Mastery Learning

Pattern · All patterns
Evidence · 28 claims (13 for, 14 mixed, 1 against) · 25 studies (7 causal, 7 quant-synthesis, 3 review, 2 associational, 2 qualitative, 2 design, 2 theoretical), q1–q4 · 7 of 25 report an effect size · 24 claims rest on one study

Description and scope

A reusable policy for organising cumulative units around a stated criterion: each unit ends in a check, a response below the criterion leads to corrective work and a further, comparable check, and advancement depends on that response rather than on the calendar alone. It instantiates the conditional principle. The programme configurations in the studies below are evidence; the response-dependent policy on this page is an untested design proposal. Use it to collect better evidence about learners' responses, not to certify that a learner "has mastered" a unit.

Inputs: establish the design brief

Field	Record or elicit
Learner and starting observation	Responses on the first unit check and on any prerequisite probe; attempt history; errors and their kind. Use expertise, age and population_band values where known, and record unknowns as unknown.
Context and activities	Setting (classroom, online-self-paced, online-instructor-led, workplace-clinical), unit length, who delivers correction (teacher, peer, software), time available for reattempts, aids permitted on checks, and what learners who pass do meanwhile.
Learner-valued goal	Ask what the learner wants from the course and on what timescale. Record divergence from the unit criterion, and the learner's tolerance for extra time, as observations.
Designer objective	The capability each unit's check stands for, its knowledge type (procedure, principle, complex-skill, ...), which later units depend on it, and the cut score with its justification.
Outcome	Unit-check performance, the course or programme examination (general-achievement), delayed retention or transfer if targeted, completion (persistence) and time to criterion (learning-time), each with its instrument and horizon.

A request to "make sure students master the material" does not specify these. Ask for them before prescribing a cut score, a number of attempts or an expected gain. Check that the unit check samples the capability the next unit needs and that a parallel form exists before relying on a recheck.

Sequence and conditional policy

1. State the criterion and the check. Name the capability, the representative task, the cut score, permitted aids, and a parallel form for rechecks. Tell learners what the criterion is.
2. Teach and practise the unit. Record what instruction and practice each learner actually received.
3. Elicit a unit-check response. Score it against the criterion and record the reasoning or error, not only the score.
4. Interpret a shortfall provisionally. Use the table below to choose a probe that separates explanations; record which remain.
5. Correct responsively. Change the form of instruction to address what the response showed, rather than replaying the unit. Record the correction delivered and its time.
6. Recheck on a parallel task. Change surface features and at least one relevant condition. Record the attempt number.
7. Decide on advancement, with the time limit explicit. Advance on criterion, or by the agreed rule when the attempt or time limit is reached, recording any unresolved gap so the next unit's first task can re-probe it.
8. Reobserve later. Check the earlier capability at the start of the dependent unit or at the stated delayed horizon.

Observation	Discriminating probe	Proposed action, with its uncertainty
Below criterion on the first check	A short prerequisite probe set apart from the unit task; one item on the same relation in another representation; ask the learner to explain one wrong step.	If the prerequisite fails, correct it first; if only the unit relation fails, correct that with a different explanation or example; if only one representation fails, address access or representation. None of these is a validated diagnosis.
Passes only on a recheck that repeats items	Give a parallel form with a changed relevant condition.	Treat item recognition as a live explanation until the parallel form is passed; do not count a repeated-item pass as meeting the criterion.
Passes the gate, then fails the next unit's first task	Compare the failed step with what the previous check sampled; re-probe the earlier capability.	Widen the previous check or add a bridging task; a narrow check, forgetting and a genuinely new demand each lead to a different action, and one observation does not separate them.
Repeated failures after correction	Change the correction's form and observe; probe an earlier prerequisite; ask about purpose and barriers.	Offer a different route (one-to-one or small-group help, a worked example, a smaller step) or revisit the cut score with evidence from other learners; do not keep cycling the same material.
Falling participation or time running out	Record time and attempts per unit; ask what the learner wants from the course.	Negotiate the limit and the advancement rule in the open; a capped number of attempts with a recorded gap may serve the learner's goal better than an unbounded gate.

Choosing configurations from evidence

• Whether to gate at all. Mastery learning programmes, 108 controlled evaluations [+M] report positive effects on examination performance against conventionally paced instruction at college, high-school and upper-elementary levels, apparently stronger for weaker students and varying with the mastery procedure, design and content. The same synthesis reports possibly more time on instructional tasks and, for self-paced college programmes, often reduced completion. Abstract only; no pooled size recorded. It supports a gate-and-correct configuration as a candidate, and puts time and completion in the brief as outcomes to watch, not side issues.
• What the correction should be. Contingent versus fixed support [+M]: in one-to-one long-division tutoring with fourth and fifth graders (n = 8 per condition), support adjusted to each response beat fixed or no support at immediate and one-month tests; a dynamic-assessment synthesis ranks explicit strategy training above contingent scaffolding. Choose correction that responds to the response, and consider explicit strategy instruction as an alternative; the one-to-one evidence does not show this works in a whole class. Not settled against its sources.
• How the check feeds action. Formative assessment syntheses [+M] report small positive overall effects in K-12 (d = .29; .20 in another synthesis), with larger estimates where self-assessment was supported. These are configuration comparisons across studies with no common horizon; they do not predict an effect for a new course.
• Software-delivered pacing. Intelligent tutoring meta-analyses [~M] associate tutoring systems with higher achievement than teacher-led large-group instruction, other computer-based instruction and textbooks, but report no advantage over individual human tutoring or small-group instruction, and gains that depend on whether tests were locally developed or standardised. If a system delivers the gate, compare it with human small-group correction, not only with whole-class pacing, and measure on an instrument independent of the system. The claim page words one entry causally ("produced") where its own description says "associated with"; read it as an association.
• Section-level comparisons. Earned credit in mastery and non-mastery sections [~W]: one college system's records favoured mastery sections in eight of nine comparisons, but students taking one mastery class earned less credit than those taking none, attributed to course selection. Retrospective and uncontrolled; do not use a local section comparison as evidence of the gate's effect without accounting for who enrolled.
• Whether the check means what the criterion says. Criterion-referenced tests need construct validation [~W] is a theoretical argument (checked by the judge on full text) that a criterion-referenced test's tasks must be shown empirically to reflect the competence of interest. The gate is only as good as that evidence.

Do not combine these into one strength, rank the syntheses by effect labels while their comparators differ, or read "positive effects" as a size. A precise abstention names the missing comparison, instrument or horizon and how to obtain it.

Fitting the design to a situation

Ask for four facts when the brief omits them: the unit criterion and how it is checked, group-paced or self-paced progression, the limit on attempts or time per unit, and the delay at which the earlier capability will be rechecked.

If the brief says…	Then change…	Basis
A course of cumulative units, with the course examination as the outcome	Use the full loop: unit check, corrective work, parallel recheck, then advance	Mastery learning improves outcomes [+M]; Group-based mastery positive in every included study [+M]
Choosing the cut score	Set a high standard for each unit; record it and tell learners	Effects vary with standard, pacing and test type [~M]
College course, self-paced (online-self-paced or Keller-style)	Add deadlines or pacing checkpoints, because self-paced mastery courses lose more students	Self-paced mastery programmes reduce course completion [-M]
Group-paced class with a teacher	Keep the class together and run correction and enrichment in a shared loop; attrition was lower in group-based mastery classes	Mastery improves engagement and lowers attrition [+W]
Software will gate progress (adaptive or practice platform)	Expect a gain against large-group teaching but not against human tutoring; check the gate on outcomes outside the platform	Intelligent tutoring systems improve achievement over non-individualised instruction [~M]; Mastery platform longitudinal positive effects [+W]
Learners may need many attempts on a unit	Set an attempt limit and change the correction after a failed recheck instead of repeating the same task; record any gap left at the limit	eSpark learners rarely reached 80% after three attempts [~W]; Socratic condition needed twice the attempts [~W]
Retention after the course matters	Recheck each earlier unit at the start of the next one and again weeks later	Gains persist at follow-up [+M]; Group-based mastery improves retention [+M]
Learners' attitudes or attrition are a concern	Mastery classes rated the instruction and the subject more favourably; tell learners what the criterion is and why	Mastery learning improves attitudes [+M]; Positive affective effects [+M]
The check will be a quiz written by the teacher	Write a parallel form for rechecks that changes surface details and at least one relevant condition	Criterion-referenced tests need construct validation [~W]
A district pacing calendar fixes when topics are taught	Agree in advance how far the loop may extend a unit, and what gap is carried forward when time runs out	Pacing calendars seen as impediments to mastery [~W]

The evidence by question

• Does making progress depend on reaching a unit criterion, with correction and a recheck, raise examination performance over conventional pacing? Mastery Learning Improves Outcomes [+M]
• Which variant: how high a criterion, group-paced or self-paced, and measured how? Mastery learning effects are larger with a higher mastery standard, group-based pacing, local tests, and less feedback for the comparison group [~M]
• What does it cost in time and coverage, and what happens when learners do not reach the criterion? Mastery learning implementation did not reduce material covered or instructional pace, contradicting the Robin Hood contention [~M]

Further evidence, not yet read against this model

Claims this page cited before it was rewritten as a conditional model. They are evidence about the relationship, but none has yet been re-read against the model above; each says how far its own sources have been checked.

• Self-regulated learning models treat self-monitoring as a core process through which learners notice gaps and adjust strategy [+M] — not settled: the text available could not confirm the entries (abstract)
• Mastery and control groups show no significant differences in entry knowledge, academic self-concept, or affect toward education [~W]
• Female students show stronger academic self-confidence, more positive attitudes, higher achievement, and fewer absences, but this advantage diminishes almost completely under mastery learning [~W]

Illustrative design instance and observation record

Illustration, invented for this page; not a tested course. An introductory statistics course for part-time adult learners (adult, postsecondary, online-instructor-led) has six cumulative units. A learner says they want to finish within the term because their employer funds one term only; the designer's objective is criterion performance on each unit. They agree a cut score for each unit check, two rechecks on parallel forms, and a rule that a learner still below criterion after the second recheck advances with the gap recorded and a short re-probe at the start of the next unit. On unit 2 (sampling distributions) the learner scores below criterion; a prerequisite probe on unit 1 (variability) is passed, and the learner explains the wrong step by confusing the spread of a sample with the spread of sample means. The tutor corrects with a simulation-based demonstration rather than replaying the unit's video; the learner passes the parallel recheck, and the start-of-unit-3 re-probe is recorded.

Record: unit, criterion and check form → instruction and practice received → response and reasoning → candidate interpretations and the probe chosen → correction delivered and its time → recheck form, attempt number and response → advancement decision and rule applied → re-probe at the next unit or stated horizon, alongside the learner's stated goal and any change in it. An agent may summarise this record but must not invent missing inputs or assign a probability that the learner has "mastered" the unit without a calibrated model.

Elements and limits

Formative assessment, feedback, reassessment, adaptive mastery learning, rubric and practice.

This pattern is scoped to cumulative units with a definable check. Performances that develop continuously without a single threshold, open-ended projects, and programme-level competency certification need other configurations and evidence. The evidence above concerns programmes and averages; whether this page's diagnostic branches improve decisions for individual learners, and what cut score or attempt limit serves a given goal, remain to be tested with learners.

Claims

• The weekly closed-book gate was attainable: 80.1% of recorded student-test observations reached the 100-point credit cap [+W]
• Teachers viewed district pacing calendars and standardized assessments as impediments when they pressured rushing through material regardless of mastery [~W]
• Successful use of mastery learning has powerful but mixed effects on teacher variables, including more positive attitudes, altered attributions, and reduced teaching confidence [~W]
• Achievement effects of group-based mastery learning are larger for elementary students than for high school or college students [~M]
• LMS process indicators documented adaptive engagement during the eight-week intervention, with most experimental students reaching mastery by unit end [+W]
• Mastery learning has overall positive effects on student affect, with the exception of slightly lower grade expectations [+W]
• Group-based mastery learning improves retention of learned material, with an average effect size of .62 across four studies [+W]
• Mastery learning effects are larger with a higher mastery standard, group-based pacing, local tests, and less feedback for the comparison group [~M]
• Mastery learning gains persist at follow-up examinations about eight weeks after instruction [+M]
• Mastery learning effects are larger for lower-aptitude students, narrowing the spread of achievement in a class [~M]
• Self-paced mastery programs (PSI) reduce college course completion, though not by enough to explain their examination advantage [-M]
• Mastery learning improves students' attitudes toward the instruction and toward the subject [+M]
• Recorded natural-language feedback was associated with lower odds of subsequent persistent failure than no feedback in the randomized course subset [~W]

Related Patterns

• Game-Based Mastery Learning (e.g., Duolingo Pattern)

Examples

• A unit where learners receive targeted reteaching and then reassess until they meet the rubric threshold.
• Increase the use of feedback and correctives through mastery learning procedures
• Mastery Learning

Key Sources

• Bloom, B. S. (1971). Mastery learning. In J. H. Block (Ed.), Mastery learning: Theory and practice. Holt, Rinehart and Winston.
• Kulik, C. L. C., Kulik, J. A., & Bangert-Drowns, R. L. (1990). Effectiveness of mastery learning programs: A meta-analysis. Review of Educational Research, 60(2), 265-299. https://doi.org/10.3102/00346543060002265
• Bonczar, Thomas P., & Easton, John Q. (1983). The Effect of Mastery Learning on Student Achievement. https://eric.ed.gov/?id=ED238505
• Guskey, Thomas R., et al. (1982). The Effectiveness of Mastery Learning Strategies in Undergraduate Education Courses. Paper presented at the Annual Meeting of the American Educational Research Association, New York. https://eric.ed.gov/?id=ED213702

Page metadata

Field	Value
type	pattern
id	mastery-learning
aliases	mastery-learning-formative-corrective-cycle
title	Mastery Learning
description	A reusable gate-correct-recheck policy for cumulative units: elicit a criterion-referenced response, interpret a shortfall, give responsive correction and recheck on a parallel task before advancing, with time, attempts and the outcome horizon stated.
status	review
grain_size	unit
author	Bloom / mastery learning tradition
moved_from	principle/superimpose-mastery-learning-on-lecture-courses
canonical	True
generated	by claude/unspecified on 2026-10-01
trust	unverified — no one has confirmed this page's content
sources	3 — kulik-1990, bonczar-1983, guskey-1982

Frontmatter is the page's machine-readable record — see the Schema & Guide.