Revisiting "The Power of Feedback" from the perspective of the learner
Mandouit L, Hattie J · 2023
grade Dreviewdeveloper-ledunreplicated
Sample
not verified — see design_notes
Population
School students (Australian; the work was funded by the ARC Science of Learning Research Centre). Exact sample size, grade levels and instrument could not be confirmed.
Design
HONEST VERIFICATION LIMIT. The article is paywalled (Learning and Instruction 84:101718) with no open-access or repository copy (unpaywall: oa_status closed, has_repository_copy false), and no abstract is deposited with Crossref, Semantic Scholar or the institutional repositories. We could not read the paper. What follows is assembled from secondary reporting and is flagged unverified. It is filed as a pointer for a future deep read, NOT as evidence that moves any verdict. This is a learner-perspective study — student reports of what they want from feedback — so it is descriptive by construction and carries no causal weight regardless of sample size.
Key findings
Hattie revisiting his own Hattie & Timperley (2007) feed-up / feed-back / feed-forward model from the receiving end. The reported finding is that students want feedback carrying task and improvement information rather than evaluation, and that the emotional/relational side of receiving feedback is under-modelled. Frequencies reported in secondary sources for what students say they look for: "how to improve" 51%, "error flagging" 50%, "done well" 41%, "feed forward / next time" 35%. NUMBERS NOT VERIFIED AGAINST THE PAPER. Directionally these corroborate the moderator structure this database already rests on (task information helps, person evaluation does not), from a different kind of evidence — which is why it changes nothing.
Genetic confound
Not applicable — self-report of preferences, no outcome claim.
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| What students report wanting in feedback (unverified secondary reporting) | % of student responses coded | how to improve 51%; error flagging 50%; done well 41%; feed-forward 35% | self-report survey | cross-sectional | none | not-applicable | non-cognitive |