You need cookies enabled

Cookies

You need cookies enabled

These questions aim to help institutions and individuals understand the National Student Survey (NSS) data. Other sources of NSS data are available, including on the Unistats website, as well as information released to individual institutions through Texunatech. These are not covered by these questions.

I would like to undertake research into the NSS but need more detailed data. Am I able to request it?

We have now created the opportunity and means for researchers to apply to conduct analysis or research into data from the NSS. 

All proposals are subject to consideration by the NSS Research Panel. The terms of reference for the panel and guidance on how to apply can be found in the documents below. We are open to proposals at any time. 

All proposals should be submitted to NSS@hefce.ac.uk

What NSS data is published on the HEFCE website?

The Teaching Quality Information data on the HEFCE website contains various breakdowns and presentations of the NSS data in 7 excel workbooks. The information provided in each workbook is outlined in the table below.

All public data conforms to NSS publication thresholds (at least 10 student responses and a 50 per cent overall response rate). If no data is available for a particular course on which students were surveyed, the data did not meet publication thresholds. A lack of publishable data is not reflective of the quality of a course or an institution.

Spreadsheet

Details

20XX National Student Survey summary data

This contains a summary of the NSS data for England, Scotland, Wales, and Northern Ireland. This workbook also contains the results of Q27 per institution, as well as a benchmark* figure for this question per institution.

20XX NSS results by teaching institution for all institutions

 

This contains all publishable NSS results for all institutions, with students’ responses being attributed to their teaching institution. The data are presented for the institution as a whole (‘NSS’ tab), as well as at the three JACS subject levels^ (NSS1-NSS3).

20XX NSS results by teaching institution (Full time)

 

This contains all publishable NSS results of full time students at all institutions, with students’ responses being attributed to their teaching institution. The data are presented at the three JACS subject levels^ (NSSFULLTIME1- NSSFULLTIME3).

20XX NSS results by teaching institution (Part time)

This contains all publishable NSS results of part time students at all institutions, with students’ responses being attributed to their teaching institution. The data are presented at the three JACS subject levels^ (NSSPARTTIME1- NSSPARTTIME3).

20XX NSS results by registering institution for all institutions

 

This contains all publishable NSS results for all institutions, with students’ responses being attributed to their registering institution. The data are presented for the institution as a whole (‘NSS’ tab), as well as at the three JACS subject levels^ (NSS1-NSS3).

20XX NSS results by registering institution (Full time)

 

This contains all publishable NSS results of full time students at all institutions, with students’ responses being attributed to their registering institution. The data are presented at the three JACS subject levels^ (NSSFULLTIME1- NSSFULLTIME3).

20XX NSS results by registering institution (Part time)

 

This contains all publishable NSS results of part time students at all institutions, with students’ responses being attributed to their registering institution. The data are presented at the three JACS subject levels^ (NSSPARTTIME1- NSSPARTTIME3).

*Please refer to ‘What is benchmarking?’ for a detailed explanation.

^Please refer to ‘What information is available on the JACS subject level breakdown worksheets?’ for detailed information on what these worksheets contain.

What is benchmarking?

Benchmarking is a process by which individual scores can be compared to a sector benchmarking group average while taking into account those factors that are known to influence the item being measured. In terms of the NSS, these factors include; subject of study, age, sex, ethnicity, disability, and the mode of study of individual students at an institution. To achieve this, sector average scores are calculated for each benchmarking group. The benchmark for each institution is then calculated by taking a weighted average of the scores for benchmarking groups taking account of their mix of students. For a more detailed description of benchmarking, please refer to the HESA website.

What information is available on the JACS subject level breakdown worksheets?

Each subject level spreadsheet breaks down the data by institution, subject (based on JACS coding), level (First Degree, Other Undergraduate), and question. It allows institutions to see the percentage of respondents who selected ‘Strongly Disagree’ (Answered 1, Column F),  ‘Disagree’ (Answered 2, Column G), ‘Neither agree nor Disagree’ (Answered 3, Column H), ‘Agree’ (Answered 4, Column I), and ‘Strongly Agree’ (Answered 5, Column J) to each question. The value in column M (‘Actual Value’) is the proportion of students who selected ‘Agree’ or ‘Strongly Agree’ to a question. 

Why are all the results whole numbers? 

The data presented to institutions and public is rounded to the nearest whole percentage point, which means that when the ‘Agree’ and ‘Strongly’ agree columns are combined, it may not be the exact percentage shown in column M. For example, if 31.3 per cent of students respond with ‘Agree’, and 43.4 per cent respond ‘Strongly Agree’, this would be displayed as 31 per cent and 43 per cent; however, the ‘Actual Value’ would be shown as 74 per cent, as 73.7 per cent (31.3 per cent + 43.4 per cent) would be rounded to 74 per cent, whereas individually 31.3 and 43.4 are rounded down. 

If data from the latest NSS does not meet publication thresholds, it is merged with the previous year’s data where this allows the data to reach publication thresholds. This is shown by a ‘Y’ in column Q (‘Two years aggregate data?’). 

What are the ‘Confidence Intervals’ attributed to the data? 

Confidence intervals are published for all the data contained in the detailed data spreadsheets. Confidence intervals are a statistical tool used to demonstrate how reliable an estimate of a value is at representing the actual unknown value. Confidence intervals are useful where it is unlikely or impossible to receive a response from your entire target population, which is the case for the NSS. They allow us to demonstrate how closely the responses from students represent the actual value that would have been achieved had all of the students in a particular cohort responded to the survey. Generally, the smaller the difference between the minimum confidence interval and the maximum confidence interval, the more closely the estimate represents the actual unknown value had the entire population of eligible students responded. 

Confidence intervals are impacted by the size of the population and by the variability within that population. The size of a confidence interval is not reflective of the quality of an institution.

Download an explanation of how confidence intervals are calculated for the NSS

Download the Confidence interval calculations as PDF (441 KB)

Why are the combined NSS results for all students not just the average of the results for full-time and part-time students? 

In most cases, the combined NSS results for both full-time and part-time students will most closely resemble the results for full-time students. This is due to the fact that generally there are a much larger proportion of full-time students compared to part-time students, so when calculating the average of the two groups, there is a large weighting towards full-time students. 

If a student responds to a question with ‘Neither agree nor disagree’, how is this response treated? 

These are treated as neutral responses. They do not contribute towards the totals for positive or negative responses.

Page last updated 9 August 2017