Using Gamification to Enhance Online Learning: A Review of the Literature

A note on what this review is

This article describes, in general terms, the kinds of findings that published research on gamification in online learning tends to report. It does not cite individual studies, authors or figures, because the aim is to help you read the research yourself, not to substitute for it. Where it says “studies often find”, that is a broad characterization and not a count of papers. For any specific claim, look up the original work and read its method section.

Gamification here means the use of game elements, such as points, badges, levels, leaderboards and progress bars, in a non-game setting. Whole games used for learning are a related but separate topic.

The overall picture: mixed and context-dependent

Researchers have reviewed the field repeatedly. The common thread in such reviews is that results are mixed. Many studies report positive effects on measures such as participation, time spent or self-reported enjoyment. Fewer, and with less consistency, report improvements in measured learning. Some find no difference, and some find negative effects for particular groups or settings.

A frequent conclusion is that the effects depend on context: the subject, the age and background of learners, how the elements are designed, how long the study ran, and what was measured. That makes simple statements such as “gamification works” or “gamification does not work” poorly supported.

Outcomes that tend to be measured

It helps to separate several kinds of outcome, since gamification can plausibly affect some more than others.

• Participation and activity. Counts of logins, posts, completed tasks. These are easiest to measure and the most likely to rise when rewards are attached to them. A rise in activity is not the same as a rise in learning.
• Engagement and enjoyment. Often measured by questionnaires. Self-reports can be influenced by novelty and by what participants think the researchers want to hear.
• Learning outcomes. Test scores or performance tasks. These are the most meaningful, but results are the least consistent.
• Retention and completion. Whether learners continue. Evidence here is limited and varied.
• Attitude and motivation. Often measured with scales. Interpretation requires care, because different studies use different instruments.

Recurring themes in the findings

Novelty effects. Several studies and reviews note that gains observed early can fade as the new features become familiar. Short studies may therefore overstate long-term benefits. A study lasting a few weeks says little about a year-long course.

Individual differences. Some learners respond positively to competition and rewards; others are indifferent or put off. Average results can hide opposite reactions in different subgroups. Studies that look at differences between learners tend to find them.

Competition is a double-edged element. Leaderboards are commonly reported to motivate some participants and discourage others, particularly those lower down the ranking.

Reward type matters. Research on motivation more broadly suggests that rewards perceived as informational, giving feedback on competence, tend to be better received than those that feel controlling. How this plays out in specific online courses varies.

Alignment with learning goals. Reviews often suggest that outcomes are better when game elements support the learning task directly, not sitting beside it. Elements that reward activity without regard to quality can encourage superficial behavior.

Design quality. Poorly implemented systems generally get weaker results. Because “gamification” covers many different designs, comparing one study with another is hard.

Problems with the evidence

Anyone reading this literature should be aware of its common weaknesses.

• Small samples and short durations. Many studies involve a single class or a few dozen participants, over weeks.
• Weak comparisons. Comparing a gamified version with a version that is otherwise worse, or with no intervention, can exaggerate benefits. A fair test compares with a well-designed non-gamified version.
• Lack of random assignment. If learners choose whether to use the gamified version, those who choose it may differ from those who do not.
• Multiple changes at once. If a gamified course also changes content, feedback and teacher attention, it is unclear which change mattered. This is a version of the principle that correlation is not causation.
• Measurement choices. Outcomes that are easy to count may not be those that matter.
• Publication bias. Studies with positive results are generally more likely to be published than those finding nothing.
• Inconsistent definitions. Different studies use the word gamification for very different things.

Questions to ask of any study

1. How many learners took part, and for how long?
2. What was the comparison group, and was it fairly designed?
3. Were learners assigned at random, or did they self-select?
4. What exactly was measured, and was it learning or only activity?
5. Were the results maintained after the novelty wore off?
6. Who funded the study, and did the authors have a stake in the product?
7. Have other researchers found similar results independently?

What can reasonably be taken from the literature

A cautious reading supports a few modest points.

• Game elements can increase participation in some settings, particularly where activities are repetitive and optional.
• Effects on actual learning are less certain and often small when found.
• Competitive features carry particular risks for some learners.
• Design and alignment with learning goals matter more than the presence of elements as such.
• Long-term effects are poorly understood.

These points leave room for sensible use. A designer might add a progress indicator and optional badges, monitor how learners respond, and stay ready to adjust. They do not justify expecting dramatic gains.

For designers and readers

If you design courses, treat the research as a source of questions, not a recipe. If you read claims made by vendors, look for independent evidence. And if you are a learner, remember that the best test of any feature is whether it helps you understand and remember the material. The literature suggests that for some people and tasks it will, and for others it will make little difference.