Proposals for a revised approach to assessing and rating quality

Key points of this proposal

• We would use a transparent set of factors to decide which providers to assess each year and when those assessments take place.
• We would normally seek to schedule providers for assessment when they have sufficient data, so they could apply for TEF ratings.
• However, we may schedule providers for assessment before sufficient data is available where other factors indicate that an earlier assessment is appropriate, including where there are concerns about compliance with the minimum requirements.
• We would take account of other factors for scheduling, including the timing of DAPs assessments, the timing of access and participation plans, and subcontractual partnerships where this would help improve alignment between regulatory processes or support more coherent assessments.
• We would normally reassess providers with a Requires improvement rating or Unsatisfactory outcome after two years.
• We would normally reassess providers with Satisfactory outcomes after three years.
• As well as linking the timing of subsequent assessments to the TEF ratings or minimum requirement outcomes, we would also take account of a range of other factors.

Explanation

Scheduling annual cohorts for assessment

49. Following stage one, we decided that for TEF assessments, we would assess providers in annual cohorts and that nearly all providers with undergraduate provision would receive a first assessment within three years. We decided that:
• we will take account of several factors when deciding when to schedule providers for their first assessments
• during the first year (2027-28) we will prioritise all providers with an existing TEF aspect rating of Bronze or Requires improvement
• during the second and third years we will assess all remaining providers with existing Silver or Gold TEF ratings, and nearly all remaining providers without existing TEF ratings.

50. We also said we would consult further on the factors we would use to schedule assessments.
51. We now propose additional factors that we would consider when scheduling a cohort of providers for assessment each year. Following the initial risk assessment (set out in Proposal 3), providers with sufficient data would be able to apply for TEF ratings. Those without sufficient data or that choose not to apply for TEF ratings would be assessed for compliance with minimum requirements. If we are concerned that a provider may not be meeting our requirements, we may carry out a quality investigation.
52. Table 2 below lists the full set of factors we would consider when scheduling annual cohorts of providers for assessments. It includes both the factors we decided following stage one, and the additional factors we are proposing, indicated as ‘New’. We also propose to prioritise the factors, for transparency.

Table 2: Scheduling factors

53. We would use these factors to select a cohort of providers for assessment each year. No single factor would determine scheduling decisions in isolation. We would consider the full range of relevant factors, giving greater weight to those identified as higher priorities.
54. Where possible, we will assess providers without TEF ratings early in the first assessment cycle if they ask us to, so they can access the benefits. If we receive more requests than we have capacity for, we will use the wider scheduling factors to decide which providers to assess first.
55. During the first year we expect to assess a smaller number of providers than in the following two years. We anticipate assessing up to 120 providers in the first year, and up to 150 per year in the following two years. This will give us time to put the new processes into practice and improve them during the first year. Starting with fewer assessments will help us and the assessment teams identify and resolve practical problems, making assessments more efficient in later years.
56. Our reasons for proposing the new factors are as follows:

• • Giving high priority to data availability would allow as many providers as possible to apply for a TEF rating based on enough evidence. This supports our decision after stage one to assess providers whenever there is enough data. We propose scheduling an assessment when there is enough data for at least one aspect and, where possible, both aspects. For example, if data is available for one aspect and we expect more data soon, we would probably delay the assessment so that it can cover both aspects.
  • Considering the timing of DAPs quality assessments would avoid unnecessary duplication and burden. For example, if a provider is due to undergo a quality assessment as part of its DAPs application, we would consider how best to sequence this with its TEF assessment. This is likely to only apply to providers that have already applied for DAPs.

57. As Proposal 11 explains, in the case of subcontractual provision, we may sometimes assess a lead provider and its main delivery partner or partners in the same year. This is because the quality of subcontractual provision may be relevant both to the assessment of the delivery partner and to the assessment of the lead provider, which remains responsible for the quality of that provision. Lead providers would be required to include evidence about subcontractual provision in their submission. Assessing partners within the same assessment cohort would therefore allow us to consider relevant evidence in a more joined up way. We would not expect to do this in all cases but may do so where it would support a more coherent assessment of quality.
58. In the stage one consultation, providers were interested in timing TEF assessments and access and participation plans so that each could inform the other. The two processes often support each other. The TEF looks at a provider’s past performance, including differences in students’ experiences and outcomes. Access and participation planning requires providers to develop ambitious and realistic plans to tackle barriers faced by underrepresented groups, and to assess how well those plans work.
59. Ideally, we would carry out a TEF assessment first so that its findings can inform a provider’s future access and participation plan. We therefore propose using this as one factor when scheduling TEF assessments.
60. However, we may have few opportunities to do this in the first year of the new TEF because approved access and participation plans run for several years and we will prioritise providers that already have Bronze or Requires improvement ratings. We may have more opportunities in later years, but we would not give this factor priority over other scheduling factors.
61. If the TEF identifies improvements that a provider could make by changing its access and participation plan, the Director for Fair Access and Participation may work with the provider to agree those changes.
62. We will review this approach and consider how to further improve the sequencing of TEF assessments and access and participation plans in the future.

Subsequent assessments

63. Following stage one we decided that:

• • we will normally reassess providers with Gold ratings after five years, Silver after four years and Bronze after three years
  • we will base the timing of reassessment on a provider’s lowest, or only, aspect rating.

64. Because a Bronze rating and Satisfactory outcome, would be broadly equivalent outcomes, we propose the same reassessment interval of three years would normally apply where a provider has either outcome.
65. In the stage one consultation, we said we would decide case-by-case when to reassess a provider with a Requires improvement rating. This would allow us to address concerns in a timely way.
66. Providers judged to require improvement represent risk to students and a need for rapid improvement. We therefore propose scheduling the provider’s next quality assessment after two years. In practice, this is the earliest we could reassess a provider. For example, a provider is selected for assessment in summer 2027. It submits evidence in December 2027, receives a provisional rating of Requires improvement in summer 2028 and its final rating in January 2029 (if it makes representations). We would normally select it for assessment again in summer 2029 and it would need to submit again in December 2029, two years after the first assessment.
67. Because a Requires improvement rating in the TEF, and an Unsatisfactory minimum requirements outcome, would be broadly equivalent, we propose the same reassessment interval of two years would normally apply where a provider has either outcome.
68. Once we select the provider for reassessment, as with all other providers, we would complete a risk assessment (see Proposal 3). This would assess the appropriate level of evidence gathering and scrutiny for the quality assessment, or whether we should investigate the provider instead.
69. We considered whether reassessment following a Requires improvement rating or Unsatisfactory outcome should focus solely on the specific concerns that led to the original judgement. This might allow for a more targeted assessment, for example where a provider was judged as unsatisfactory on specific criteria, the reassessment could focus solely on those criteria. However, our view is that quality should be reassessed in the round. A provider’s overall performance may have changed since the previous assessment and new evidence may have become available.
70. For that reason, our preferred approach is that reassessment should normally be a full assessment rather than a more limited review.
71. We would expect providers to demonstrate how they have responded to the concerns of the earlier assessment and what improvements they have made. However, the subsequent assessment would continue to consider all relevant evidence and criteria.
72. Where a provider improves and is awarded a higher rating or outcome than Requires improvement or Unsatisfactory, it would return to the standard reassessment interval associated with its new outcome. Where a provider does not improve at its next assessment, it would normally experience increasingly significant regulatory action (see Proposal 21).
73. Following stage one, we decided to link the timing of subsequent TEF assessments to the ratings awarded and ongoing risk monitoring. We now propose the expanded set of scheduling factors above would apply both to providers’ first assessments under the future quality system and to subsequent assessments.
74. In developing the operational detail of the system, we now consider that many of the factors that we decided are relevant to scheduling a provider’s first assessment would remain relevant for subsequent assessments. For example:

• • If a provider applies for TEF ratings and receives Silver ratings and undergoes a merger four years later when it would be due for reassessment, we would consider whether it is appropriate to delay its next assessment until after the merger has concluded.
  • If a provider has Satisfactory outcomes and wants to access the benefits of TEF ratings at its next assessment, we would consider the availability of data when scheduling its next assessment.

75. We therefore propose to apply a single set of factors when deciding the timing of providers’ first and subsequent assessments.
76. We would continue to give high priority to a provider’s lowest TEF rating or minimum requirements assessment outcome. We would normally reassess providers after five years for Gold, four years for Silver, three years for Bronze or Satisfactory, and two years for Requires improvement or Unsatisfactory. This may then be adjusted depending on the other scheduling factors.