In short
- Pooled research finds no reliable overall effect of competition on performance: it helps people who aim to win and harms people who aim to avoid losing.
- Grouping by reading level, not by age or class alone, gives every pupil rivals of similar ability and a top place within reach.
- Score reading completed and accuracy on a pupil's own levelled text, so a lower-level reader can outscore a higher-level one.
- Publish team totals and top movers, keep individual standing private, and remember that level-matched reading competitions have rarely been tested directly.
If a reading competition ranks a whole year group on one list, the strongest readers start ahead. A pupil who reads two levels below them has little chance of catching up, however hard they work. The answer is not to drop competition. It is to change who competes against whom, and what earns points. This article reports what research found about competition, social comparison and leaderboards. Then it turns those findings into a design you can run.
What makes a reading competition fair?
A reading competition is fair when every pupil who takes part has a realistic chance of winning or moving up. That means grouping pupils by reading level, not by age or class alone. It also means scoring what each pupil does at their own level: the reading they complete and their accuracy on their own text. Under those rules, a lower-level reader who reads carefully every day can outscore a higher-level reader who does not.
Fair does not mean that everyone wins. It means that effort decides the result, not the level a pupil started at.
What research says about competition and performance
Competition is neither good nor bad for performance in general. Murayama and Elliot (2012) pooled the existing studies in a meta-analysis. They found no noteworthy overall relation between competition and performance.
Their explanation is that competition does two opposite things. It leads some people to aim at doing better than others, and that goal helps performance. It leads other people to aim at not doing worse than others, and that goal harms performance. In the average, the two cancel out. A second meta-analysis and three new studies supported this account.
This was not a study of reading competitions. It does give you a question to ask of any design. Does it lead most pupils to try to win, or to try not to come last?
Bandura (1997) offers a reason why pupils may differ. He argued that how much effort people give, and how long they persist, depends on whether they believe they can succeed. He named a person's own experience of success as one source of that belief. By that reasoning, a pupil who has never finished near the top has little ground for expecting to.
Why the gap to the top matters
Garcia, Tor and Schiff (2013) reviewed research on social comparison, the habit of judging ourselves against others, as a source of competitive behaviour. They set out the conditions under which comparison and rivalry grow stronger:
- The task matters to the person.
- The rival is similar in ability.
- The competitors stand near a standard, such as first place, and not far away from it.
- There are few competitors, not many.
Read these as design rules. A pupil ranked 140th in a year group of 150 is far from first place, among many competitors, with no close rival that matters. A pupil ranked fourth in a league of eight similar readers meets all four conditions. The review draws on research from many settings, not on school reading competitions, so treat this as an inference.
Mixed groups may carry a second cost. Marsh (1987) described the big-fish-little-pond effect. Equally able students had lower academic self-concepts in high-ability schools than in low-ability schools. Students judged themselves against the group around them. In the same analysis, academic self-concept had a direct effect on later school performance, beyond ability and earlier results. His data came from a large existing survey, the Youth in Transition study, not from reading classes.
The likely lesson is this. An A2 reader ranked on the same list as B2 readers is told every day that they are a poor reader. They are not. They are a reader at an earlier stage.
What classroom leaderboard studies found
Hanus and Fox (2015) followed university students on two courses through a 16-week semester. One course used a leaderboard and badges. The other had the same curriculum without them. Students on the leaderboard course showed less motivation, satisfaction and empowerment over time. They also scored lower on the final exam, and the lower scores were explained by their lower intrinsic motivation.
That is one study of university students on two courses. It cannot settle the matter for school pupils. It does show that a ranking can cost more than it gives.
The wider picture is more positive. Sailer and Homner (2020) pooled studies of gamification in a meta-analysis. They found small positive effects on learning, on motivation and on behaviour. The effects on motivation and behaviour were less stable when only the more rigorous studies were counted. Combining competition with collaboration was particularly effective for behavioural outcomes. The authors add that the factors behind successful gamification are still somewhat unresolved.
Where the evidence is thin
We found no study that compares a level-matched reading competition with an unmatched one for EFL pupils aged 8 to 16. The case for level matching rests on general research into competition, comparison and self-belief, applied by reasoning. The practical advice below has not been tested as a package, so check it against your own pupils' reading records.
Ways to match competitors
There are three main ways to match pupils: by level band, by starting point and by improvement. A fourth balances teams. The table compares them with the usual default.
| Method | How it works | What it needs | Where it fails |
|---|---|---|---|
| By level band | Pupils compete only against pupils at the same reading level, across classes and ages. | A reading level for every pupil, and enough pupils in each band to form a league. | Small schools may have bands of three or four pupils. Publicly labelled bands can expose older pupils at low levels. |
| By starting point | Each pupil gets a personal target set from their own first weeks. The score is the share of the target reached. | Two or three weeks of baseline reading records. | A pupil can hold back during the baseline to earn an easy target. Targets go stale and need resetting each term. |
| By improvement | Pupils are ranked by gain over their own last period: the top movers. | Comparable records from one period to the next. | Strong, steady readers have little room to rise. A bad week followed by a normal one looks like progress. |
| Mixed-level teams | Teams are balanced so each has a similar spread of levels. Members score at their own level and the team total counts. | Levels for every pupil, and teacher time to balance the teams. | If raw output counts, weaker members get blamed. If only the total is shown, some pupils can coast. |
| By age or class alone | Everyone in the class or year is ranked on one list. | Nothing. | The gap between levels is at its widest, so the starting level decides the result. |
Level bands are the base. Starting points and improvement scores work best inside a band, where they separate pupils by effort. Team formats suit contests between classes or houses; see house reading competitions. For contests across schools, match whole schools by level, as the guide to running a competition between schools explains.
What to score so that every level can win
Matching decides who competes. Scoring decides what counts. Aim for a score that a pupil earns by reading well at their own level, so that the level itself gives no advantage.
- Reading done at your own level. Count stories or units completed at the pupil's assigned level. Do not count pages or words, because higher-level texts tend to be longer.
- Accuracy on your own text. Use the quiz score on the story the pupil actually read. An A2 pupil answering questions on an A2 story can reach full marks.
- Regularity. Count the days on which the pupil read. Every pupil has the same seven days in a week.
- No credit for reading down. Count only texts at or above the assigned level, so nobody gains by choosing easy books.
- Speed with care. If you score speed, score it together with accuracy, and only among pupils reading the same text.
Classroom example (an illustration, not a research result)
Lina and Omar are both 11 and in the same class. Lina reads at A2. Omar reads at B1.
The rule: one point for each story finished at your own level, plus up to one point per story for quiz accuracy.
Lina finishes four A2 stories and averages 90% on the quizzes: 4 + 3.6 = 7.6 points.
Omar finishes two B1 stories and averages 80% on the quizzes: 2 + 1.6 = 3.6 points.
Lina wins the week. Under a pages-read rule Omar might have won, because his stories are longer.
How to show results without shaming
Display matters too. Hanus and Fox (2015) studied a course built around a leaderboard and badges, and motivation fell. The options below keep the contest visible and protect individual pupils.
- Publish team totals. Class, house or school scores move the comparison from pupils to groups. Sailer and Homner (2020) found competition combined with collaboration particularly effective for behavioural outcomes.
- Name the top movers. A weekly list of the pupils who improved most is a list that lower-level readers can top.
- Keep individual standing private. Let each pupil see their own position and trend. Do not pin a full ranked list to the wall.
- Show the top, never the bottom. If you display names, show the first three in each league and stop there.
- Reset often. Start a new round every few weeks, so a slow start does not decide the whole term.
Certificates for specific achievements can do some of this work with no ranking at all. The article on whether reading certificates motivate students covers that evidence.
How to set levels in the first place
Level matching is only as good as the levels. The CEFR Companion Volume (Council of Europe, 2020) describes reading ability in bands from Pre-A1 to C2, with descriptors for overall reading comprehension. It also defines plus levels, such as A2+ and B1+, for finer steps.
- Place every pupil with a reading measure, not with last year's grade. See how to assess a student's English reading level.
- Confirm the level with a real text. A pupil placed at A2 should read an A2 story with good understanding. See matching students to graded readers.
- Form leagues from the bands. Use plus levels only where you have enough pupils to fill them.
- Give leagues neutral names, so the name does not announce the level.
- Review levels each term. Move a pupil up when their accuracy at the current level stays high.
Expect some wrong placements. A pupil who wins every round by a wide margin is probably in the wrong league. Treat that as information and move them up.
How iRead handles this
In iRead, every story, quiz and game is levelled to the CEFR. In the daily run, students on the same global pack read the same levelled story and play the same daily game, scored for accuracy and speed. Reading is the only way to score. Schools are matched only against schools at the same CEFR level. Public leaderboards show school-level results only, and a student sees their own standing privately inside the app.
The core principle: Match pupils by reading level, score what they do at that level, and keep individual standing private. Then effort decides who wins.
See it in iRead: Schools are matched only against schools at the same CEFR level. Public leaderboards show school-level results only, and each student sees their own standing privately inside the app.
See the reading competitionKey takeaway
A fair reading competition gives every pupil a rival they can beat and a score they can earn. Set levels first, then choose the matching method and the scoring.
Frequently asked questions
How do you make a reading competition fair?
Group pupils by reading level, not by age or class alone. Score the reading each pupil completes and their quiz accuracy on their own levelled text. Publish team totals and top movers, and keep individual standing private. Then a careful lower-level reader can outscore a stronger reader who does less.
What rules should a reading competition have?
State who competes against whom, what earns points and how results are shown. Useful rules: only texts at or above your assigned level count, and quiz accuracy counts alongside stories finished. Scores reset every few weeks. No full ranked list of individual pupils is displayed.
Is competition in the classroom good or bad for learning?
Neither, on average. Murayama and Elliot (2012) found no noteworthy overall relation between competition and performance. Competition helped when it led people to aim at doing better than others. It harmed when it led them to aim at not doing worse. A design should encourage the first goal.
Should I display a reading leaderboard in class?
Be careful with individual rankings. Hanus and Fox (2015) found that university students on a course with a leaderboard and badges lost motivation over a semester and scored lower on the final exam. Team totals and top-mover lists are safer. Let pupils see their own standing privately.
Sources
- Murayama, K., & Elliot, A. J. (2012). The competition-performance relation: A meta-analytic review and test of the opposing processes model of competition and performance. Psychological Bulletin, 138(6), 1035-1070. doi.org/10.1037/a0028324
- Garcia, S. M., Tor, A., & Schiff, T. M. (2013). The psychology of competition: A social comparison perspective. Perspectives on Psychological Science, 8(6), 634-650. doi.org/10.1177/1745691613504114
- Marsh, H. W. (1987). The big-fish-little-pond effect on academic self-concept. Journal of Educational Psychology, 79(3), 280-295. doi.org/10.1037/0022-0663.79.3.280
- Hanus, M. D., & Fox, J. (2015). Assessing the effects of gamification in the classroom: A longitudinal study on intrinsic motivation, social comparison, satisfaction, effort, and academic performance. Computers & Education, 80, 152-161. doi.org/10.1016/j.compedu.2014.08.019
- Sailer, M., & Homner, L. (2020). The gamification of learning: A meta-analysis. Educational Psychology Review, 32(1), 77-112. doi.org/10.1007/s10648-019-09498-w
- Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman. archive.org/details/selfefficacyexer0000band
- Council of Europe (2020). Common European Framework of Reference for Languages: Learning, teaching, assessment. Companion volume. Strasbourg: Council of Europe Publishing. rm.coe.int/common-european-framework-of-reference-for-languages-learning-teaching/16809ea0d4