Assessing the Accuracy of Elementary School Test Scores as Predictors of Students’ High School Outcomes

Oct 3, 2026

Upload by Fans

Testing students and using test results to hold schools and, in some cases, teachers accountable is a central school-improvement strategy, while tests also serve diagnostic roles such as identifying students at risk of dropping out of high school. Policy debate centers on whether testing is effective and whether scores represent meaningful schooling outcomes. This working paper uses panel data from North Carolina, Massachusetts, and Washington State to evaluate how accurately early test scores predict later high-school outcomes, including 10th grade achievement.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages76

Reading Time3 h 12 min

AccessFree to download

Contents

Abstract

1. Introduction

2. Literature on Predicting Long-Term Student Outcomes

3. Data Sources, Sample Inclusion, and Measures

4. Empirical Approach

4.1 Analytic Approach

4.2 Measuring the Accuracy of Out-of-Sample Predictions

5. Results

5.1 Baseline Findings on Long-Term Predictions

5.2 Out-of-Sample Prediction Accuracy

5.3 Assessing the Potential Implications of Sample Attrition

5.4 Comparison with Machine Learning Methods

6. Conclusion

References

Tables and Figures

Appendix A

Appendix B

|
NATIONAL
CENTER for ANALYSIS of LONGITUDINAL DATA in EDUCATION RESEARCH

TRACKING EVERY STUDENT’S LEARNING EVERY YEAR
A program of research by the American Institutes for Research with Duke University, Northwestern University, Stanford University, University of Missouri-Columbia, University of Texas at Dallas, and University of Washington | | --- | | | Assessing the Accuracy of Elementary School Test Scores as Predictors of Students ’ High School Outcomes Dan Goldhaber Malcolm Wolff Timothy Daly WORKING PAPER No . 23 5-0 5 20-2 • August 202 1 Assessing the Accuracy of Elementary School Test Scores as Predictors of Students’ High School Outcomes Dan Goldhaber American Institutes for Research/CALDER University of Washington Malcolm Wolff University of Washington Timothy Daly EdNavigator Contents ......................................................................................................................................................... i Acknowledgments......................................................................................................................................... ii Abstract ........................................................................................................................................................ iii 1. Introduction ........................................................................................................................................... 1 2. Literature on Predicting Long-Term Student Outcomes ....................................................................... 3 3. Data Sources, Sample Inclusion, and Measures.................................................................................... 6 4. Empirical Approach .............................................................................................................................. 9 4.1 Analytic Approach .............................................................................................................................. 9 4.2 Measuring the Accuracy of Out-of-Sample Predictions ................................................................... 11 5. Results ................................................................................................................................................. 15 5.1 Baseline Findings on Long-Term Predictions................................................................................... 15 5.2 Out-of-Sample Prediction Accuracy ................................................................................................. 18 5.3 Assessing the Potential Implications of Sample Attrition................................................................. 20 5.4 Comparison with Machine Learning Methods.................................................................................. 21 6. Conclusion .......................................................................................................................................... 22 References ................................................................................................................................................... 25 Tables and Figures ...................................................................................................................................... 31 Appendix A ................................................................................................................................................. 45 Appendix B ................................................................................................................................................. 67 Acknowledgments This work is supported by the Carnegie Foundation and the National Center for the Analysis of Longitudinal Data in Education Research (CALDER) . All opinions expressed in this paper are those of the authors and do not necessarily reflect the views of our funders or the institutions to whic

FAQs

What question does the paper address about test scores?

The paper evaluates how accurately early elementary school test scores predict later high school outcomes, including 10th grade achievement.

What data sources are used in the study?

The analysis uses panel data from three states: North Carolina, Massachusetts, and Washington State.

Why is the accuracy of test-based predictions politically significant?

Because there is policy debate over whether using tests for diagnostic and accountability purposes is effective and whether test results reflect meaningful outcomes for students.

Oct 3, 2026

Upload by Fans