Trends in U.S. Students’ Average
Math, Reading, and Science Scores on
NCES Large-Scale Assessments:
1998–2024
Activate Research, Inc.
June 12, 2025
Executive Summary
Introduction
This consolidated review of National Center for Education Statistics (NCES) student assessment data offers a presentation of trends in average test scores across multiple NCES studies. Specifically, the analyses investigate U.S. students’ average test scores in math, reading, and science between 1998 and 2024 to present robust findings about student performance on NCES-sponsored large-scale assessments¹ during the last quarter century, a period marked at the beginning by an increased focus on accountability ushered in by the passage of the No Child Left Behind Act (NCLB) in 2002 and at the end by the aftereffects of the COVID-19 pandemic.
​
The figures presented here use publicly available data from national and international assessments. Our analyses build on work conducted by NCES and its contracting partners.
Key Findings
1
For both grade 4 and grade 8 average math and reading scores, generally, we observe score increases from about 2000 through the first half of the 2010s. Scores largely hold steady or decrease in the second half of the 2010s, and most scores in the 2020s are either lower than or not measurably different from the scores from about 2000.
2
We see a similar pattern for grade 4 average science scores on the Trends in International Mathematics and Science Study (TIMSS): scores increase through the mid-2010s, then decrease, and ultimately end the period in 2023 not measurably different from the score in 2003. For TIMSS grade 8 average science scores, we do not find the same general pattern, as the scores from one assessment administration to the next are not measurably different between 2003 and 2023. It is not possible to compare these TIMSS results with those from National Assessment of Educational Progress (NAEP) science assessment data, as insufficient NAEP science data points exist to model a trend.
Implications for Research and Practice
These analyses raise a variety of questions for future investigation, such as whether average math and reading scores will level out as we move further away from the heightened federal focus on school accountability and from the decreases observed following the COVID-19 pandemic. They also highlight a critical issue: limited data points hinder accurate long-term analysis. Without the reconstitution of NCES, this challenge could extend beyond the issues noted with science in this report to all subject areas, jeopardizing the ability to track and research student outcomes nationwide.
NCES data are critical to our understanding of U.S. students’ academic performance and allow for comparisons both at home and around the world. While nearly all U.S. states are required to assess their students’ performance annually, the assessments used for this purpose are generally not comparable from state to state and tend to change frequently, making them ill-suited for comparative or longitudinal analyses. Thus, NAEP Long-Term Trend (NAEP-LTT) and NAEP serve as the key means by which researchers and policymakers can make cross-state and over-time comparisons. On the global front, NCES’s international assessments serve as the main sources of data for understanding how U.S. students measure up against their international peers. Without NCES, these analyses would not be possible.
The full report, Trends in U.S. Students' Average Math, Reading, and Science Scores on NCES Large-Scale Assessments: 1998-2024, is available for download.
Figures
Figure 1.​ Percentage change in average mathematics scale scores for National Assessment of Educational Progress Long-Term Trend (NAEP-LTT, age 9), NAEP (grade 4), and Trends in International Math and Science Study (TIMSS, grade 4), relative to earliest presented year: 1999 through 2024

NOTE: For each survey, each value is the percentage change between the respective scale score and the overall average scale score for the first data collection that is presented in the figure. The first year presented for NAEP-LTT is 1999; for NAEP, it is 2000; and for TIMSS, it is 2003. Not all apparent differences between estimates are statistically significant. For detailed information, including the results of statistical testing and findings on the individual assessments, please see the full report.
Figure 2.​ Percentage change in average mathematics scale scores for National Assessment of Educational Progress Long-Term Trend (NAEP-LTT, age 13), NAEP (grade 8), and Trends in International Math and Science Study (TIMSS, grade 8), relative to earliest presented year: 1999 through 2024

NOTE: For each survey, each value is the percentage change between the respective scale score and the overall average scale score for the first data collection that is presented in the figure. The first year presented for NAEP-LTT is 1999; for NAEP, it is 2000; and for TIMSS, it is 1999. Not all apparent differences between estimates are statistically significant. For detailed information, including the results of statistical testing and findings on the individual assessments, please see the full report.
Figure 3.​ Percentage change in average reading scale scores for National Assessment of Educational Progress Long-Term Trend (NAEP-LTT, age 9), NAEP (grade 4), and Progress in International Reading Literacy Study (PIRLS, grade 4), relative to earliest presented year: Selected years, 1998 through 2024

NOTE: For each survey, each value is the percentage change between the respective scale score and the overall average scale score for the first data collection that is presented in the figure. The first year presented for NAEP-LTT is 1999; for NAEP, it is 1998; and for PIRLS, it is 2001. PIRLS 2021 is not included in this analysis because, due to the COVID-19 pandemic, it assessed students in grade 5 whereas all other PIRLS administrations assessed students in grade 4. Not all apparent differences between estimates are statistically significant. For detailed information, including the results of statistical testing and findings on the individual assessments, please see the full report.
Figure 4.​ Percentage change in average reading scale scores for National Assessment of Educational Progress Long-Term Trend (NAEP-LTT, age 13), and NAEP (grade 8), relative to earliest presented year: Selected years, 1998 through 2024

NOTE: For each survey, each value is the percentage change between the respective scale score and the overall average scale score for the first data collection that is presented in the figure. The first year presented for NAEP-LTT is 1999; and for NAEP, it is 1998. Not all apparent differences between estimates are statistically significant. For detailed information, including the results of statistical testing and findings on the individual assessments, please see the full report.
Figure 5.​ Percentage change in average science scale scores for National Assessment of Educational Progress (NAEP, grade 4) and Trends in International Math and Science Study (TIMSS, grade 4), relative to earliest presented year: 2003 through 2023

NOTE: For each survey, each value is the percentage change between the respective scale score and the overall average scale score for the first data collection that is presented in the figure. The first year presented for NAEP is 2009; and for TIMSS, it is 2003. Not all apparent differences between estimates are statistically significant. For detailed information, including the results of statistical testing and findings on the individual assessments, please see the full report.
Figure 6. Percentage change in average science scale scores for National Assessment of Educational Progress (NAEP, grade 8) and Trends in International Math and Science Study (TIMSS, grade 8), relative to earliest presented year: 1999 through 2023.

NOTE: For each survey, each value is the percentage change between the respective scale score and the overall average scale score for the first data collection that is presented in the figure. The first year presented for NAEP is 2009; and for TIMSS, it is 1999. Not all apparent differences between estimates are statistically significant. For detailed information, including the results of statistical testing and findings on the individual assessments, please see the full report.
Footnote
1. Assessments include the National Assessment of Educational Progress Long-Term Trend (NAEP-LTT, students ages 9 and 13), NAEP (students in grades 4 and 8), the Progress in International Reading Literacy Study (PIRLS, students in grade 4), and the Trends in International Mathematics and Science Study (TIMSS, students in grades 4 and 8).
Authorship
Steven Bahr: Formal analysis (supporting); Validation (supporting). Nancy Collins: Formal analysis (lead); Methodology (equal). Kimberly Curtis: Writing – original draft and editing (equal). Jane Hall: Writing – original draft and editing (equal). Robbie Kaplan: Project coordination (supporting); Web publication (lead). Mark Low: Validation (lead); Methodology (equal). Kathleen Mulvaney-Panjwani: Validation (supporting). Tim Oltman: Data curation (lead); Visualization (lead). Melissa Patton: Formal analysis (supporting); Validation (supporting).
About Activate Research
Activate Research, Inc. is a woman-owned small business that provides expert social science consulting services and works with government agencies and private-sector clients to conduct rigorous research and data analysis. Its program evaluation, quantitative and qualitative data collection, and technical assistance services help clients make informed decisions and meet their strategic goals. Activate delivers high-quality solutions tailored to clients’ unique needs by emphasizing technical accuracy, compliance with standards, and a commitment to impactful, actionable insights.