Explore the approach
Follow the evidence, from comment to report.
A comment can praise teaching and criticise assessment feedback. Explore how those experiences stay separate, how they contribute to a summary and how to find the words behind a count.
This is a teaching example with 12 synthetic comments and manually assigned labels. It illustrates the workflow, not a live model, actual survey result or product interface. No upload or sign-in is needed.
02 · Make the denominator visible
Count comments, then inspect the evidence.
Denominator: 12 synthetic comments. Each comment counts once per topic.
Positive and negative columns count comments containing each sentiment for that topic. A mixed comment can appear in both, so those columns need not add to the total. Topic shares also overlap. These counts are not the sentiment index used in our public NSS research.
| Topic | Comments | Share | Positive | Negative | Neutral |
|---|---|---|---|---|---|
| Academic support | 3 | 25% | 3 | 1 | 0 |
| Assessment feedback | 4 | 33% | 2 | 2 | 1 |
| Community | 3 | 25% | 1 | 1 | 1 |
| Learning resources | 3 | 25% | 3 | 2 | 0 |
| Teaching | 3 | 25% | 3 | 0 | 0 |
Change the survey context to recalculate the denominator. The topic filter narrows the evidence shown above; it does not change the survey denominator.
Understand topic-level sentiment · Read the public NSS measurement definitions
03 · Take a question to the right team
Turn a pattern into a practical review.
In this example, assessment feedback includes positive accounts of useful advice, a complaint about timing and a request for a worked example. Treating all three as the same problem would lose the detail a course team needs.
- Frame the question: can students understand and use the feedback before their next assignment?
- Review the evidence: compare the relevant comments, assessment brief and return dates with students and staff.
- Agree a response: identify the specific change, a responsible person and when it will happen.
- Check back: ask whether the revised experience addresses the original issue, and explain the decision to students.
This is a suggested review process for the synthetic example, not a claim about a university’s results.
Continue with the assessment-feedback pathway →What would this look like for your survey?
We can discuss the questions, categories and reporting outputs appropriate to your institution.
01 · Keep the original context
Read the words and their labels together.
Each extract below keeps its original comment and survey. Some sentences relate to more than one topic; repeated mentions of the same topic still count as one comment in the summary.
C01 · Module evaluation — The seminars helped me understand the topic. Feedback on my essay arrived too late to...
C02 · Module evaluation — The feedback explained how to improve my argument. I used it in the next assignment.
C03 · Module evaluation — The reading links were difficult to find, but the librarian helped me get access.
C04 · Module evaluation — Small-group discussions made it easier to ask questions.
C05 · Module evaluation — I could not tell what the marking criteria meant. A worked example would help.
C06 · Module evaluation — The recorded explanations were useful when I needed to revisit a difficult concept.
C07 · Postgraduate survey — My supervisor helped me plan the next stage. I found it difficult to meet other resea...
C08 · Postgraduate survey — The writing group gave me people to discuss ideas with.
C09 · Postgraduate survey — I was passed between teams when I asked about support. Eventually someone explained t...
C10 · Postgraduate survey — The library resources supported my project, but remote access was unreliable.
C11 · Postgraduate survey — I would like more opportunities to meet researchers outside my department.
C12 · Postgraduate survey — Written feedback helped me decide what to revise before our next supervision meeting.