshall, C. M., & Durand, M. (2004). Hidden language impairments in children:
Parallels between poor reading comprehension and specific language impairment? Journal of Speech,
Language & Hearing Research, 47(1).
Nation, K., & Snowling, M. J. (2004). Beyond phonological skills: Broader language skills contribute to
the development of reading. Journal of research in reading, 27(4), 342-356.
National Mathematics Advisory Panel [NMAP], (2008) Foundations for Success: The Final Report of the
National Mathematics Advisory Panel (No. ED04CO0015/0006): U.S. Department of Education:
Washington, D.C.
National Reading Panel (US), National Institute of Child Health, & Human Development (US). (2000a).
Report of the National Reading Panel: Teaching children to read: an evidence-based assessment of the
scientific research literature on reading and its implications for reading instruction: reports of the
subgroups. National Institute of Child Health and Human Development, National Institutes of Health.
National Reading Panel (US), National Institute of Child Health, & Human Development (US). (2000b).
Teaching children to read: An evidence-based assessment of the scientific research literature on
reading and its implications for reading instruction. National Institute of Child Health and Human
Development, National Institutes of Health.
NCTM. (2000). Principles and standards for school mathematics. Reston, VA: National Council for
Teachers of Mathematics.
Neuenschwander, R., Rothlisberger, M, Cimeli, P., Roebers, C.M. (2012). How do different aspects of
self-regulation predict successful adaptation to school? Journal of Experimental Child Psychology,
113, 353-371.
Neuman, S.B. & Roskos, K. (2005). The state of pre-Kindergarten standards. Early Childhood Research
Quarterly, 20, 125-145.
Nichols, W. D., Rickelman, R.J., & Rupley, W. H. (2004). Examining phonemic awareness and concepts of
print patterns of Kindergarten students. Reading Research and Instruction. 43, 56-82.
McNamara, D. S., & Kintsch, W. (1996). Learning from texts: Effects of prior knowledge and text
coherence. Discourse processes, 22(3), 247-288.
Northwest Evaluation Association (NWEA) (2011). Measures of Academic Progress (MAP). Portland, OR.
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York: McGraw-Hill.
Nydick, S. W., & Weiss, D. J. (2009). A hybrid simulation procedure for the development of CATs. In
Proceedings of the 2009 GMAC Conference on Computerized Adaptive Testing. Retrieved from www.
psych.umn.edu/psylabs/CATCentral.

Page | 138
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Oakhill, J., & Cain, K. (2007). Introduction to comprehension development. In K. Cain & J. Oakhill (Eds.),
Children’s comprehension problems in oral and written language: A cognitive perspective (pp. 3–40).
New York: Guilford Press.
Oakhill, J. V., Cain, K., & Bryant, P. E. (2003). The dissociation of word reading and text comprehension:
Evidence from component skills. Language and cognitive processes, 18, 443-468.
Okamoto, T. (1996, July). On relationships between statistical zero-knowledge proofs. In Proceedings
of the twenty-eighth annual ACM symposium on Theory of computing (pp. 649-658). ACM.
Paris, S. G. (2005). Reinterpreting the development of reading skills. Reading Research Quarterly, 40(2),
184-202.
Perfetti, C. A. (1994). Psycholinguistics and reading ability. In M. Gernsbacher (Eds.), Handbook of
psycholinguistics (pp. 849-894). San Diego, CA: Academic Press.
Perfetti, C. A. (1992). The representation problem in reading acquisition.
Porges, S. (2003). The polyvagal theory: Phylogenetic contributions to social behavior. Physiology and
Behavior, 79, 503–513.
Phillips, B. M., & Torgesen, J. K. (2006). Phonemic awareness and reading: Beyond the growth of initial
reading accuracy. Handbook of early literacy research, 2, 101-112.
Pratt, K., Martin, M., White, M.J., & Christ, T.J. (2010). Development of a FAIP-R First Grade Probe Set
(Technical Report No. 3). Minneapolis, MN: University of Minnesota, Department of Educational
Psychology.
Purpura, D. J., & Lonigan, C. J. (2013). Informal numeracy skills: The structure and relations among
numbering, relations, and arithmetic operations in preschool. American Educational Research Journal,
50, 178-209.
RAND Reading Study Group (2002). Reading for understanding: Toward an R&D program in reading
comprehension. Santa Monica, CA: RAND.
Raver, C. C. & Knitzer, J. (2002). Ready to enter: What research tells policymakers about strategies to
promote social and emotional school readiness among three- and four-year-old children. Promoting
the Emotional Well-being of Children and Families. Policy Paper No. 3. Columbia, New York: National
Center for Children in Poverty.
Rayner, K., Pollatsek, A., Ashby, J., & Clifton, C. Jr. (2012). Psychology of reading (2nd ed.). New York:
Psychology Press.
Raudenbush, S. W. and Anthony S. Bryk (2002). Hierarchical linear models: Applications and data
analysis methods, Second Edition. Newbury Park, CA: Sage
Page | 139
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Reschly, A. L., Busch, T.W., Betts, J., Deno, S. L., and Long, J.D. (2009). Curriculum-Based measurement
oral reading as an indicator of reading achievement: A meta-analysis of the correlational evidence.
Journal of School Psychology, 47, 427-469.
Ricketts, J., Nation, K., & Bishop, D. V. (2007). Vocabulary is important for some, but not all reading
skills. Scientific Studies of Reading, 11(3), 235-257.
Riddle Buly, M., & Valencia, S. W. (2002). Below the bar: Profiles of students who fail state reading tests.
Educational Evaluation and Policy Analysis, 24, 219-239.
Renaissance Learning. (1998b). STAR Math. Wisconsin Rapids, WI: Renaissance Learning.
Rubin, D. C. (1995). Memory in oral traditions: The cognitive psychology of epic, ballads, and countingout-rhymes. New York: Oxford University Press.
Rydell, A., Berlin, L., & Bohlin, G. (2003). Emotionality, emotion regulation, and adaptation among 5- to
8-year-old children. Emotion, 3, 30–47.
Samejima, F. (1994). Some critical observations of the test information function as a measure of local
accuracy in ability estimation. Psychometrika, 59(3), 307-329.
Samuels, S. J. (2007). The DIBELS tests: Is speed of barking at print what we mean by reading fluency?
Reading Research Quarterly, 42, 563–566.
Sadler, D. (1989). Formative assessment and the design of instructional systems. Instructional Science,
18(2), 119-144.
Scarborough, H. (2001). Connecting early language and literacy to later reading (dis)abilities: Evidence,
theory, and practice. In S. B. Neuman & D. Dickinson (Eds.), Handbook of early literacy research (pp. 97–
110). New York: Guilford Press.
Schank, Roger C., and Robert P. Abelson. 1977. Scripts, plans, goals, and understanding: An inquiry into
human knowledge structures. Hillsdale, NJ: Lawrence Erlbaum.
Schmeiser, C. B., & Welch, C. J. (2006). Test development. In R. L. Brennan (Ed.), Educational
Measurement (4th ed., pp. 623-646). Westport, CT: American Council on Education and Praeger
Publishers.
Severson, H. H., Walker, H. M., Hope-Doolittle, J., Kratochwill, T. R., & Gresham, F. M. (2007). Proactive,
early screening to detect behaviorally at-risk students: Issues, approaches, emerging innovations, and
professional practices. Journal of School Psychology, 45, 193-223.
Shinn, M. R. (Ed.). (1989). Curriculum-based measurement: Assessing special children. New York:
Guilford Press.

Page | 140
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Skiba, R. & Peterson, R. (2000). School discipline at a crossroads: From zero tolerance to early response.
Exceptional Children, 32, 200-216.
Snow, C. E., Burns, M. S., & Griffin, P. (Eds.). (1998). Preventing reading difficulties in young children.
National Academies Press.
Stahl, S. A. (2001). Teaching phonics and phonological awareness. Handbook of early literacy
research, 1, 333-347.
Snow, C. (1991). The theoretical basis for relationships between language and literacy in development.
Journal of Research in Childhood Education, 6, 5–10.
Snow, C. E., Burns, M. S., & Griffin, P. (1998). Preventing reading difficulties in young children. National
Academies Press.
Stanovich, K. E. (1981). Attentional and automatic context effects in reading. In A. M. Lesgold & C. A.
Perfetti (Eds.), Interactive processes in reading. Hillsdale, N.J.: Erlbaum.
Stanovich, K.E. (1984). The interactive-compensatory model of reading: A confluence of
developmental, experimental, and educational psychology. Remedial and Special Education, 5, 11-19.
Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of individual differences in the
acquisition of literacy. Reading Research Quarterly, 360-407.
Stanovich, K. E. (1990). A call for an end to the paradigm wars in reading research. Journal of Reading
Behavior, 22, 221-231.
Stanovich, K. E., & West, R. F. (1989). Exposure to print and orthographic processing. Reading Research
Quarterly, 24(4), 402-433.
Stein, N. L. & Glenn, C. G. (1975). An Analysis of Story Comprehension in Elementary School Children: A
Test of a Schema (Report No. PS 008 544). Washington University. (ERIC Document Reproduction
Service No. ED121474).
Stein, N. L. & Trabasso, T. (1982). What's in a story: An approach to comprehension and instruction. In
R. Glaser (Ed.), Advances in the psychology of instruction (Vol. 2). Hillsdale, N.J.: Erlbaum.
Storch, S. A., & Whitehurst, G. J. (2002). Oral language and code-related precursors to reading:
evidence from a longitudinal structural model. Developmental psychology, 38, 934-947.
Sugai, G., Horner, R., & Gresham, F. (2002). Behaviorally effective school environments. In Shinn, M.R.,
Walker, H.M.,& Stoner, G. (Eds.), Interventions for academic and behavior problems II: Preventive and
remedial approaches (pp. 315–350). Bethesda, MD: National Association of School Psychologists.

Page | 141
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Sugden, D.A. (1986). The development of proprioceptive control. Themes in Motor Development, 35,
21-39.
Swets, J. A., R. M. Dawes, & J. Monahan (2000), “Psychological Science Can Improve Diagnostic
Decisions”, Psychological Science in the Public Interest 1: 1–26. Test of Silent Reading Efficiency and
Comprehension (TOSREC).
Taylor, B. M., Pearson, P., Clark, K., & Walpole, S. (2000). Effective schools and accomplished teachers:
Lessons about primary-grade reading instruction in low-income schools. The Elementary School
Journal, 101(2), 121-165.
Taylor, B. M., Pearson, P., Peterson, D. S., & Rodriguez, M. C. (2003). Reading growth in high-poverty
classrooms: The influence of teacher practices that encourage cognitive engagement in literacy
learning. The Elementary School Journal, 104(1), 3-28.
Taylor, B. M., Pearson, P., Peterson, D. S., & Rodriguez, M. C. (2005). The CIERA School Change
Framework: An evidence-based approach to professional development and school reading
improvement. Reading Research Quarterly, 40(1), 40-69.
Thompson, S. J., Johnstone, C. J., Thurlow, M. L., & Clapper, A. T. (2004). State literacy standards,
practices, and testing: Exploring accessibility (Technical Report 38). Minneapolis, MN: University of
Minnesota, National Center on Educational Outcomes.
Trabasso, T. & Nickels, M. (1992). The development of goal plans of action in the narration of a picture
story. Discourse Processes, 15, 249-275.
Trabasso, T. & Stein, N. L. (1997). Narrating, representing, and remembering event sequences. In P. W.
van den Broek, P. J. Bauer, & T. Bourg (Eds.), Developmental spans in event comprehension and
representation: Bridging fictional and actual events (pp. 237-270). Mahwah, NJ: Erlbaum.
Tumner, W. E., Herriman, M. L., & Nesdale, A. R. (1988). Metalinguistic abilities and beginning Reading.
Reading Research Quarterly, 23, 134-158.
Van de Walle, J. A., Karp, K. S. & Bay-Williams, J. M. (2013). Elementary and middle school mathematics:
Teaching developmentally (8th ed.). New York, NY: Pearson.
van den Broek, P. & Trabasso, T. (1986). Causal networks versus goal hierarchies in summarizing text.
Discourse Processes, 9, 1-15.
van den Broek, P. (1994). Comprehension and memory of narrative texts: Inferences and coherence. In
M. A. Gernsbacher (Ed.) Handbook of Psycholinguistics (pp. 539-588). New York: Academic Press.
VanDerHeyden, A. M. (2011). Technical adequacy of response to intervention decisions. Exceptional
Children, 77, 335-350.

Page | 142
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
VanLoy, W. J. (1996). A comparison of adaptive self-referenced testing and classical approaches to the
measurement of individual change (Doctoral dissertation, University of Minnesota).
Vellutino, F. R. (2003). Individual differences as sources of variability in reading comprehension in
elementary school children. In A. P. Sweet & C. E. Snow (Eds.), Rethinking Reading Comprehension (pp.
5-81). New York: Guilford Press.
Vellutino, F. R., & Scanlon, D. M. (1987). Phonological coding, phonological awareness, and reading
ability: Evidence from a longitudinal and experimental study. Merrill-Palmer Quarterly: Journal of
Developmental Psychology, 33, 321-363.
Vellutino, F. R., & Scanlon, D. M. (1991). The preeminence of phonologically based skills in learning to
read. In S. Brady & D. Shankweiler (Eds.), Phonological processes in literacy: A tribute to Isabelle
Liberman (pp. 237–252). Hillsdale, NJ: Erlbaum.
Vellutino, F. R., Scanlon, D. M., Small, S. G., & Tanzman, M. S. (1991). The linguistic basis of reading
ability: Converting written to oral language. Text, 11, 99–133.
Wayman, J., Cho, V. & Johnston, M. (2007). The Data-Informed District: A District-Wide Evaluation of
Data Use in the Natrona County School District. Austin: The University of Texas. Retrieved 3/12/08 from
http://edadmin.edb.utexas.edu/datause/
Weiss, D. J. (2004). Computerized Adaptive Testing for Effective and Efficient Measurement in
Counseling and Education. Measurement & Evaluation in Counseling & Development (American
Counseling Association), 37(2).
Weiss, D. J., & Kingsbury, G. (1984). Application of computerized adaptive testing to educational
problems. Journal of Educational Measurement, 21(4), 361-375.
Weiss, D. J., & Kingsbury, G. (2005). Application of computerized adaptive testing to educational
problems. Journal of Educational Measurement, 21(4), 361-375.
Westberg, K. L. (1993). An Observational Study of Instructional and Curricular Practices Used with
Gifted and Talented Students in Regular Classrooms. Research Monograph 93104.
Westberg, K. L., & Daoust, M. E. (2003). The results of the replication of the classroom practices survey
replication in two states. The National Research Center on the Gifted and Talented Newsletter, 3-8.
Whitehurst, G. J., & Lonigan, C. J. (1998). Child development and emergent literacy. Child
development, 69(3), 848-872.
Williams, K.T. (2001). Group Reading Assessment and Diagnostic Evaluation (GRADE).Circle
Pines, MN: AGS Publishing.

Page | 143
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

References
Williams, K.T. (2004). Group Mathematics Assessment and Diagnostic Evaluation (GMADE). Circle Pines,
MN: AGS Publishing.
Williford, A.P., Maier, M.F., Downer, J.T., Pianta, R.C., & Howes, C. (2013). Understanding how children’s
engagement and teacher’s interactions combine to predict school readiness. Journal of Applied
Developmental Psychology, 34, 299-309.
Zeno, S.M., Ivens, S.H., Millard, R.T., Duvvuri, R. (1995). The educator’s word frequency guide. Brewster,
NY: Touchstone Applied Science Associates.
Zickar, M. J., Overton, R. C., Taylor, L. R., & Harms, H. J. (1999). The development of a computerized
selection system for computer programmers in a financial services company. In F. Drasgow & J. B.
Olson-Buchanan (Eds.), Innovations in computerized assessment (pp. 7-33). Mahwah, NJ: Erlbaum.
Zwaan, R. A. & Rapp, D. N. (2008). Discourse comprehension. In M. J. Traxler & M. A. Gernsbacher (Eds.),
Handbook of Psycholinguistics, 2nd Edition (pp. 725-764). New York: Elsevier.

Page | 144
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Appendix A: Benchmarks and Norms Information

Appendix A: Benchmarks and Norms Information
Norms and benchmarks are reported in the FastBridge Learning Benchmarks and Norms Guide, 20152016.

Page | 145
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

Appendix C: Diagnostic Accuracy Supplement for Reading

Appendix B: FastBridge Learning Reading Diagnostic
Accuracy
Table 105. Summary of Diagnostic Accuracy AUC Statistics and Validity Evidence

Grade
K

Measure
earlyReading

1

earlyReading

2

aReading
CBMreading
aReading
CBMreading
aReading
CBMreading
aReading
CBMreading
aReading
CBMreading
aReading
aReading
aReading
aReading

3
4
5
6
7
8
9
10

Subtest
Concepts of Print
Onset Sounds
Letter Names
Letter Sounds
Composite
Letter Sounds
Word Rhyming
Word Segmenting
Nonsense Words
Sight Words
Sentence Reading
Composite
—
—
—
—
—
—
—
—
—
—
—
—
—
—

Note. F = Fall; W = Winter; S = Spring

Page | 146
www.fastbridge.org
© 2015 Theodore J. Christ and Colleagues, LLC. All rights reserved.

F to W
AUC
.80 -.81
.83 - .84
.78 - .82
.80 - .82
.82 - .84
.73 - .99
—
.87
.66 - .89
.69 - .92
.71 - .94
.85 - .89
—
—
—
—
—
—
—
—
—
—
—
—
—
—

W to S
AUC
—
—
—
—
.84
.76–.77
.71–.75
—
—
—
.82–.92
—
—
—
.83–.97
.76–.97
.77–.94
.78–.90
.71–.93
.71–.93
.77–.92
.81–.88
.91–.92
.92–.95
—
—

F to S
AUC
—
—
—
—
.73–.76
.75–.78
—
—
—
—
.72
—

.76–.90
.79–.97
.75–.92
.81–.89
.68–.93
.71–.91
.83–.89
.83–.88
.81–.85
.69–.85
.85–.87
.77–.82

Concurrent
Validity
(Spring)
pending
pending
pending
pending
pending
pending
pending
pending
pending
pending
pending
pending
.96–.97
.97 –.99
.78–1.00
.77 –1.00
.79–1.00
.71–1.00
.78–.96
.69–.97
.86–1.00
.81–1.00
pending
pending
pending
pending



---

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.5
Linearized                      : Yes
Encryption                      : Standard V4.4 (128-bit)
User Access                     : Print, Extract, Print high-res
Author                          : 
Create Date                     : 2015:12:05 15:00:47Z
Modify Date                     : 2019:06:17 13:52:30-07:00
Subject                         : 
XMP Toolkit                     : Adobe XMP Core 5.2-c001 63.139439, 2010/09/27-13:37:26
Creator Tool                    : Nitro Pro 9  (9. 5. 1. 5)
Metadata Date                   : 2019:06:17 13:52:30-07:00
Format                          : application/pdf
Creator                         : 
Title                           : 
Description                     : 
Keywords                        : 
Producer                        : Nitro Pro 9  (9. 5. 1. 5)
Document ID                     : uuid:bdb1fee1-9d55-44a5-84fa-f926b1c2df6e
Instance ID                     : uuid:0fc55691-7318-41fa-91ac-2b6d498e307a
Page Mode                       : UseNone
Page Count                      : 147

EXIF Metadata provided by EXIF.tools