CoChat
Sign Up
Cited by 15Education during COVID-19 pandemicSchool Choice and PerformanceCOVID-19 Pandemic Impacts

Unequal Learning Loss: How the COVID-19 Pandemic Influenced the Academic Growth of Learners at the Tails of the Achievement Distribution

Open access
Peters, Scott J., Langi, Meredith, Kuhfeld, Megan, Lewis, Karyn·2025·Journal for the Education of the Gifted
DOI 10.1177/01623532241301079
Open this paper in CoChat — free accountOpen PDF

Abstract

The COVID-19 pandemic resulted in substantial unfinished learning for U.S. students, but to differing degrees for various subgroups. Students of color, from low-income families, or who attended high-poverty schools experienced greater unfinished learning. In this study, we examined the degree of unfinished learning for students who went into the pandemic scoring in the top or bottom 10% of achievement in math or reading. Our results show that students who scored at or below the 10th percentile grew less during the pandemic than their similarly scoring, pre-COVID peers and, as of the end of the 2021–2022 school year, had yet to rebound toward pre-COVID levels of growth or achievement. Conversely, students who scored at or above the 90th percentile largely grew at rates closer to their pre-COVID peers. These students were harmed less academically and have recovered more quickly than their peers scoring at or below the 10th percentile.

Integrity signals

Registry

Crossref · 10.1177/01623532241301079

Cites

22 of 23 references verified1 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

5 of 28 cited sources checked · 2 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    The COVID-19 pandemic substantially harmed the average American student's achievement.

    “The COVID-19 pandemic had substantial negative effects on the average American student's achievement (Betthäuser et al., 2023;Cohodes et al., 2022).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Betthäuser et al., 2023 source

      The Discussion reports a substantial average learning deficit across the countries studied. The section ‘Learning deficits arose early in the pandemic and persist’ says persistent deficits also appear in US estimates, but it does not establish that the estimated effect represents the average American student or give a US-specific magnitude described as substantial. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites Cohodes et al., 2022

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Findingown data

    In the COVID sample, bottom-decile students' reading growth in Year 2 was lower than that of pre-COVID bottom-decile peers.

    “students who started in the bottom decile grew less in reading and math in Year 2 than did their pre-COVID peers”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: School closures affected instruction in a way similar to summer break.
    • Rests on an assumption: Without the pandemic, the pre-COVID growth trends would have continued for the COVID sample.
  • Findingown data

    Top-decile math students in the COVID sample grew 89% to 104% as much as pre-COVID peers in Year 2.

    “Top decile math students during COVID grew between 89% and 104% as much as their pre-COVID peers in Year 2.”
    Quote in the paperSubstantial claim
  • Findingown data

    In the COVID sample, bottom-decile students' math growth in Year 2 was lower than that of pre-COVID bottom-decile peers.

    “students who started in the bottom decile grew less in reading and math in Year 2 than did their pre-COVID peers”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: School closures affected instruction in a way similar to summer break.
    • Rests on an assumption: Without the pandemic, the pre-COVID growth trends would have continued for the COVID sample.
  • Findingown data

    Across cohorts, bottom-decile students' Year 3 math growth in the COVID sample was 88% to 100% of pre-COVID peers' growth.

    “Year 3 growth for bottom decile students in the COVID sample was between 88% and 100% as much as their pre-COVID peers in math”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. 12
  • May combine two claims. 21
  • The quote only partly states this claim. 1

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Jakubowski et al. (2024) found that in schools with average to very short closures, the highest achievers lost the least ground across content areas.

    “in schools with average to very short closures, the highest achieving students lost the least ground in terms of academic achievement across all content areas.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Jakubowski et al. (2024) source

      The indexed repository passage for the cited paper (no section or page given) says losses were greater in schools with relatively longer closures, supporting the shorter-closure comparison. It does not report that the highest achievers lost the least ground in schools with average-to-very-short closures, or that this held across all content areas. A nearby passage about lower-achieving students belongs to a separately listed publication, not the cited paper. (LLM, conf=0.73)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationreasoning

    The findings provide further evidence of a Matthew Effect, in which pre-pandemic high performers maintained or increased their advantage because bottom-decile students' growth and achievement fell.

    “The present findings are further evidence of a Matthew Effect (e.g., Morgan et al., 2008), whereby students who were already performing at the highest levels before the pandemic maintained their advantage or became relatively even more advantaged due to the drop in growth and achievement for students in the bottom decile.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Morgan et al., 2008 source

  • May combine two claims.

    Premisecitation

    Black, Hispanic, and low-income students were disproportionately harmed by school closures and related disruptions and experienced greater learning loss.

    “research has shown Black, Hispanic, and low-income students were disproportionately harmed by school closures and related academic disruptions (Goldhaber et al., 2022), and as a result experienced even greater learning loss due to the pandemic”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Goldhaber et al., 2022 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Jakubowski et al. found that in schools with long or very long closures, the most advanced students lost more than average or low-scoring peers.

    “this relationship flipped in schools that experienced long of very long closures-the most-advanced students lost more in terms of academic achievement than their average or even low-scoring peers.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Jakubowski et al. (2024) source

      The indexed passages are from the repository listing (no paper section or page is available), rather than the paper’s results. Searches for long or very long closures and comparisons between high- and low-scoring students found no passage from the cited paper establishing which group lost more. The decisive part of the paper may be missing from our indexed copy. (LLM, conf=0.68)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    Bottom-decile students' Year 2 growth relative to pre-COVID peers was .62 in math and .61 in reading.

    “these students grew less than two-thirds as much as their pre-COVID peers (Growth Ratios of .62 in math and .61 in reading)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Bottom-decile students' Year 3 growth was nearer to pre-COVID trends but still below them in both reading and math, with ratios of .92.

    “Year 3 growth was closer to pre-COVID trends (Growth Ratios of .92 in reading and in math) but still fell short of pre-COVID levels”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Findingown data

    In Year 3, bottom-decile students showed less evidence of rebounding to pre-COVID levels than top-decile students.

    “during Year 3 students in the bottom decile showed less evidence of rebounding back to pre-COVID levels.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    In math, the gap was about 10% larger in early grades and unchanged in middle grades.

    “For math, the gap was roughly 10% larger in the early grades and unchanged in the middle grades (see Kuhfeld & Lewis, 2022;Lewis et al., 2022, for additional details).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Kuhfeld & Lewis, 2022 source
    • cites Lewis et al., 2022

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Study design and conductcitation

    MAP Growth uses a norming procedure to estimate nationally representative achievement and growth norms, with the 2020 norms based on pre-pandemic performance.

    “It also applies a sophisticated norming procedure to estimate nationally representative achievement and growth norms, the most recent of which were released in 2020 and are based on pre-pandemic student performance (Thum & Kuhfeld, 2020).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Thum & Kuhfeld, 2020

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    Lewis and Kuhfeld (2021) found in reading that the top 20% met or exceeded growth expectations while the bottom 20% grew more slowly than pre-COVID norms.

    “In reading, students in the top 20% of achievement met or exceeded their growth expectations, whereas students in the bottom 20% showed slower growth compared to pre-COVID norms (32nd to 39th conditional growth percentiles”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Lewis and Kuhfeld (2021)

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Full-information maximum likelihood allowed students to be included even when they were not observed at every time point.

    “Full-information maximum likelihood (FIML) was used to estimate our models, which allows for the inclusion of students in the analyses even if they are not observed at all time points.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Top-decile students were less harmed academically and recovered faster than bottom-decile peers.

    “These students were harmed less academically and have recovered more quickly than their peers scoring at or below the 10th percentile.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Top-decile students made substantial progress toward academic recovery, while bottom-decile students had not yet turned the corner and remained far from pre-COVID levels.

    “students in the top deciles have made significant progress towards academic recovery, whereas students that started in the bottom deciles have yet to turn the corner and are much further from attaining pre-COVID levels of growth or achievement.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Lewis and Kuhfeld found in math that even the top quintile grew less than expected, and the bottom quintile grew even more slowly.

    “even the top quintile of math achievers in Fall 2019 still grew less than expected (40th-45th CGP).The bottom quintile showed even slower growth (26th-32nd GCP).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Lewis and Kuhfeld (2021)

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The data cannot directly explain why top-decile students were more insulated from the pandemic's negative achievement effects.

    “The data do not speak directly to why top decile students were more insulated to the negative achievement effects of the COVID-19 pandemic, but we can think of two possible explanations.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductnone given

    Few students took the Spring 2020 test, and those who did were not representative of the pre-COVID sample.

    “very few students tested in Spring 2020 and those who did are not representative of the pre-COVID sample.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductcitation

    MAP Growth is vertically scaled in Rasch units and is usually given at least twice a year.

    “the assessment is vertically scaled in Rasch units (i.e., RIT scores) and is typically administered at least two times per year (Fall and Spring; NWEA, 2019).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites NWEA, 2019

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    Less is known about how students' prior achievement shaped the size of unfinished learning.

    “less is known about the role of students' prior achievement in shaping the magnitude of unfinished learning.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Schult et al. found the 5th percentile group scored 28 points lower and the 95th percentile group scored 11 points higher than pre-COVID peers in operations.

    “the 5th percentile group scored 28 points lower and the 95th percentile group scored 11 points higher than their pre-COVID peers”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Schult et al. (2022) source

  • May combine two claims.

    Interpretationreasoning

    Bottom-decile students incurred more pandemic learning loss in Year 2 and made less recovery progress in Year 3 than top-decile students.

    “bottom decile students incurred more learning loss from the pandemic during Year 2 and made less progress towards recovery in Year 3 compared to top decile students.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Compared with Fall 2019, Fall 2021 scores were .20 to .27 SD lower in math and .09 to .18 SD lower in reading.

    “Kuhfeld et al. (2022) reported that compared to Fall 2019, Fall 2021 test scores were .20 to .27 standard deviations lower in math and .09 to .18 standard deviations lower in reading”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Kuhfeld et al. (2022) source

  • May combine two claims.

    Premisecitation

    Kuhfeld, Lewis, et al. (2020) found Fall 2020 MAP Growth results were reasonably comparable for Grades 3-8 but possibly not for Grades K-2.

    “Their findings suggest that results are reasonably comparable for students in Grades 3-8 but may not be so for Grades K-2.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Kuhfeld, Lewis, et al. (2020)

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Study design and conductnone given

    Even though students did not always remain beyond the percentile cut-off, the achievement levels of the groups stayed widely separated at all time points.

    “Although the students in each group did not necessarily stay at or below the set percentile cut-off within each subsequent time point, the achievement levels of the groups remained very separated across all time points.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    More students are likely to score at or below the 10th percentile because of COVID, while roughly the same proportion will score at the 90th percentile.

    “Although more students are likely to score at or below the 10th percentile as a result of COVID, roughly the same proportion of students will score at the 90th percentile.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The total sample sizes differed because different numbers of students were tested with MAP Growth before and after COVID.

    “The total sample sizes were also different because different numbers of students were tested with MAP Growth pre-and post-COVID.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The study lacked data on which individual students were labeled as having a disability or as gifted and talented.

    “Similarly, although there are implications for exceptional learners, we did not have data on which individual students were labeled as having a disability or as gifted and talented.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Black students made up about 15% of the COVID sample and 17% to 18% of the pre-COVID sample.

    “For example, Table 1 reports that Black students made up approximately 15% of the COVID sample but 17%-18% of the pre-COVID sample.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    MAP Growth's vertical, grade-independent scale and computer adaptive design made it especially useful for this research question.

    “Instead, they were especially useful for this question because of MAP Growth's vertical, grade-independent scale and computer adaptive design.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    According to MAP Growth norms, an RIT score of 210 corresponds to the 94th percentile for third grade and the 50th percentile for fifth grade.

    “According to MAP Growth norms (Thum & Kuhfeld, 2020), this represents the 94th percentile for a third-grade student or the 50th percentile for a fifth-grade student.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Thum & Kuhfeld, 2020

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductreasoning

    The data come from a large national sample of NWEA MAP Growth test-takers, but the results are not necessarily representative of the nation or any single state.

    “These data come from a large national sample of students who completed the NWEA MAP Growth test but that does not mean the results are necessarily representative of the nation as a whole or any single state.”
    Quote in the paperNot a substantial claim
30 of 30

  • The authors state this as an assumption.

    School closures affected instruction in a way similar to summer break.

    “This was done intentionally based on the hypothesis that school closures had an impact on instruction like that of summer break.”

  • The paper relies on this without support.

    Without the pandemic, the pre-COVID growth trends would have continued for the COVID sample.

    “This design allows us to isolate the timing of COVID-19 in the longitudinal growth trajectories of students at the bottom and the top of the distribution.”

  • The authors state this as an assumption.

    The study assumes a linear rate of growth for both math and reading.

    “However, to be consistent with prior seasonal learning research (Kuhfeld & Lewis, 2022;Kuhfeld et al., 2021;von Hippel et al., 2018), we assume linear rate of growth for both subjects.”

  • The paper relies on this without support.

    Most percentile cut scores used to flag students for interventions are based on pre-COVID samples.

    “Most of the percentiles used to make these decisions are based on pre-COVID samples.”
    “the 10th percentile cut score, and similar cut scores used to flag students for additional interventions on other assessments, are based on pre-COVID data.”

  • The paper relies on this without support.

    High-dose tutoring implementation has been challenging, and the neediest students are least likely to participate.

    “For example, although high-dose tutoring has received substantial attention as a possible way to mitigate COVID unfinished learning, implementation has proven challenging with many of the neediest students being least likely to participate.”

  • The paper relies on this without support.

    Prior research has established that growth slowed for all students during COVID and that COVID worsened preexisting subgroup disparities.

    “Substantial research has already established that (a) growth slowed for all students during COVID and (b) COVID exacerbated preexisting disparities across subgroups.”
6 of 6

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Reported effect lies outside its accompanying interval

    Minor

    The reported effect size is −0.14, but the accompanying interval runs from −0.017 to −0.010. The effect size cannot lie within that interval.

  • Math demographic totals conflict with their subgroups

    Minor

    Table 3 reports zero for every listed racial and ethnic category in the full pre-COVID math sample for Grades 6-8, yet its bottom-decile subgroup has nonzero proportions in those categories. A subgroup with those demographics cannot be part of a full sample whose corresponding proportions are all zero.

2 of 2

Cites

Flagged first, then by citations

Research works printed but not linked: 9

  1. Barker E., 2022, State Education Standard, 22, 16
  2. Cohodes S., Goldhaber D., Hill P., Ho A., Kogan V., Polikoff M., Sampson C., West M. (2022). Student achievement gaps and the pandemic: A new review of evidence from 2021–2022. https://crpe.org/wp-content/uploads/final_Academic-consensus-panel-2022.pdf
  3. Kuhfeld M., 2022, Student achievement in 2021–2022: Cause for hope and continued urgency
  4. Kuhfeld M., 2020, Comparability analysis of remote and in-person MAP Growth testing in fall 2020
  5. Lewis K., 2021, Learning during COVID-19: An update on student achievement and growth at the start of the 2021–22 school year
  6. NWEA, 2019, MAP® Growth™ technical report
  7. Renaissance Learning, 2021, How kids are performing: Tracking the school-year impact of COVID-19 on reading and mathematics achievement
  8. Thompson S., 2018, Npj Science of Learning, 3
  9. Thum Y. M., 2020, NWEA 2020 MAP Growth achievement status and growth norms for students and schools

Cited by

Sorted by citations

Related

Authors
Peters, S. J., Langi, M., Kuhfeld, M., & Lewis, K.
Affiliations
  • Northwest Evaluation Association (US)
  • Wested (US)
Corresponding author
Peters, Scott J.
ORCID
Venue
Journal for the Education of the Gifted
Publisher
SAGE Publications
Published
2024-11-20
Volume
48
Issue
1
Pages
25-47
ISSN
  • 0162-3532 (ISSN-L)
  • 2162-9501
DOI
10.1177/01623532241301079
OpenAlex
W4404567541
Semantic Scholar
fdbc7042e9f94631c145def57d9c629213c2cc6a
Cited by
15
References
19
Citations by year
2026: 8 · 2025: 5 · 2024: 2
FWCI
19.91
Type
journal article
Language
en
Open access
Licence
journals.sagepub.com/page/policies/text-and-data-mining-license (tdm)
Locations
journal for the education of the gifted (publishedVersion)
Topics
  • Social Sciences › Education › Education during COVID-19 pandemic
  • Social Sciences › Education › School Choice and Performance
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › COVID-19 Pandemic Impacts
Keywords
COVID-19 pandemic, academic progress, achievement distribution, low-achieving students, high-achieving students, students of color, low-income students, high-poverty schools

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-07.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-07.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review