Beyond the scorecard: what exam analytics should tell you

Every exam produces a large amount of data, and most institutions use very little of it. Typically they look at each candidate’s score, perhaps a pass rate and a merit list. The same data can also show which questions worked, where teaching fell short and which students need help now.

Start with the questions

The report most institutions skip is item analysis, a breakdown of how each question performed.

• Difficulty index is the share of candidates who answered correctly. A question that almost everyone gets right, or almost nobody does, tells you little about ability.
• Discrimination shows whether stronger candidates did better on that question than weaker ones. If they didn’t, the question may be unclear, wrongly keyed or testing something else.
• Distractor analysis shows how often each wrong option was picked. An option nobody chooses isn’t doing its job.

Reviewing these numbers after each exam steadily improves your question bank and makes later exams fairer.

Then look at topics and groups

Once you trust the questions, step back:

• Topic-wise accuracy shows which concepts a batch struggled with, so teachers know where to focus.
• Section analysis shows whether time pressure or one weak area pulled scores down.
• Cohort comparison shows how this batch compares with earlier years or with other centres.

Operational reports matter too

Analytics is also about running exams smoothly. These reports help:

• Attendance, by centre, slot and shift
• Subjective evaluation status, so no answer is left unmarked
• Scaled scores, showing normalised results across papers and shifts

Put all your results in one place

Many institutions run online tests, CBT exams and OMR-marked paper exams, each tracked in its own spreadsheet. Bringing them together saves hours of manual work, and it means you can finally compare like with like.

Use the data to change something

Analytics is only worth the effort if it changes decisions. A simple routine after each exam:

1. Check item analysis and fix or retire weak questions.
2. Share topic-wise reports with teachers so they can adjust their plans.
3. Identify students who need extra support.
4. Compare cohorts to see whether last term’s changes worked.

RealExam Analytics is built around this routine, so each exam helps you improve the next one.