Consultation
Published 22 October 2026
Proposals for a revised approach to assessing and rating quality
Published 22 October 2026
Proposal 8: Student experience indicators
Stage one decisions
- NSS-based indicators will continue to be used and will be expanded to include an indicator based on the NSS ‘Learning opportunities’ theme.
- NSS-based indicators will be benchmarked to show how students’ views compare with those of similar students on similar courses across the sector.
What we are proposing now
- We propose how the NSS-based indicators would be mapped to the assessment criteria, reported for different provider and student populations, and benchmarked using an updated set of benchmarking factors.
Why we are proposing this
- This approach would provide consistent and comparable evidence about students’ experiences while supporting fair assessment of providers with different student populations.
Key points of this proposal
- Assessment of each student experience criterion would be informed by a corresponding NSS-based indicator.
- NSS indicators would be based on the most recent four years of data.
- NSS indicators would be presented in a combined view of taught and subcontracted-out provision, as well as separate views for taught, subcontracted-out and validated-only provision.
- The indicators would also be ‘split’ by subject, course characteristics and student characteristics, including some additional splits to align with access and participation data.
- Benchmarking factors would be revised to replace separate student characteristics with a single composite measure of student characteristics (ABCS).
Explanation
- NSS-based indicators provide evidence about students’ views of different aspects of their academic experience. Following our stage one decisions, these indicators would continue to form an important part of the evidence used in assessing the student experience aspect. As set out in Proposal 4, a provider would need to have sufficient NSS indicators to apply for a TEF student experience rating. All TEF student experience ratings would therefore be informed by NSS indicators, alongside submissions and other applicable evidence.
- Where a provider undergoes an assessment against only the minimum requirements for the student experience aspect, we would consider any NSS indicators that are available as part of the assessment, and apply the same approach as set out in this proposal.
- We propose a direct mapping between each student experience criterion and an NSS indicator. This is based on mapping the relevance of the assessment criteria to the questions asked within each NSS theme.6 This mapping would help ensure that the quantitative evidence aligns closely with the areas being assessed and makes it clearer how NSS evidence contributes to judgements.
Table 9: Proposed mapping of NSS indicators to student experience criteria
|
Student experience criterion |
NSS indicator |
|
1. Course content |
Learning opportunities |
|
2. Course delivery |
Teaching on my course |
|
3. Assessment and feedback |
Assessment and feedback |
|
4. Student support |
Academic support |
|
5. Learning resources |
Learning resources |
|
6. Student engagement |
Student voice |
Structure of the NSS indicators
- We propose to present a combined view of the NSS indicators for taught and subcontracted-out provision to provide overall indicators covering all provision that will affect the assessment against the criteria. This would enable assessors to start with an overall indication of performance.
- NSS indicators would also be presented separately for taught provision and subcontracted-out provision to help understand differences between them. They would also be presented separately for validated-only provision to support our approach to assessing this provision in a different way (see Proposal 6). The indicators for subcontracted and validated provision would also be split by named partner provider. We would continue to present split indicators for different student and course characteristics. This would help assessors understand whether students’ experiences are consistent across subjects and areas of a provider’s provision, and across different groups including students from disadvantaged backgrounds.
- We are considering ways in which we can align the TEF and access and participation plans to better align the processes and support improvement at providers. This includes aligning the TEF and access and participation data, to enable the two activities to better complement each other. To help improve this alignment we propose in the TEF data to include the more detailed breakdown by type of disability that we use in the access and participation data. We also intend to reinstate the split for free school meals (FSM) eligibility, which was temporarily removed from the current version of the TEF dashboard while we addressed a technical challenge with implementing the required approach to data suppression.
- We also propose an additional split for Associations between characteristics of students (ABCS) measures. This is because we propose below to use these as a benchmarking factor and for transparency we include all benchmarking factors as split indicators.
- As set out in Proposal 6 we are also interested in views about whether it would be helpful to include an additional split for students studying on fully flexible courses.
Benchmarking factors
- We decided following stage one that NSS indicators would continue to be benchmarked. Benchmarking allows providers’ results to be considered in the context of the characteristics of their students and courses, providing an indication of how students’ views compare with those of similar students studying similar courses elsewhere in the sector.
- We have reviewed the benchmarking factors used for NSS indicators and, as part of this, have developed ABCS measures for NSS for full-time and part-time students. ABCS allow us better to understand how survey responses vary for groups of students holding different sets of characteristics. It provides us with a single composite measure we can use in place of the separate student characteristics currently used to benchmark the NSS indicators. We are proposing to use ABCS because:
- It would be consistent with the approach we take to benchmarking student outcomes indicators.
- ABCS recognise that students often have multiple characteristics that affect their experiences and outcomes. For example, a student may be female, mature and from a disadvantaged background at the same time. ABCS help us take account of these combined characteristics rather than treating each characteristic in isolation.
- It would reduce the number of benchmarking factors. This would make our benchmarking more effective by reducing the number of distinct benchmarking groups and hence the extent of ‘self-benchmarking’ that currently occurs, while still allowing us to take account of the factors that have the greatest effect.
Benchmarking groups and self-benchmarking
Our benchmarking method involves creating benchmarking groups for each unique combination of categories that exists within the benchmarking factors. We place students from all providers into these groups. For example, we currently have a benchmarking group which includes all under 21, female, black students without a disability studying law on a full-time basis.
To construct a provider’s benchmark, we calculate a weighted average based on the percentage of its students that fall into each of these groups and the average rate for each group.
The provider’s own students are included in the benchmarking groups. If the characteristics of the students at the provider do not frequently occur elsewhere in the sector, there won’t be many other students in the benchmarking groups we use to calculate the provider’s benchmark. In these cases, the provider’s students heavily influence the weighted average we calculate and this is known as self-benchmarking. High levels of self-benchmarking make the benchmark less meaningful as we are essentially comparing the provider with itself rather than the wider sector. Having fewer distinct benchmarking groups decreases the likelihood of high levels of self-benchmarking.
The proposed benchmarking factors are available in Part 4.9 of the draft guidance.
For further information about the benchmarking factor review and ABCS for NSS, see the benchmark analysis published alongside the draft guidance.
Indicating performance
- We propose to retain the broad approach used in TEF 2023 to indicating performance in relation to benchmarks. Performance materially above benchmark would indicate outstanding quality, performance broadly in line with benchmark would indicate high quality, and performance materially below benchmark would indicate performance below the level of high quality. Where performance is below benchmark, assessors would consider the extent of that difference when considering if it indicates that quality may be satisfactory or unsatisfactory.
- We have carried out analysis to identify whether we should make changes to how we define a material difference from benchmark, published alongside the draft guidance. The findings do not suggest there is a strong reason to move away from our current approach and we would therefore continue to define a material difference from benchmark as 2.5 percentage points either above or below benchmark.
- We are also not proposing any changes to the way in which we guide assessors to consider the statistical certainty of performance in relation to benchmarks. This is explained in the draft guidance.
- We do not propose to introduce absolute performance thresholds for NSS indicators. The NSS measures students’ perceptions of their experience, and results across the sector are generally concentrated at relatively high levels. We therefore consider benchmarked performance to provide a more meaningful basis for distinguishing different levels of quality.
- We recognise that because the NSS is a survey of students’ perceptions, high or low NSS scores may in some cases be in tension with other elements of quality within the student experience criteria. Our proposals aim to mitigate this.
- We are proposing that NSS indicators would normally be weighted equally to the student experience submissions (50-50). This would mean that NSS scores could not, in isolation, lead to a particular rating or outcome.
- In Proposal 3 we set out that we would carry out an initial quality risk assessment. Where intelligence suggests a risk that students may not be receiving an appropriate level of academic challenge or sufficiently rigorous assessment, for example, we may carry out an enhanced assessment. Such risks could arise from intelligence we receive, rapid growth in student numbers, significant changes in patterns of recruitment, or other indicators that appear inconsistent with very high scores in NSS indicators. In these circumstances we would normally gather additional evidence through an enhanced assessment, as set out in Proposal 11. If we have serious concerns, we may carry out a quality investigation instead of a minimum requirements or TEF assessment. Where we gather additional evidence and this conflicts with the NSS indicators, we would put less weight on the indicators.
- If we have concerns about a provider unduly influencing NSS responses, such as bulk responses, misinformation given about the purpose of the NSS, or undue incentivisation of positive responses, we propose that we would put less weight on NSS data as part of the assessment. We plan to review our published approach to monitoring and investigating allegations of inappropriate influence on the NSS to ensure that this process is appropriately robust, given that TEF ratings and minimum requirements outcomes would be linked to restrictions and in the case of TEF, incentives. In the event that we find inappropriate operation of the NSS, we may suppress publication of the affected NSS data and disregard it in the assessment. We may also judge the student engagement criterion to be unsatisfactory, as inappropriate operation of the NSS would likely be in conflict with the student engagement requirements in condition B2. This could result in a rating of Requires improvement or an outcome of Unsatisfactory.
Further detail on the proposed approach to data used in assessment, including interpretation, consistency, structure and benchmarking factors is set out in Part 4 of the draft guidance.
Published 22 October 2026
Describe your experience of using this website
