
Gamifying Financial Literacy: What Works in the Classroom
Gamified classes score higher than comparison classes on average, by about half to four-fifths of a standard deviation in meta-analyses published between 2019 and 2024. The average hides a wide spread, though, and the one analysis that split results by age found almost no difference for secondary-school students (eight studies).
What works best is less settled than that headline suggests. The studies compared gamified and plain classes; they did not test ownership, stakes, or most individual mechanics. Our own reading, labelled as judgment wherever it appears below, is that games where students make decisions and see the results are a better bet than points and badges added to a worksheet.
What does gamified financial literacy actually mean?
Gamified financial literacy means teaching money skills through game mechanics: challenges, progression, feedback, competition, and roles. Done well, students are not reading about investing. They are making decisions and watching outcomes.
At one end sits surface gamification: points for finishing a module, a badge for logging in, a leaderboard of who clicked fastest. At the other sits deeper gamification: a simulated portfolio the student owns, a role to play, a competition with an outcome they care about. Both get called "gamified." Whether the deeper version teaches more is something the studies below only partly answer.
Does gamification actually work? What the research says
On average, gamified classes outperform plain ones. These meta-analyses pooled studies that compared the two:
| Meta-analysis | What it pooled | Average difference | Worth noticing |
|---|---|---|---|
| Sailer and Homner, 2020 | 19 samples for test-type (cognitive) outcomes, 16 for motivation, 9 for behavior | g = 0.49 cognitive, 0.36 motivation, 0.25 behavior | Only the cognitive difference stayed significant in the most rigorous subset |
| Kim and Castelli, 2021 | 18 quasi-experimental studies from 2010 to 2019 on behavioral change | d = 0.48 overall (0.30 on test scores, 0.60 on participation) | Only two of the 18 studies involved K-12 students |
| Li, Ma and Shi, 2023 | 41 studies (49 samples, 5,071 participants) on learning outcomes | g = 0.82 (95% CI 0.57 to 1.08) | Results varied enormously from study to study |
| Zeng and colleagues, 2024 | 22 experimental studies from 2008 to 2023 on academic performance | g = 0.78 | The authors call this a moderately positive effect |
These are standardized effect sizes. If scores are roughly bell-shaped, a g of 0.5 puts the average gamified student near the 69th percentile of a comparison class, and a g of 0.8 near the 79th. Li, Ma and Shi note that their figure sits well above the 0.40 to 0.60 reported in earlier work.
Design shows up in the data too. In Li, Ma and Shi, gamification built only from mechanics (points, rewards, rankings) averaged g = 0.53 across 15 studies. Designs that also included dynamics (feedback, competition) and aesthetics (narrative, companionship) averaged g = 1.29 across another 15. That compares studies rather than running an experiment, so other differences may be doing some of the work.
Motivation research adds a warning. Li, Hew and Du (2024) pooled 35 interventions (about 2,500 participants). Gamified classes scored slightly higher on intrinsic motivation overall (g = 0.26). Students in them rated their autonomy (g = 0.64) and relatedness (g = 1.78, with a very wide confidence interval) higher, but their perceived competence only a little higher (g = 0.28, about the size of the overall difference). Gamification seems to lift the feeling of freedom and connection far more than the feeling of competence. The authors' systematic review of 31 studies named low perceived competence and low perceived autonomy as the main problems in gamified classes. Our reading is that this gap is where shallow gamification lives.
A classroom study of a course gamified with a leaderboard and badges reported that students became less motivated, satisfied, and empowered over the semester than students in a non-gamified section (Hanus and Fox, 2015). It is one study of one design, so read it as a caution.
Motivation research offers a possible mechanism. A classic meta-analysis of 128 experiments found that tangible rewards significantly undermined intrinsic motivation, measured as free-choice time on the task (d = -0.34), and more so for children than for college students (Deci, Koestner, and Ryan, 2001). Points are not cash or prizes, so this is a caution about the mechanism, not proof that points backfire.
Where is the gamification evidence thin?
Plenty is missing, and the gaps matter in a grade 6 to 12 classroom.
- The average describes no particular classroom. Studies in Li, Ma and Shi disagreed sharply (I-squared of 94 percent, meaning nearly all the variation reflected real differences between studies rather than chance).
- School-age evidence is thin and inconsistent. In that analysis, 12 elementary samples averaged g = 1.29 and 29 higher-education samples 0.87, while the 8 secondary-school samples averaged 0.01 (95% CI -0.26 to 0.29). Kim and Castelli found only two K-12 studies, and Sailer and Homner three school-setting studies with large test-type differences.
- Personal finance is not a category. Li, Ma and Shi sorted studies by discipline, and none was personal finance. The nearest, business, had five studies and averaged g = 0.03 with an interval from -1.13 to 1.19, too wide to read either way.
- Duration has no clean answer. Sailer and Homner found no significant duration difference for test-type or behavior outcomes. Kim and Castelli found interventions under an hour averaged d = 1.57 against 0.39 for multi-week ones. Li, Ma and Shi found the shortest (g = 1.87) and the longest (over a semester, g = 3.30, five studies) both beat the middle.
- Rigor shrinks the numbers. In Sailer and Homner's strongest-design subset, the test-type difference held at g = 0.42, but the motivation (0.22) and behavior (0.27) differences were no longer statistically significant.
- They test gamification in general. Nothing here speaks to any particular product, ours included.
Which game mechanics have evidence behind them?
Use this table to audit any gamified tool. The middle column separates what the research tested from our own design judgment, because several mechanics were never isolated in the meta-analyses.
| Game mechanic | What the research says | How to use it well |
|---|---|---|
| Points for completion | Not tested alone. Designs built only from points, rewards, and rankings averaged less than fuller designs (Li, Ma and Shi). | Award points for demonstrated reasoning, not for logging in or clicking through |
| Badges | Not isolated. Our design judgment. | Reserve them for mastery milestones students choose to chase |
| Leaderboards | Mixed. Competition alone was the weaker form in Sailer and Homner's data. | Rank on growth or process, or use team boards instead of top-scorer boards |
| Levels and progression | Not isolated. Our design judgment. | Make advancement require applying a concept, like moving from loan officer to bank president |
| Simulation with visible results | Not isolated. Our design judgment. | Give students a portfolio, business, or budget they own, with results the class can see |
| Narrative or game story | Supported for behavior: g = 0.41 with a game story, 0.10 without (3 studies against 7). No difference for test scores or motivation (Sailer and Homner). | Cast students in a role such as Fed Chair, loan officer, or investor. The role-play itself is our judgment, since the studies coded story, not role-play |
| Rewards and goals students care about | Not isolated. Our design judgment, with a caution: expected tangible rewards can undermine free-choice engagement (Deci et al.). | Build the goal into the activity, such as a target to explain or a budget to defend, before reaching for prizes |
| Competition plus collaboration | Best supported, from small numbers. On behavior, g = 0.70 for mixed designs against 0.12 for competition alone (3 studies against 6). Among the most rigorous studies, mixed designs also beat competition alone on motivation (g = 0.61 against -0.09; 2 studies against 5), per Sailer and Homner. | Use team challenges so students compete and coach each other |
Two rows rest on a tested comparison; the rest are mixed or our own judgment. Our working rule is to favor mechanics that ask students to apply a concept and see the result over mechanics that reward activity.
Points versus ownership
This section is design reasoning informed by self-determination theory, not a tested finding.
A point is extrinsic. A student can chase it, bank it, and forget the lesson. A portfolio the student built works differently. When a student buys a share of a company they actually use and watches it move, the result belongs to them.
Self-determination theory holds that motivation grows when students feel autonomous ("I chose this"), competent ("I can do this"), and connected ("we are in this together") (Ryan and Deci, 2000). Designing for those three needs is a sensible response to the motivation findings above. Simulated money carries no real risk, so the stakes here are a goal and a result everyone can see.
How to gamify financial literacy well: a teacher framework
You do not need to build a video game. Put ownership and visible results into what you already teach, and make progression mean something:
- Give students something to own. A simulated portfolio, a business to run, a budget with a goal.
- Give the work a visible result. Ask students to defend decisions and explain outcomes, since a grade for the highest return mostly rewards luck.
- Make progression mean mastery. Levels should unlock because a student demonstrated a skill, not because time passed.
- Mix competition with collaboration. Mixed designs beat competition alone on behavior, and on motivation among the most rigorous studies (2 studies against 5), in Sailer and Homner's data (Sailer and Homner, 2020). Team leaderboards were not tested as such, so treat them as a sensible application, not a proven one.
- Grade the thinking, not the score. Reflection and reasoning are where the learning shows up, and grading them protects the careful student with an unlucky week. Here is how to grade a stock market simulation without rewarding luck.
A live simulator instead of a points economy
This research guides how we design Rapunzl: a live simulator rather than a points economy. Students get a simulated $10,000 portfolio of stocks and crypto with live Nasdaq pricing, choose the companies, and watch the market move their results during class. No real money is at risk, so the stakes are a goal and a visible result.
Rapunzl also runs a free national scholarship competition from January to late April, and recent competitions have drawn nearly 100,000 students from 1,000+ schools. A scholarship is a tangible reward, the kind the research above says can undermine interest, and we have no evidence on how it affects motivation. A classroom challenge with a reward students care about can serve a similar purpose on a smaller scale.
Frequently Asked
Questions
On average, classes that use gamification score higher than plain ones, by about half to four-fifths of a standard deviation in the meta-analyses (Sailer and Homner, 2020; Li, Ma and Shi, 2023). Evidence for secondary-school students is thin, and none of these analyses had a personal finance category, so treat the numbers as a rough guide.
On their own, rarely. Tangible rewards can undermine intrinsic motivation, especially for children (Deci, Koestner, and Ryan, 2001), and designs built only from points, rewards, and rankings averaged smaller differences than fuller ones (Li, Ma and Shi, 2023). Tie points to demonstrated reasoning or mastery, never to logging in or clicking through.
It depends on the design. Competition by itself was the weaker form in the pooled data, while mixing competition with collaboration did better on behavior and, in the most rigorous studies, on motivation (2 studies against 5; Sailer and Homner, 2020). Team leaderboards were not tested specifically, so ranking on growth or process, or using team boards, is a reasonable application, not a proven one.
A simulation gives students a portfolio, business, or budget they own and a result they can see, while a points system rewards activity. The studies above did not isolate that difference. The reasoning rests on self-determination theory, which links autonomy to intrinsic motivation (Ryan and Deci, 2000).
Choosing tools for a gamified unit is easier with someone to compare notes with, so start a conversation with us about what Rapunzl could do for your classroom.
By Maria Rodriguez, Curriculum Designer at Rapunzl, designing gamified and Spanish-language financial literacy curriculum for grades 6–12.
Get the free Rapunzl Starter Unit
Module 1, ready to teach: a slide deck, a quiz, and the answer key. Free.











