Self Assessment Improves Self Regulated Learning¶

Claim · All claims
Evidence · 1 study · quant-synthesis r? · q4 · i1 small · n=2,305 students across 19 studies
Currency · newest study 2017 · not yet checked against its sources · not reviewed by a person

When learners appraise the quality of their own work against criteria, they gain the evaluative information needed to plan, monitor, and adjust their studying — the core operations of self-regulated learning.

Subclaims¶

q4 i1 A meta-analysis of self-assessment interventions found a small but statistically significant positive effect on students' use of self-regulated-learning strategies (d=0.23, 12 effect sizes, 369 participants), with converging support from a reduction in maladaptive self-regulatory behaviors and a large effect on self-efficacy. → Panadero et al. 2017

Evidence¶

Panadero et al. 2017¶

Panadero, E., Jonsson, A., & Botella, J. (2017). Effects of self-assessment on self-regulated learning and self-efficacy: Four meta-analyses. Educational Research Review, 22, 74–98. doi:10.1016/j.edurev.2017.08.004

q4 · random-effects meta-analysis (four sub-analyses) · i1 · small effect on Learning SRL, d=0.23; medium/non-significant on the other two SRL measures; large on self-efficacy, d=0.73 · n=2,305 students across 19 studies · quant-synthesis · r?

Four random-effects meta-analyses (Hedges & Olkin method) pooled 19 experimental and quasi-experimental studies of self-assessment interventions in K-12 and higher-education classrooms. For the "Learning SRL" outcome — measures of strategy use such as planning, monitoring and self-regulatory strategy deployment — 12 effect sizes from 369 students produced a combined effect of d=0.23, 95% CI [0.08, 0.39], a small but statistically significant positive effect. A second meta-analysis on "Negative SRL" (self-regulatory behaviors detrimental to learning, e.g. procrastination) found d=−0.65 (k=6, n=347; CI crossed zero), meaning self-assessment reduced these behaviors; a third on SRL measured qualitatively found d=0.43 (k=3, n=338; CI crossed zero, underpowered). A separate meta-analysis of 27 effect sizes on self-efficacy found a larger effect, d=0.73, 95% CI [0.39, 1.08], moderated by gender (larger for girls) and by whether the self-assessment intervention included a monitoring component.

Studies bearing on this claim¶

Should learners assess their own work, and can their self-ratings be trusted? This table gathers the studies behind this claim and the 11 narrower claims that bear on it: 16 distinct studies (7 causal, 3 quant-synthesis, 3 review, 1 associational, 1 theoretical, 1 qualitative), q2–q4, 3 of 16 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

Mechanism. Self-assessment is theorized to support self-regulated learning by making performance standards explicit and by generating feedback that learners can act on without waiting for an external evaluator. It connects closely to assessment for learning, in which assessment information is used formatively rather than only to certify achievement. In self-regulated learning terms, self-assessment feeds the monitoring and reflection phases: learners who can accurately judge where their work stands relative to a standard are better positioned to set appropriate goals, select strategies, and persist or redirect effort — see self-regulated learning.

Boundary conditions to watch. Self-assessment quality depends heavily on the availability of clear criteria, exemplars, or rubrics; without external reference points, learners — especially novices — tend to overestimate their competence, and inaccurate self-estimates can lead to premature termination of study or misplaced practice effort. Training in self-assessment and calibration is generally treated in the literature as a prerequisite rather than an optional extra. Rubrics and worked exemplars help here by reducing the extraneous load of judging one's own work — see cognitive load reduction improves learning. Self-assessment also appears to benefit self-regulation more directly than it benefits task performance itself; gains in outcomes are typically mediated by improved strategy use rather than by direct effects on scores.

Calibration. A recurring theme in the literature is that self-assessment is only as useful as it is accurate. Miscalibrated learners — particularly low performers — may believe they have mastered material they have not, which undermines the monitoring function self-assessment is meant to serve. Interventions that pair self-assessment with external feedback or with comparison against exemplars tend to improve calibration, which in turn is expected to strengthen the downstream self-regulation benefits. This mirrors the broader pattern in which feedback improves learning only when it is actionable and tied to criteria learners understand.

Open questions. One synthesis is recorded: Panadero et al. (2017), whose four meta-analyses of 19 studies found a small positive effect on strategy use (d = 0.23, k = 12) and confidence intervals crossing zero for the other two SRL measures. No primary study and no study of calibration accuracy is recorded here yet. Open empirical questions include how durable the self-regulation gains are over time, whether effects differ across age groups and domains, and how self-assessment interacts with formative feedback practices.

Merged from “Self Assessment Against Criteria Supports Self Regulated Learning” (self-assessment-against-criteria-supports-self-regulated-learning): Mechanism. Self-regulated learning theory describes a cycle of forethought, performance, and self-reflection (Zimmerman, 2002; see Self-regulated learning). Self assessment against criteria operationalizes the self-reflection phase: criteria give learners a standard to compare their performance against, which makes self-monitoring accurate rather than impressionistic [+M]. Without explicit criteria, learners tend to overestimate their competence, and the resulting miscalibration undermines planning and effort allocation [~M]. Andrade (2010) frames self assessment as the mechanism by which students become "the definitive source of formative assessment" — internalizing the evaluative criteria that teachers otherwise apply for them. Rubric-based self assessment has been associated with improved learning and with more accurate self-regulatory judgments, particularly when students use the criteria before and during drafting rather than only after completion [+M] (Andrade & Valtcheva, 2009; Panadero & Alonso-Tapia, 2013). This positions self assessment as a form of assessment for learning rather than merely a grading convenience.

Boundary conditions. The quality of the criteria matters. Vague or holistic criteria invite leniency bias and self-serving judgments [~M]; specific, task-relevant criteria (e.g., analytic rubrics, quality checklists, comparison to annotated exemplars) support more accurate self-evaluation [+M]. Accuracy also improves when learners have prior exposure to exemplars of the target quality, and when self assessment is practiced repeatedly rather than used once [+M]. There is a known tension with feedback: if self assessment replaces external feedback entirely, learners may lack the information needed to close the gap between their judgment and the standard — self assessment works best alongside, not instead of, teacher or peer feedback [~M] (see Action-oriented feedback). A further boundary condition is that checklists used as evaluation tools can fail in certain online contexts (see Checklist evaluation ineffective online) [-M], so the format of the criteria is not neutral. Novices and younger learners are also more prone to leniency and miscalibration, so structured criteria and modeling of the self-assessment process matter most where self-regulation is weakest [~M].

Design implications. To realize the self-regulatory benefit, criteria should be introduced early enough to guide forethought — not only consulted after work is complete — and the self-assessment step should ask learners to identify evidence in their own work against each criterion, not merely assign a score. Co-constructing criteria with students is a promising but under-documented variant. Where self assessment is used, pair it with a mechanism for acting on the judgment (revision, goal-setting) so that accurate monitoring translates into regulated effort.

Open questions. Of the works cited in this merged paragraph, only Panadero's group is recorded under Evidence (as the 2017 meta-analyses); Zimmerman (2002), Andrade (2010), Andrade & Valtcheva (2009) and Panadero & Alonso-Tapia (2013) are not, so the bracketed tags above rest on no entry on this page. Key moderators to document include learner age and domain expertise, whether criteria are co-constructed with learners, and whether effects on self-regulation transfer beyond the assessed task.

Learner Variables¶

• Self-Regulation — outcome: instruction changes it

Related Claims¶

• Assessment for learning improves achievement. — formative use of assessment information is the broader practice within which self-assessment sits.
• Cognitive load reduction improves learning. — well-designed criteria and rubrics reduce the extraneous load of judging one's own work.
• Feedback improves learning. — external feedback paired with self-assessment improves calibration and makes self-monitoring actionable.
• Self-regulated learning — the theoretical framework this claim connects self-assessment to.
• Checklist evaluation ineffective online — a boundary condition: checklists used as evaluation tools can fail in certain online contexts.
• Action-oriented feedback — external feedback that complements learner self-judgment.
• Assessment — the element family that self-assessment tasks belong to.
• Adding small peer feedback groups to web-based SRL training and a diary (Group TDP) produced statistically significant gains in SRL knowledge, the SRL overall score, self-efficacy and the mathematics focus score, the last of which Group TD did not show — related
• Self-efficacy for self-regulated learning predicts academic achievement above and beyond prior achievement, gender, SES, intelligence, personality traits, and self-esteem — related
• Language teachers in two workshops responded positively to the evaluation framework, leading to a four-category revision — related
• In a game-based problem-solving study, trait self-regulation scores related significantly only to knowledge map scores, with higher planning, self-monitoring, effort and self-efficacy going with better knowledge map performance — related
• Formative-assessment-based LA shows positive effects on achievement, motivation, and understanding, with personalized feedback reading time mattering — related
• Learning paths used for SRL improved accuracy and performance without increasing effort, and tutoring-system benefits required sustained practice — related
• Feedback about self as a person is the least effective feedback level; task, process, and self-regulation feedback support learning and combine for greater impact — related
• A within-subjects pilot with 27 professional software engineers found statistically significant self-reported self-efficacy gains across all 10 assessed skill areas after completing the cc-self-train curriculum — a narrower finding that bears on this claim
• AI feedback supports or undermines self-regulated practice depending on whether it supports or substitutes for internal monitoring — related
• The framework's broader relevance rests on a shared learning structure of embodied skill, expert feedback, and self-monitoring, not direct equivalence across domains — related
• AI-section students showed tool-linked self-efficacy and attenuated metacognitive calibration, with the human-pair contrast group showing larger concept-map gains in cross-links and hierarchy — related
• AI-assisted practice applications improved music students' performance outcomes, self-efficacy, and self-regulated learning — related
• A randomized controlled trial with 86 K-12 students found a significant gain in self-regulated-learning strategy use from agentic coaching systems — related
• LLMs can deliver scalable, personalized, interactive feedback almost instantly, reducing dependency on teacher availability and supporting self-regulated learning in large courses — related
• Correlations with self-regulation tend to be larger for course grades than for standardized test scores in several coded studies — related
• Self-monitoring intervention effects rest on four key components: selecting a target behavior, defining it, self-assessment, and self-recording — related
• Teachers' self-assessed reflection skill exceeds their externally assessed reflection level, with only a low correlation between the two — related
• Rubric use has a small positive effect on students' self-efficacy (g = 0.18), based on only four studies — related
• Rubric use has a small, statistically non-significant pooled effect on self-regulated learning (g = 0.23) — related
• For self-regulated learning outcomes, rubrics combined with both self- and peer assessment show a significantly larger effect than rubrics with self-assessment alone — a narrower finding that bears on this claim
• Regulating GA proceeds via metacognitive monitoring of adequacy and accuracy plus control through outward- and inward-oriented strategies — a narrower finding that bears on this claim
• Formative assessment shows gains in student self-efficacy and motivation and particular benefits for disadvantaged and low-achieving students — related
• Across both experiments, self-assessment feedback failed to improve self-assessment accuracy in the absence of feedback, and learners' self-assessments did not track practice-driven performance gains — related
• Monitoring tools show low effects on self-regulated learning (d = 0.19) — related