2025年仍未恢复：PISA中的疫情学习落后与停课时长的有限作用

英文标题/原题

No recovery by 2025: Pandemic learning shortfalls in PISA and the limited role of school closure length

作者

刘康; 杨晓娟

通讯作者

刘康

研究方向

教育评估；疫情学习损失；国际大规模测评

关键词

PISA; COVID-19 learning loss; school closures; international large-scale assessment; trend analysis; linking error

投稿/发表信息

发表期刊/目标

Educational Assessment, Evaluation and Accountability

投稿/成果状态

投稿中

投稿日期

2026-10-23

发表日期

-

原地址/投稿系统地址

https://submission.springernature.com/new/submission/5ce6aba0-4517-4110-aad4-b92c4e40c7f7

期刊与检索情况

检索情况

SSCI、Scopus、ERIC（Springer 期刊页，2026-10）

摘要/说明

School closure length is widely treated as the dose of pandemic learning loss. Using PISA 2003–2025 for 61 education systems, this preregistered study evaluated recovery as each system's deviation from its own pre-pandemic trend, with inference at the country level. In mathematics, systems were 15.6 points below trend in 2022 and 22.7 points below trend in 2025. The change between the two rounds (−7.2 points, 95% CI −15.1 to 0.6) gave no evidence of recovery. Principal-reported closure length was not associated with this change (0.04 points per week, −0.14 to 0.27), and in an exploratory analysis within 76 systems, schools closed longer did not score lower in 2022. The socioeconomic gap narrowed after 2022 because advantaged students declined, not because disadvantaged students recovered. Evaluations of recovery should benchmark outcomes against pre-pandemic trends, and closure exposure is a poor basis for targeting recovery resources.

基金信息

无

论文备注

项目目录：E:\论文选题_本地数据集\P03_疫情后恢复_停课天数与2025反弹。交付：05_稿件/manuscript.pdf、supplement.pdf、title_page.pdf、cover_letter.pdf、highlights.pdf；投出版在 05_稿件/投稿包_SEE/。预注册 https://osf.io/49ney/（2026-10-21，封存至 2027-10-21）。2026-10-22 被 SEE 以范围不符初审拒稿，准备改投。 2026-10-23 改投 EAEA（Educational Assessment, Evaluation and Accountability），投稿包 05_稿件/投稿包_EAEA/，项目目录改名为【投稿中_EAEA】。