Reviewed — a delegated second read (D-295, D-356)
Reviewed on 2026-10-20 against the content and evidence rules: every sentence was read against them, and at least one citation in each unit that cites was re-resolved against the raw record it names. The owner delegated the read on 2026-10-20 and the Lead signed on his behalf (D-356, by the method of D-295) — it is a delegated review, not the owner's own reading of the page.
We recommend it as the required companion to the Mindset card on this shelf: the frame is worth having and the popular claim is stronger than the measurement, and a member deserves both halves.
Two meta-analyses in one paper: how strongly the mindset construct relates to achievement, and how much mindset interventions move it. Both averages are small, and the interventions that did show an effect were concentrated among academically at-risk and low-income students rather than spread across everybody.
How a moderator analysis rescues the useful part of a finding from an average that hides it; why a small average effect and a real effect in a subgroup are compatible claims; and what publication bias checks look like when they are done properly.
Two things to try
- Read the two effect sizes and write them down, then compare them with how the idea was described to you.
- Find the moderator table and write which group the interventions actually helped.
- Take one educational programme you know of and ask which subgroup its evidence came from.
Level, and what it means here
This is a primary source written for specialists, and it assumes the earlier items in its collection. It is a description of the paper, not of you.