Skip to content

Gamification

Principle · All principles
Evidence · 36 claims (16 for, 18 mixed, 2 against) · 43 studies (15 causal, 12 quant-synthesis, 10 review, 3 associational, 2 theoretical, 1 design), q1–q4 · 12 of 43 report an effect size · 28 claims rest on one study

Conditional relationship

This page is about adding a game layer to ordinary instruction: points or XP, levels, progress bars, badges, streaks, team scores, a theme or story, around practice the learner would otherwise do without them. For learners whose bottleneck is keeping going through practice they must repeat many times (vocabulary and other verbal-association targets, fact fluency, a procedure drilled to accuracy, coding exercises), a game layer added to instruction that already works is expected to produce a small gain in what learners can do afterwards (immediate-performance, general-achievement) over the same instruction without it. The expectation is conditional on three things: the layer rewards progress on the target (accuracy on new or harder items, improvement on the learner's own past) rather than mere activity or completion; it leaves learners meaningful choice; and its theme and decoration are tied to the content rather than added to entertain.

What gamification is usually adopted for is the least supported part. Gains in interest and enjoyment (motivation-affect) and in participation (persistence) were smaller in the one meta-analysis here and were not significant once the analysis was restricted to the more rigorous studies; the cognitive gain was. And two conditions can reverse the expectation: an expected tangible reward for doing or finishing a task that learners already find interesting can lower their later, unrewarded engagement with it; and interesting but irrelevant decoration can cost comprehension, most for learners with little prior knowledge or who find irrelevant material hard to ignore.

What tests this, and what does not. One claim tests gamified against non-gamified instruction on average, across 40 experiments in school, university, work and informal settings (Gamification has small positive effects on cognitive, motivational and behavioral learning outcomes [+M]). No claim tests the conditional part: whether mechanics that reward mastery beat those that reward activity, whether leaderboards help or harm, or which element (a streak, a badge, a story) carries an effect. Those conditions are carried from neighbouring evidence that was not gamified: laboratory reward experiments, cooperative-learning studies in schools, and studies of decorative material in texts. Each is labelled where it is used.

How the converted neighbours sit beside this page. Game-based Learning owns learning through a full game, where the game is the learning activity; its numbers and this page's do not transfer to each other. Autonomy owns the relationship between choice and motivation, which this page applies to the choices a game layer offers. Cooperative Learning and Social Interdependence own how group goals and individual accountability shape achievement, which this page uses for team scores. Engagement owns the distinction between behavioural, emotional and cognitive engagement, which a gamification evaluation needs. Immediate Feedback owns feedback timing. The building block itself is the Gamification element.

Default design, while the relationship is untested

The earlier page's guidance, kept as a concrete default a designer can act on and revise. Each step says what supports it.

  1. Start from practice that works without the game. Name the repeated practice to be gamified (items, attempts, exercises), check that it is sound instruction on its own (feedback after each attempt, items matched to the target), and add the layer only to that. Untested proposal; from the earlier page ("a motivational layer on top of sound instruction, not a replacement for it").
  2. Give points for accuracy and improvement, not for activity. Award points or XP for correct attempts on new or harder items and for beating the learner's own previous score; give none for logging in, time spent, or items completed regardless of accuracy. Evidence: in laboratory reward experiments, tangible rewards contingent on engaging in, completing, or performing a task lowered later free-choice engagement with it (d = −0.28 to −0.40), with children more affected than college students (Expected tangible rewards reduce later free-choice engagement with a rewarded task, while verbal praise does not [~S]); undermining was weaker where the task had a clear, well-defined standard of quality (Rewarding an already-intrinsically-motivating activity can reduce future engagement with it [~S]). Neither tests points or badges in a course, so treat this as carried evidence.
  3. Pair every point with an informational sentence. "Correct: you used the past tense because the action is finished" says what was done well; a bare "+10" does not. Word badges and messages about effort and strategy ("ten days of practice", "solved without hints"), not ability ("genius"). Evidence: positive verbal feedback raised later free-choice engagement (d = 0.33) in the same meta-analysis (rewards claim [+M], lab tasks); praise for intelligence after success left fifth graders less persistent and lower-scoring after a later failure than praise for effort (Praise for intelligence, rather than for effort, after success undermines children's motivation and performance after a later failure [~M], children, carried to badge wording).
  4. Make a level mean a skill passed. A level or badge unlocks when the learner meets a criterion on the target without hints, for example 8 of 10 correct on a short check (an untested proposal; set the criterion locally and record it). Show progress against the learner's own past (a personal best, a mastery map) as the default comparison. Untested; from the earlier page (self-referenced progress, as in Khan Academy's mastery levels).
  5. No public individual leaderboard by default; if scores are shared, share team scores built from each member's learning. Form small mixed teams; each member takes an individual quiz; the team score is the sum of members' improvement over their own previous scores, as in Teams-Games-Tournament or Student Teams-Achievement Divisions. Evidence: in school studies of four weeks or more, cooperative methods whose group rewards rested on every member's individual learning had a median effect of +.32 on achievement against +.07 without (In school studies, cooperative and peer-learning methods that reward groups on every member's individual learning raised achievement more than methods without such rewards [+M]); a review of Johnson and Johnson's meta-analysis reports cooperative learning ahead of competition and individual work (Meta-analyses by Johnson and Johnson find cooperative learning promotes higher achievement than competition or individual work across ages, subjects, and tasks [+M]). None of it was gamified, and no claim here compares a persistent public leaderboard with private progress. The one gamified competition study in the wiki qualifies this without lifting it: across eight low-stakes competitive Kahoot review rounds in one undergraduate course, students who disliked competition scored no lower in the game than those who liked it and enjoyed the rounds as much (Kahoot performance over a semester did not differ between high- and low-competitive undergraduates, though scores varied across time points [~W]; Students across competitive-preference levels reported enjoying the Kahoot activities consistently throughout the semester, with no detectable wear-out effect [~W]). It grouped students by competitive preference, not by rank, had no non-competitive comparison, and is associational, so it says a brief ungraded quiz round need not be avoided, not that a standing ranking is harmless. If a leaderboard is required, show each learner a small bracket of similar scorers or a team table, and let them opt out of display (untested proposal).
  6. Offer bounded choice. Two or three options at a time: which topic next, which of two challenge levels, which avatar or route through the material; hold the target fixed. Evidence: autonomy support with non-controlling feedback goes with higher intrinsic motivation (Self-determination theory holds that autonomy support raises intrinsic motivation, and a meta-analysis finds expected tangible rewards undermine it [+M], narrative review and the rewards meta-analysis, not gamified); in one 21-student study of a role-playing vocabulary app, perceived choice and interest/enjoyment were the lowest-rated motivation dimensions (Interest/enjoyment and perceived choice were the weakest motivation dimensions, suggesting Mangomon's RPG mechanics did not yet fully serve learner autonomy [~W]).
  7. Keep theme and story out of the moment of practice. Tie any theme to the content (the story's problems are the target problems); put narrative, animation and celebration between items or at the end of a level, not on the screen while the learner works an item. Evidence: interesting but irrelevant additions can impair comprehension and transfer, with effects that vary by material and learner; decorative pictures weakened the benefit of instructional pictures for low-prior-knowledge 7th and 8th graders, and harmed comprehension only in students with low attentional inhibition on harder texts (Interesting but irrelevant details can impair comprehension and transfer, with effects that vary with the material and the learner [~M]); added load matters most for complex material (Cognitive Overload Degrades Learning [~M]). Neither was tested on a gamified course.
  8. Plan the dose and the arc across sessions. All untested proposals; the only durations on the claim pages are a four-week app study, a semester of eight competitive quiz review rounds in which self-reported enjoyment stayed flat (enjoyment claim [~W]), and the meta-analysis's finding that duration did not moderate cognitive or behavioural outcomes.
  9. Session length: 10–20 minutes of gamified practice, three to five times a week, for at least four weeks.
  10. Week 1: short levels (about five items), points and an informational message after every item, an easy first level so every learner has an early success.
  11. Weeks 2–3: longer levels with mixed and harder items; points weighted toward harder and previously missed items; a level unlocks only after a check on a later day, not the same session.
  12. Week 4 onward: move points from every item to every level; shift attention to personal records and goals the learner sets; reduce celebratory effects.
  13. Ending the layer: before the course ends, run at least a week with the points hidden or removed and see whether practice continues; a drop then is the undermining or novelty risk showing itself, and a reason to change what points were given for.
  14. Moving on: a learner moves to the next level on the step 4 criterion, not on points accumulated.
  15. Judge the layer outside the game. Measure the capability on a task with no game around it, at the horizon the objective sets, and record enjoyment and participation separately. If possible, compare with the same practice ungamified for one group or one unit. Untested proposal; the meta-analysis's effects were heterogeneous, so a local check is what tells whether this layer worked.

When a condition is not met: if the practice underneath is not sound (no feedback, misaligned items), fix that first, because nothing here suggests a game layer compensates for it; if learners already find the task interesting and practise freely, use step 3 and step 4 only, without points or rewards.

Fitting the design to a situation

The three facts that most change the decision:

  • Whether learners already find the task interesting. Rewards for doing a task risk lowering later engagement most when interest is already there; a dull, necessary drill carries less of that risk. If unstated, ask: do learners do this task now without being made to, and do they say they enjoy it?
  • Which outcome must change. A capability on a later test, continued practice over weeks, and enjoyment are different outcomes, and the evidence is firmest for the first. If unstated, ask: what should learners be able to do afterwards, and is the real problem that they stop practising?
  • Who sees the scores. Private progress, team scores and a public individual ranking are three different designs with different risks for those ranked low. If unstated, ask: will anyone other than the learner see points, ranks or badges, and will they affect a grade?
If the brief says… Then change… Basis
Learners already find the task interesting, or practise it voluntarily Do not attach expected points, prizes or badges to doing or finishing it; give informational feedback and a progress view only Rewards claim [~S] and overjustification claim [~S] (lab tasks, free-choice time; carried to courses)
The task is a dull, necessary drill (fact fluency, verb forms, keyboard practice) Points for accuracy and improvement are reasonable; set a clear quality standard for earning them; plan the week without points (step 8) untested here: the rewards claim's Discussion argues there is little interest to undermine, but no recorded entry tests uninteresting tasks; the overjustification claim [~S] reports less undermining for well-defined tasks with a clear standard
Learners are children (under 12) Prefer verbal, informational feedback and effort-focused badges to tangible prizes; avoid prizes for completion; no public individual ranking Rewards claim [-S] (tangible rewards more detrimental for children than college students, lab); praise claim [~M] (fifth graders)
Learners are novices, or the material is complex Strip theme, animation and decoration from the screen while items are worked; celebrate only between levels Seductive details [~M] (school and university readers, not gamified); cognitive overload [~M]
Learners are advanced, or already accurate Let them test out of early levels on the step 4 check; give harder challenge levels rather than more points on easy items untested proposal
adults in a workplace-clinical setting (compliance, onboarding) Tie levels to job tasks passed at a criterion; no badges for completion or seat time; no public ranking among colleagues; keep the theme sober untested proposal; the earlier page's view that adults prefer progress views to badges has no claim behind it
Goal is verbal-association (second-language vocabulary, terms) A quiz or points layer is a reasonable vehicle; test recall a week after the last session, not only in the app Mangomon app, 21 Thai undergraduates [+W] (pre-post, no control); Quizizz [+W] and mixed-methods study [+W] (both known only through one review)
Goal is a concept, principle or complex-skill Award points for correct explanations and for solving changed problems, not for speed; avoid timed, speed-ranked quiz rounds untested; from the earlier page (timed leaderboard formats reward speed and suit fluency more than reasoning)
classroom with one teacher and a large group Use team scores from individual quizzes (step 5), updated once a week on a class board; the teacher reviews the board, not every point Group rewards on individual learning [+M] (school studies, four weeks or more, not gamified)
online-self-paced, no teacher present The software must give the informational message after each item (step 3) and run the level checks (step 4); if streaks are used, allow a missed day without losing the streak, so one absence does not end practice untested proposal; the meta-analysis pooled informal and online settings with school ones
Mixed modes: some learners remote, some in the room, or phone-only Run all scoring and feedback through one tool every learner reaches the same way; make teams mixed across room and remote; avoid speed-ranked live rounds, which penalise slow connections and small screens untested proposal
Scarce facilitator time (volunteers, a busy shift, a tutor who checks in briefly) Automate points, feedback and level checks; the facilitator spends about five minutes a week on the team or progress board and contacts learners whose practice stopped untested proposal
Time is a single-session Do not build a points economy; at most one team quiz round with feedback after each question and a short debrief; do not claim a motivational effect untested; no claim here tests a single gamified session
Learners have low confidence or low literacy No public ranking; self-referenced progress only; badges and messages short, with icons and audio; first level designed so everyone succeeds untested proposal; the earlier page's warning that leaderboards demotivate low-ranked learners has no claim behind it
Learners are uncomfortable choosing (some high power-distance settings, anxious novices) Offer two options with a recommended default rather than an open menu untested; noted in the autonomy claim [~W]'s Discussion with no study recorded
A live competitive quiz round (Kahoot-style) is wanted for review in a course Use it for low-stakes review: ungraded, feedback after each question, run across the term rather than once; do not read in-game scores as the capability, and check it with a separate quiz No performance gap by competitive preference [~W] and enjoyment stable over eight rounds [~W]; in-game scores predicted the first course quiz only for students who liked competition, and no later quiz (Kahoot–quiz relationship differed by competitiveness only on quiz 1 [~W]) (one undergraduate course, associational, no non-competitive comparison)
Points or badges feed a grade, or stakes are high Keep points formative and separate from the grade; grade on an aligned task outside the game untested; from the earlier page (effects weaker in high-stakes settings, no claim behind it)
The brief names its own routine (house points, a class reward chart, an existing app's XP or streaks) Keep it; change only what the evidence says matters: give points for accuracy and improvement rather than attendance or completion, add an informational sentence to each award, and base any team score on each member's learning Rewards claim [~S]; group rewards on individual learning [+M]; the combination is untested

Seven of the eighteen rows rest on a claim, and only two involve a gamified activity: the vocabulary row (a pre-post app study with no control and two second-hand reports) and the live-quiz row (one associational semester study in one course); the other five carry laboratory reward experiments, praise experiments with children, non-gamified cooperative learning in schools, or reading studies to this setting, and say so. Two more rows (the dull-drill and the choice rows) link a claim only for a limit or a note and are untested proposals; the remaining nine are untested proposals.

Observation, state and explanation

What can be observed: points and levels earned, accuracy per item, response times, sessions started and completed, practice after the layer is removed, a learner's choices, and performance on a task outside the game. "Is motivated", "has mastered it" and "is gaming the system" are inferred states. Points are an observation about the game; they say something about the capability only if they are earned for accuracy on items the learner has not already mastered.

Response Plausible explanations Observation that could change the interpretation
Points and levels rise; the outside test does not Points can be earned without the target (repeating easy items, guessing); the items are misaligned with the test; the gain is in the app's format only; in-game scores track the test for some learners and not others (in one course, Kahoot scores predicted the first quiz only for students who liked competition, and no later quiz: quiz 1 claim [~W]) Correlate points with accuracy on new items; check what share of points came from already-mastered items; give the test once in the app's format and once outside it.
Practice drops when points are removed, or after a few weeks Rewards displaced interest the learner had; novelty wore off; the material became too hard or too easy Compare practice before, during and after the layer; check whether the drop began before removal; ask what the learner did instead.
Low-ranked learners stop practising Public comparison discourages them; the items are too hard; they never valued the task Offer a private, self-referenced view and an easier entry level, separately, and see which changes practice; note where they stop.
Fast answers with falling accuracy Speed or volume is what earns points; the learner is chasing a streak; the items are too easy to need thought Compare response time with accuracy; remove the speed element for one round and watch accuracy.
A learner ignores the game layer but practises steadily Values the task, not the game; finds the theme childish; the layer adds nothing for them Ask; offer a plain mode; check the outside test, which is what counts.

These rows are proposals, untested as diagnostics. Removing or changing a mechanic to test an explanation is itself an intervention, so record each change and its date.

Evidence and qualifications

Learners, tasks, settings and outcomes differ across these claims (gamified courses pooled across settings, laboratory puzzles, school cooperative learning, reading texts), so do not rank them by effect label. None follows a gamified course after the layer is removed.

The evidence by question

Further evidence, not yet read against this model

Claims read against this model that do not change it. Each says what it is and why it stays here.

  • Belonging Interventions Improve Outcomes [+M] — (not about this page) two randomised brief belonging interventions at entry to college (92 students; 26,911 students at 22 institutions), which reframe doubts about belonging; the earlier page cited it for team and community mechanics, but neither study involves game mechanics.
  • Game-based vocabulary practice produced larger gains than traditional instruction for newly arrived migrant children. [+W] — 48 Ukrainian children aged 6–7, Kahoot! and outdoor games against traditional teaching over four weeks (171.9 against 139.3 words, no standardised effect size); confounds format, novelty, competition and being outdoors, and is a game used as the activity rather than a layer added to practice. Belongs mainly to Game-based Learning; too confounded to support the child row.
  • Planned contingency-of-reinforcement designs can raise academic achievement above normal classroom levels across a wide range of children [+W] — a 1970 narrative survey of behaviour-modification studies, no effect sizes; the behaviourist side of the reward question. It does not separate rewards for accuracy from rewards for activity and does not test later free-choice engagement, so it does not change step 2.
  • Most of the 45 reviewed studies report positive gamification effects on motivation and engagement in mathematics; only four report negative effects [+W] — attached 2026-10-21 from Ratinho et al. (2026), which proposed "Align gamification design with students' psychological needs, limiting excessive competition and prioritizing personal progress and cooperative tasks".

Objective and learner-valued goal

The designer's objective is usually a capability (vocabulary recalled a week later, a procedure done accurately) and, often, continued practice. Elicit separately what the learner wants from the layer and from the practice: to win, to keep a streak, to see progress, to avoid being ranked, to get the practice over with, or to be able to use the skill. Agreement and divergence are both observations. A learner who values the streak may log in daily for one easy item; one who values not being ranked may stop when a leaderboard appears; one who values the skill may find points childish and still practise well.

For example, an adult learning Italian before a move may want to hold a conversation with neighbours, while the designer's objective is 300 high-frequency words at 80% recall after a week. Points for correct recall serve the designer's objective; a weekly level that ends with a short spoken exchange using the week's words serves the learner's. Record both, so that a long streak is not read as the capability the learner came for.

What would revise this model?

The expectation should weaken if controlled comparisons of the same practice with and without a game layer, on outcomes measured outside the game, show no cognitive gain, or a gain that disappears once the layer is removed. The alignment condition should be dropped if mechanics rewarding activity do as well as those rewarding accuracy and improvement. The caution about rewards should weaken if gamified courses, unlike the laboratory tasks, show no fall in practice after points are removed, and strengthen if they do. If public leaderboards are tested and do not harm low-ranked learners, the default against them is a preference, not a finding. Do not protect the model by calling every failure "poorly aligned mechanics" after the fact; record which mechanics were used before the outcome is known.

Design Decisions

What should a gamification layer be expected to change?

What should points, badges and rewards be given for?

Individual competition or team-based mechanics?

How much choice should the mechanics give learners?

How much decoration, story and theme?

Is it worth gamifying second-language vocabulary practice?

How long can it run before the effect fades?

Claims

  • Autonomy — gamification works when it expands meaningful choice, not when it controls learners through rewards
  • Assessment for Learning — badges and progress indicators are most valuable when they function as low-stakes feedback on mastery
  • Adaptive Learning — level structures and difficulty curves depend on matching challenge to current ability
  • Active Learning — game mechanics amplify engagement with the activity itself; they cannot compensate for passive content
  • Game-based Learning — learning through a full game, where the game is the activity; this page is about a layer added to ordinary instruction
  • Cooperative Learning and Social Interdependence — group goals and individual accountability, the basis for team scores
  • Engagement — behavioural, emotional and cognitive engagement, which a gamification evaluation should record separately
  • Immediate Feedback — feedback timing, which points and messages inherit

Examples

Illustrative

Duolingo — Language-learning app built on streaks, XP, leagues, and hearts mechanics layered over spaced practice. Its design team publishes research on how streaks and leagues drive retention; the mastery-aligned structure (skills unlock in sequence) exemplifies mechanics tied to learning progression.

Kahoot! — Classroom quiz game with timed multiple-choice rounds, music, and live leaderboards. Best used for retrieval practice and formative review; the timed leaderboard format rewards speed, so it suits fluency goals more than deep reasoning.

Khan Academy — Points, badges, and mastery levels tied to mastery-based practice in math and other subjects; progress indicators are self-referenced rather than competitive, illustrating the safer comparison structure.

Classcraft — A classroom management layer that turns coursework into a persistent team-based role-playing game, with teams earning powers through academic and behavioral goals — an example of collaborative (rather than individual-competitive) mechanics.

Zombies, Run! — Audio narrative gamification of exercise; a model of narrative framing that transforms a repetitive activity rather than bolting rewards onto it.

Key Sources

  • Deterding, S., Dixon, D., Khaled, R., & Nacke, L. (2011). From game design elements to gamefulness: Defining "gamification." Proceedings of the 15th International Academic MindTrek Conference: Envisioning Future Media Environments, 9–15. doi:10.1145/2181037.2181040
  • Hamari, J., Koivisto, J., & Sarsa, H. (2014). Does gamification work? A literature review of empirical studies on gamification. Proceedings of the 47th Hawaii International Conference on System Sciences, 3025–3034. doi:10.1109/HICSS.2014.377
  • Sailer, M., & Homner, L. (2020). The gamification of learning: A meta-analysis. Educational Psychology Review, 32(1), 77–112. doi:10.1007/s10648-019-09498-w
  • Landers, R. N. (2014). Developing a theory of gamified learning: Linking serious games and gamification of learning. Simulation & Gaming, 45(6), 752–768. doi:10.1177/1046878114563660
  • Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125(6), 627–668. doi:10.1037/0033-2909.125.6.627
  • DiFrancesca, D., & Spencer, D. (2025). The effects of competitive classroom activities on undergraduates with varied levels of competitive preference. Instructional Science. doi:10.1007/s11251-025-09741-8
Page metadata
Field Value
type principle
id gamification
title Gamification
description For learners whose bottleneck is keeping going through repeated practice, a game layer (points, levels, badges, team scores, a theme) added to sound instruction is expected to give a small gain in later performance when it rewards accuracy and improvement rather than activity, leaves meaningful choice and keeps decoration tied to content; one meta-analysis tests gamification on average, and no claim tests these conditions.
status review
canonical True
generated by claude/unspecified on 2026-10-19
trust unverified — no one has confirmed this page's content

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