How to read the statistics
Average, distribution, difficulty, discrimination and Cronbach's alpha.
The test's Statistics step shows overall figures, the distribution of results and an analysis of each question.
| Measure | Meaning | How to interpret |
|---|---|---|
| Difficulty | Share of respondents who answered correctly (0–1) | Below 0.3 — hard; above 0.8 — easy |
| Discrimination | Difference in results between the top and bottom 27% groups | Above 0.3 — good; negative — check the question |
| Correlation | Relationship between the question score and the total score | Positive and high is good |
| Cronbach's alpha | Internal reliability of the test | Above 0.8 — high; 0.6–0.8 — acceptable |
Option analysis
Option analysis on a question row shows how many respondents chose each answer option. A wrong option that nobody picks is not working — replace it with a more plausible one.
A negative discrimination value means stronger respondents did worse on this question than weaker ones. This usually means the correct answer is marked wrongly or the question is ambiguous.
You need at least 20–30 results for the statistics to be reliable.
Related articles
Didn't find an answer?
Send us your question and we'll reply within business hours.