how-to
How to Measure Student Growth in Game-Based Learning
Quick Summary
Student growth in game-based learning is the change from a student's starting point to their current performance on specific, intended learning outcomes. Measure it with a matched pre-assessment and post-assessment, track in-game evidence against the skill, and never rely on a single signal like points or self-reports.
- Write observable learning outcomes before measuring, such as solving two-digit addition with regrouping.
- Run a short baseline check before play, then a parallel post-check with matched difficulty.
- Track transfer to new problem types, not just repeated items, to separate real learning from memorized clicks.
- Keep engagement metrics like time on task separate from learning evidence.
- Triangulate at least two sources before reporting growth to students or families.
Table of Contents
- Defining Game-Based Learning and Game-Based Assessment
- Start With Clear Learning Outcomes and Instructional Objectives
- Build a Pre- and Post-Assessment Plan Around Your Game
- Game-Based Assessment Examples You Can Adapt
- Choosing Student Growth Tracking Tools and Interpreting Metrics
- Measuring Engagement Without Confusing It for Learning
- Protecting Student Data and Ensuring Fair Measurement
- Run Your First Growth Measurement Cycle This Term
- Frequently Asked Questions
Measuring student growth in game-based learning is harder than it looks. A game can show that a student finished a level, but that does not prove the student learned the math or coding skill behind it.
Game scores often reflect completed challenges or reached targets, which is useful but limited. Researchers working on game-based assessment, such as Udeozor and colleagues in their paper "Measuring learning in digital games: Applying a game-based assessment framework" (British Journal of Educational Technology), propose a framework for measuring learning in digital games.
Here is the core idea for the rest of this article. Measure growth against clear learning goals, collect evidence before and after play, and never trust a single signal, including how much a student says they learned.
Defining Game-Based Learning and Game-Based Assessment
Game-based learning uses games to teach specific skills and knowledge.
Game-based assessment is the practice of using in-game performance data, such as challenges completed, targets achieved and choices made, to estimate a student's learning progress.
The difference matters. A game that is fun does not automatically measure anything.
Watch out: a game can reward speed or repetition without building understanding.
Start With Clear Learning Outcomes and Instructional Objectives
Before you measure anything, write down what the student should be able to do. These are your intended learning outcomes, sometimes called instructional objectives.
- Name one skill per goal (for example, "adds two-digit numbers with regrouping")
- Use observable verbs such as solve, explain, sequence or debug
- Tie each goal to a standard your school already uses, such as Common Core or CSTA
Growth is the change from where a student started to where they are now, measured against these goals. Without a clear target, you are only counting activity.
Build a Pre- and Post-Assessment Plan Around Your Game
To show growth, you need a starting point. Give a short check before students begin, then repeat a similar check after a set period of play.
Keep the plan simple:
- Run a short pre-assessment that matches your learning outcomes
- Play the game for a set number of sessions, such as six over three weeks
- Run a post-assessment with comparable but not identical questions
Use matched difficulty on both checks so you do not mistake an easier test for growth.
Game-Based Assessment Examples You Can Adapt
Generic advice to "track accuracy" is not enough. Below are three worked examples that show the full measurement cycle: baseline, gameplay evidence, follow-up, and interpretation. Each example uses a simple table you can recreate in a spreadsheet. The numbers are illustrative, not prescriptive, but the structure is what matters.
Example 1: Math fluency game (Grade 3 addition with regrouping)
Learning outcome: Students will solve two-digit addition problems with regrouping with 80% accuracy.
Baseline (pre-assessment): Give a 10-item paper or tablet quiz matched to the outcome.
Follow-up (post-assessment): Give a parallel 10-item quiz with the same difficulty distribution but different numbers.
Sample student record:
| Measure | Baseline | Session 3 | Session 6 | Post-test |
|---|---|---|---|---|
| Regrouping accuracy | 4/10 | 6/10 | 8/10 | 8/10 |
| Avg attempts per item | 2.8 | 2.1 | 1.4 | n/a |
| New problem types solved | 0/3 | 1/3 | 3/3 | 3/3 |
Interpretation: Accuracy rose, attempts fell, and transfer to new problem types improved. That pattern suggests real growth, not just memorized clicks.
Example 2: Coding puzzle game (debugging a loop)
Learning outcome: Students will identify and fix an off-by-one error in a loop within three attempts.
Baseline: Ask students to debug two short code snippets on paper and explain the fix aloud. Score with a simple 0-2 rubric (0 = no fix, 1 = fix with help, 2 = independent fix with explanation).
Gameplay evidence: Log attempts to first successful fix, whether the fix holds on a new puzzle, and whether the student can explain the change in a short exit ticket.
Follow-up: Give two new debugging tasks of matched difficulty and rescore with the same rubric.
Sample class summary:
| Student | Baseline rubric | Attempts to fix (last session) | New-puzzle fix holds? | Post rubric |
|---|---|---|---|---|
| A | 1 | 2 | Yes | 2 |
| B | 1 | 5 | No | 1 |
| C | 0 | 3 | Yes | 2 |
Attempts to fix (last session)
Interpretation: Student B spent time in the game but did not transfer the fix to a new puzzle. That is a signal to reteach, not to celebrate time-on-task.
Example 3: Reading or story game (inference and evidence)
Learning outcome: Students will support an inference with evidence from the text.
Baseline: Read a short passage and ask one inference question.
Gameplay evidence: Record dialogue choices and whether the student selects evidence-based options, plus a short written or verbal justification after key choices.
Follow-up: Use a parallel passage and the same rubric.
Interpretation: A score alone would not tell you whether the student can explain the choice.
Watch out: Do not compare a pre-test and post-test that differ in difficulty or format.
These three examples share a common structure you can reuse: define one observable outcome, collect a baseline with a rubric or quiz, log the specific in-game behaviors that map to that outcome, run a matched follow-up, and interpret the pattern across accuracy, attempts, and transfer.
Choosing Student Growth Tracking Tools and Interpreting Metrics
Accuracy over time is a start, but the harder question is which metric fits which learning objective, and how to tell progress from mere activity. This section gives you a selection framework and an interpretation checklist you can apply before you trust any dashboard.
A metric-selection framework
Match the metric to the type of learning you are measuring. Three broad categories cover most game-based learning goals:
| Learning goal type | Example | Best-fit metrics | Weak-fit metrics |
|---|---|---|---|
| Procedural fluency | Math facts, typing, basic coding syntax | Accuracy on new items, attempts to first correct response, time per correct item | Total points, levels completed, time on task |
| Conceptual understanding | Regrouping, loops, inference | Transfer to new problem types, explanation quality on exit tickets, rubric scores | Repetition counts, streak length |
| Strategic or collaborative skill | Planning, peer feedback, debugging strategy | Quality of choices in novel scenarios, peer-interaction logs, teacher observation rubric | Session completion, voluntary return rate |
Use this table to avoid the most common mistake: choosing a metric because the game exports it easily, not because it maps to the learning outcome. If the game only exports points and levels, you may need to add a short rubric-scored task outside the game to capture the skill you actually care about.
Establishing a baseline that means something
A baseline is only useful if it uses the same construct and comparable difficulty as your follow-up. Three practical rules:
- Use the same rubric or item blueprint for pre- and post-measures, changing only surface details.
- Collect the baseline before gameplay begins, not after a practice session.
- Record the conditions (device, time of day, support given) so you can tell a real change from a context effect.
A common pattern is to skip the baseline and compare students to each other. That measures standing, not growth. Growth is always a within-student comparison from baseline to follow-up.
Distinguishing progress from activity
Activity metrics (time on task, sessions completed, voluntary return) tell you about engagement. They do not tell you about learning. Use this checklist to separate the two:
- Transfer check: Does the student succeed on new items or only repeated ones?
- Explanation check: Can the student explain the reasoning behind a correct answer?
- Retention check: Does performance hold after a gap of a week or more?
- Triangulation check: Do at least two independent sources agree (e.g., game log and post-test, or game log and teacher observation)?
If activity is high but all four checks fail, you have engagement without evidence of learning. That is useful information, but it is not growth.
Interpreting metrics without overclaiming
Self-reports on their own are weak evidence. If a student says they feel better at fractions but their scores have not moved, trust the scores and look closer at where the breakdown occurs.
A simple interpretation sequence helps:
- Compare baseline and follow-up on the same rubric or item blueprint.
- Check whether gains appear on new items, not just repeated ones.
- Look for agreement between in-game data and the follow-up measure.
- Note any explanation or observation evidence that supports or contradicts the scores.
- Decide whether the pattern is growth, activity, or a measurement artifact (easier test, different conditions).
For a deeper look at tool options and rubric design, see our evaluate student coding projects fairly guide, which includes a reusable rubric you can adapt to game-based tasks.
Watch out: Do not treat a single metric as proof of growth. A rising accuracy score with no transfer and no explanation is weak evidence.
Measuring Engagement Without Confusing It for Learning
Engagement is important, but it is not growth. A student can be busy, focused and happy while learning nothing new.
Track engagement and learning separately:
- Engagement: time on task, session completion, voluntary return to the game
- Learning: accuracy on the target skill, performance on new items, post-assessment results
Katrina Serrano's literature review of digital game-based learning and student learning (University of Northern Iowa) is a readable starting point on what research says about learning games. Whatever the motivation boost, the goal is skill gains you can measure, not just engagement.
Protecting Student Data and Ensuring Fair Measurement
Game data can reveal a lot about a child. Limit what you collect to what you need for the learning goals, store it securely, and explain to families what is tracked and why.
Fairness matters too. Students with slower internet, less screen experience or reading difficulties may score lower for reasons unrelated to math or coding. Check whether a low result reflects the skill or the barrier, and offer alternative ways to show what a student knows.
Run Your First Growth Measurement Cycle This Term
Start small. Pick one skill, one game and one class. See real outcomes from comparable programs to set a benchmark for your first cycle. Write the learning outcome, run a pre-assessment, play for four to six sessions, then run a matched post-assessment. Review the results with a colleague and adjust the next cycle.
Get started with CodeAlgo Academy to see how the platform supports measurable math growth across K-8 classrooms.
Frequently Asked Questions
How do you measure student growth in game-based learning?
Start by defining the learning outcomes the game targets, then collect evidence before and after play. A common approach pairs a short pre-assessment with a post-assessment that uses similar items, so the change can be attributed to learning rather than familiarity. In-game data such as challenges completed, targets achieved and levels reached can supplement this, but scoring metrics show performance inside the game and do not automatically prove growth. Triangulate with independent measures to make the case for growth.
Are self-reported gains a reliable measure of student growth?
Not on their own. Self-reported gains can differ from measured results. Students may feel they have improved without measurable change, or the reverse. Use student self-reports as a complement to pre- and post-assessments and in-game evidence, and interpret them with caution when reporting to administrators or families.
What should teachers assess during game-based learning?
Teachers should assess the specific learning outcomes the game is designed to build, such as a math skill or a computational thinking concept, rather than general enjoyment or completion. Useful evidence includes accuracy on targeted problems, the strategies students use when stuck, and whether they transfer the skill to a non-game task. Separating these from time-on-task or engagement signals keeps the assessment focused on learning.
How can teachers use pre- and post-assessments with educational games?
Give a brief assessment before the first game session and repeat a parallel version after the unit ends. Keep items aligned to the same standards and difficulty so scores are comparable. Record the results in a simple tracker with each student's starting and ending scores. Pair this with in-game performance data to see whether gains match the assessment, and review any students with large in-game progress but flat test scores.
