Why we still put student feedback analysis into Excel

Student Voice Weekly, episode 37, with Dr Stuart Grey.

Dr Stuart Grey follows a colleague preparing a learning and teaching committee document using Student Voice AI Excel reports. He explains how the workbooks complement generative summaries, connect categories and sentiment to original survey comments, and support institutional research teams in applying their local knowledge. He also covers routinely generated institution, faculty, department and school sheets, with comment redaction as standard.

Chapters

00:00 - Why Excel still matters for student feedback
00:43 - Preparing a committee document
01:20 - What the Excel reports include
01:48 - Excel and generative summaries
02:09 - Filtering, charts and editable workbooks
02:43 - From categories to the original comments
03:42 - Sentiment and question wording
04:04 - Programme views and institutional knowledge
04:48 - Keeping the evidence behind a finding
05:20 - Institution, faculty, department and school sheets
05:48 - Redaction as standard
06:01 - Making the analysis useful
06:25 - Subscribe and share

Features and further reading

The committee example illustrates how colleagues can move from a category summary to the survey comments behind it, retain the evidence for a finding, and adapt the workbook to their own institutional context.

Subscribe to The Student Voice Weekly. Browse the podcast.

Transcript

Hello, and welcome to Student Voice Weekly. I'm Dr Stuart Grey, founder of Student Voice AI, and today I'd like to talk about why we provide student feedback analysis in Microsoft Excel.

It might sound like a fairly ordinary feature to focus on, but the format of the analysis makes a real difference to who can work with it and how people can work with it.

So over the last couple of weeks, we've looked at local reports and sector comparisons. This week we're talking about what happens when the analysis reaches a colleague who needs to prepare a document, investigate a theme, or discuss findings with colleagues in a school or department. What can they actually do with these comments, with these categorised comments to make their lives easier?

So let's imagine a colleague in a university planning or surveys team preparing a document for a learning and teaching committee. So the committee wants to understand what students are saying about, say, organisation and which issues need further investigation.

So our colleague needs to establish the wider pattern, examine relative differences, and then find the comments that explain what students are experiencing. So they also need to be able to answer any follow-up questions that might occur in the meeting specifically about how they've reached their interpretation. So it has to go back to the actual comments. It can't just be an empty summary.

So that is the kind of work our Excel reports are designed to help. So they bring together the summaries, the categories, sentiment filters, the pivot tables, all on top of the original survey comments. So where there's also the information supplied with the survey, we have things like school, department, programme, year of study, demographic data. So it's all there in one spreadsheet, and the end user can really dive in and get the most out of that data.

So the Excel sheets are actually a brilliant complement to the generative summaries we create. So the summaries bring together the main themes into a written account covering perhaps a large number of surveys. Whereas the Excel sheets allow you to filter in more detail and really dive into the specific comments that you need to build up an evidence base to write documents, for instance, at this level.

So with Excel, we get a lot of things for free. We can filter rows, we can search, we can build new tables, prepare charts. All of our Excel files are entirely open, so you can change them as you see fit, copy and paste, adapt what's there already.

So it's a really flexible way for people to use the data and also really a democratic way 'cause nearly everyone has experience with Excel to different degrees and can really dive into making it their own. So they can explore the structure and also bring their own knowledge to it and adapt it for their needs.

So let's go back to our colleague. They start looking at the data broadly. Okay, what are the main categories? What are the proportions? We can see where organisation features within the wider response set. We can dive into the sentiment, positive and negative under that category. And then we can see, okay, given that, we now know where to look at the individual comments. We can dive into those sheets and filter and very quickly get all the comments about organisation with certain sentiment thresholds.

So these examples, comments might cover, say, timetable changes or assessment schedules or uncertainty about communication. So we want to dive into actual comments, and we can drag those out. We can copy out those comments, again, put them in our document as examples of what we're trying to get across.

So the original wording is really useful here. Verbatim student comments are gold, but because we typically have many thousands of comments, we need a way to quickly get to the right ones, and that's what the Excel sheets let us do.

The Excel workbooks also include views of positive and negative comments. Again, depending on the question asked and also the sentiment is independently applied, so you can find where people were talking negatively when asked a positive question and vice versa, and we see a significant proportion of respondents do that. So our sentiment models allow us to capture that data regardless of which question is asked.

So based on all this information, the colleague now has a really clear idea of what they're asking, what they're trying to find out. They can dive into the Excel sheet themselves. They can filter it by programmes. They can see, okay, which schools have different issues; what are cross-cutting issues right across the whole institution.

So fundamentally, the workbook makes the information available, and the institutional research team decides how to use it based on their experience and their knowledge of working in that institution. So I think that's a really important part. It's a tool to enable people to speed up their work and add the real value that they can add of their institutional knowledge, their understanding of the issues and of student experiences, et cetera.

We can also use these spreadsheets as evidence. So when we do supply documents to committees, et cetera, we can say, "Look, this is what we based it on." You can show them. You can copy and paste the comments out. You can show what the sentiments were, the categorisations were. So again, you have that full evidence chain right the way back to back up the points you're making.

So colleagues might describe the patterns and give certain examples that they used in their exploration. But the workbook itself contains all that data still, and you can point people towards it if they want to have a look themselves.

So we also generate these Excel sheets at institution, faculty, department, and school level automatically, and this means institution-level sheets go to planning and quality and surveys teams, so they get a broad view of the institution. Whereas faculty, department, and school teams receive the sheets focused on their subunit so they can do the same thing and dive into just their comments. And we find this works really well for getting the right people in the institution in the right place to read the right comments.

Also, with this in mind, we carry out redaction of comments as standard, so people on the ground level don't see names, et cetera. So these can be shared with colleagues without worrying about, okay, who's mentioned where.

Overall, we are really pleased with the quality of our Excel spreadsheets and the utility, and our customers really rate them. And I think there's a lot to be said for giving people the data, all the data, the analysed data and the underlying comments in a form they can use, they can adapt, they can communicate with colleagues with to fundamentally improve the student experience.

So that's all for this week. If you'd like to find out more, you can sign up for the newsletter on studentvoice.ai. If you like this podcast, please share it with anyone, and if you have any ideas for future episodes, please do let me know. Thanks.