Pandemic-related performance change was more positive in Cluster 2 than Cluster 3 countries across all domains, and more positive than Cluster 1 in reading and science¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Model-based pandemic-related performance change (ATE) favoured Cluster 2 over Cluster 3 in mathematics (6.9 points, p = .04) and reading (9.9 points, p = .03), but not science (5.4 points, p = .11). → Giang Pham 2026 q2 i? Cluster 2 showed more positive pandemic-related change than Cluster 1 in reading (12.9 points, p = .07) and science (21.3 points, p = .02), with no significant difference in mathematics. → Giang Pham 2026 (2)

Evidence¶

Giang Pham 2026¶

Giang Pham, Franziska Maria Locher, Dimitra Kolovou, Eliane Arnold. (2026). COVID-19: A cross-national analysis with PISA data. Learning and Instruction 105 (2026) 102366. https://doi.org/10.1016/j.learninstruc.2026.102366

q2 · i? · associational · r2

Regression on the potential-outcome-based Pandemic-Related Performance Change (N = 47 countries), contrasting Cluster 2 vs Cluster 3. The article reports "6.9 PISA points in mathematics, p = .04" and 9.9 points in reading (p = .03); the science difference (p = .11) was not significant.

"Pandemic-Related Performance Change was more positive in Cluster 2 countries (Accentuated Teacher-Led Synchronous Instruction) compared with Cluster 3 countries (Accentuated Instructional Disruption) across all domains (6.9 PISA points in mathematics, p = .04; 9.9 points in reading, p = .03; 5.4 points in science, p = .11)."

Giang Pham 2026 (2)¶

Giang Pham, Franziska Maria Locher, Dimitra Kolovou, Eliane Arnold. (2026). COVID-19: A cross-national analysis with PISA data. Learning and Instruction 105 (2026) 102366. https://doi.org/10.1016/j.learninstruc.2026.102366

q2 · i? · associational · r2

Same regression framework contrasting Cluster 2 vs Cluster 1 on pandemic-related performance change. The science difference was significant (21.3 points, p = .02); reading (p = .07) was marginal; mathematics and all Cluster 1 vs Cluster 3 contrasts were not significant.

"A more positive effect of belonging to Cluster 2 over Cluster 1 (Family-Supported Asynchronous Instruction) was also observed in reading (12.9 points, p = .07) and science (21.3 points, p = .02). No significant differences were found between Cluster 1 and Cluster 2 in mathematics, nor between Cluster 1 and Cluster 3 across all domains."

Discussion¶

Related Claims¶

• Countries with Accentuated Teacher-Led Synchronous Instruction showed better PISA 2022 performance than countries with Accentuated Instructional Disruption across all domains, with modest effects — possibly the same claim (merge candidate)
• Cluster 2 countries outperformed Cluster 1 countries in PISA 2022 only in science (about 20 PISA points, p = .002), with no significant differences in mathematics or reading — related
• Pandemic-era teaching and learning conditions across 47 countries fall into three clusters: Family-Supported Asynchronous Virtual Instruction, Accentuated Teacher-Led Synchronous Instruction, and Accentuated Instructional Disruption — a broader claim this one bears on