Academy · Classroom Impact Lab
Look carefully at what changed
Enter class results from before, straight after and some time after teaching. The lab describes the differences and lists what else could explain them.
PrototypeThis is a descriptive tool, not a research study. One class and one assessment cannot show that a strategy caused a change. Enter group results only, never student names. Everything is saved in this browser only.
Loading your saved work…
1. Context
Describe the class and the assessment. Group results only: do not enter student names.
2. Results
Enter results for each assessment point you have. A baseline and at least one later point are needed to compare.
Baseline
Before teaching
Immediate
End of the lesson or unit
Delayed
Days or weeks after teaching
3. Misconceptions
For each misconception, count how many learners showed it at each point. Leave a cell blank if you did not check.
4. What else happened
Cognitive Lab · Classroom Impact Lab
Evidence summary
Descriptive classroom evidence entered by the teacher. It does not establish cause.
What the results show
Averages are shown as the percentage of available marks.
- BaselineNo result
- ImmediateNo result
- DelayedNo result
Comparisons
Baseline to immediate
Both assessments need enough valid results to compare.
Baseline to delayed
Both assessments need enough valid results to compare.
Immediate to delayed
Both assessments need enough valid results to compare.
Question by question
Enter results question by question to see this table.
Misconceptions over time
No misconceptions recorded.
Before you draw conclusions
Things that could explain the results other than the strategy.
A classroom comparison is not an experiment
Without a comparison group or random allocation, a change between assessments can reflect ordinary teaching, maturation, practice with the test or who was present, as well as the strategy. Describe what changed; do not conclude what caused it.
Interpretation guide
Reading classroom evidence
Engagement, performance, learning, retention and transfer are different things. Each needs its own kind of evidence.
What different evidence can tell you
- Engagement
- Students are attending to and taking part in the task.
- Evidence: Observation of on-task behaviour, the kind of thinking a task asks for.
- Understanding
- Students can explain, apply or reason with an idea, not only repeat it.
- Evidence: Explanations, worked solutions, questions that need reasoning.
- Retention
- Students can still recall or use the idea after time has passed.
- Evidence: A delayed check, ideally a week or more later, without fresh revision just before it.
- Transfer
- Students can use the idea in a new context or a problem they have not seen.
- Evidence: Unfamiliar problems that need the same underlying idea.
- Correlation
- Two things tend to occur together.
- Evidence: On its own, it says nothing about which caused which, or whether a third factor caused both.
- Causation
- One thing brings about another.
- Evidence: Usually needs a comparison group, random allocation or repeated, controlled replication. A single class rarely provides it.
Collecting better evidence
- Decide what you want to know before you choose a measure: engagement, understanding, retention and transfer need different evidence.
- Use the same or a closely parallel assessment at each point, and keep the conditions similar.
- Include a delayed check if you care about learning that lasts, not only performance on the day.
- Look at question-level results and misconceptions, not only the overall average.
- Note anything else that changed: absences, extra revision, a shortened lesson.
Practise interpreting results
Choose the conclusion that best fits each scenario. Feedback explains what the evidence can and cannot show.
Increased participation, no improvement in delayed recall
A teacher introduces a lively card-sort activity. Almost every student takes part, compared with about half in earlier lessons. On a delayed quiz three weeks later, the class average is about the same as for the previous unit.
Improved immediate scores, weaker retention
After a week of blocked practice, a class scores 82% on an end-of-week test, up from 45% at baseline. Four weeks later, on a parallel test, they score 54%.
High confidence ratings, persistent misconceptions
At the end of a unit, most students rate their understanding 4 or 5 out of 5. On the same day, a two-question diagnostic shows that two-thirds still hold the same misconception they had at the start.