For the past few years HESA has been producing data to support student:staff ratio (SSR) calculations for use by league table compilers and other interested parties.
The last review of this methodology took place back in 2015. Since then HESA has become the Designated Data Body for English higher education and has transferred a number of its former roles to Jisc. So Jisc now administers the supply of data for these purposes, and HESA has determined that it is no longer appropriate for it to “own” the SSR methodology.
With Jisc’s role being to administer and share the data, rather than to lead on these sorts of definitions, the Higher Education Strategic Planners Association (HESPA), through its HE Data Insight Group (HEDIG), has agreed to take up the mantle as custodian of this methodology.
Strategic planners in the sector have long been interested in the use of data to support SSRs and Matt Finn’s Wonkhe article back in May eloquently expresses some of their views. HESPA was an active partner in the last SSR methodology review and its members firmly believe that there is value in shared, sector-owned, and transparent methodologies for this sort of metric. Like them or loathe them, there is a clear consensus that moving away from a single reproducible methodology to multiple variants, such as a different approach for each league table, is undesirable.
Therefore, a working group made up of representatives from HESA, Jisc and a range of different types of higher education providers has undertaken an in-depth review of the current methodology. The review explored opportunities to improve the methodology, investigating a number of different options and potential improvements were categorised according to those that could be implemented with no change to data collection requirements and those that would require such changes. The former were prioritised for this review, with the remainder being noted for input to consultations on future collection changes.
Would an entirely different metric work better?
SSRs are generally used as a proxy for class size or contact time with academic staff, so potential alternative metrics which might better fulfil such purposes were explored. Options considered included use of attendance data or timetable data, but these would require additional data collection.
Another option considered was to calculate teaching staff cost per student FTE, however this could be misleading because high scores might be indicative of high numbers of staff, but they might equally represent smaller numbers of highly paid staff. The review group concluded that, at this time, no suitable alternative metrics were readily available.
Improving the accuracy of the staff FTE calculation
Reducing the FTE of staff classified as having responsibilities for both teaching and research would account for the fact that these staff will not spend 100 per cent of their time on student-related activities. Staff classified as teaching-only would remain as 100 per cent and those classified as research-only as zero per cent. There are a range of academic workload models in use across the sector so the proportion of time spent away from student-related activities varies between institutions, but it was felt that some reduction was appropriate.
Alternatives to a blanket FTE reduction were also examined. The review group looked at whether data on sources of basic salary could be used to identify the teaching element of staff time.
The relationship between the academic staff FTE and research funding was examined to identify whether a formula could be created to reduce staff FTE to account for research activity. Neither approach was found to be sufficiently robust.
Improving the accuracy of the student FTE calculation
A number of students spend a significant proportion of their time off campus learning in workplace settings and being taught by professionals in those settings rather than academic staff. As these professionals are not employed by the HEP, they are not included in the staff side of the calculation, so this is not offset in that way.
Course lengths vary, but will typically be over 40 weeks pa for these types of courses leading to professional registration, and placements can account for up to 70 per cent of time on some teacher training courses. A reduction in FTE for these students would improve accuracy and complement the approach to reducing the FTE of students on full-year industrial placements by 50 per cent.
Cost centre coding is currently the only subject-related classification that is common to all of the student, staff and finance HESA records, but increasing use is being made of the new approach to subject coding through HECoS and the common aggregation hierarchy.
The review group considered whether there would be benefits to changing SSRs to use HECoS, but concluded that this would not improve the accuracy of SSRs because it would be likely to create as many problems as it would solve. For example, many courses allow students to take one or more optional modules outside their specialism. It would also require more robust collection of HECoS data within the three records and would therefore take time to implement.
Having explored a broad set of options, the review group is consulting HESPA members, and others with an interest in this work, on two distinct proposals. If you would like to feed into this consultation, please complete this form by 31 October 2022 and we will share results in due course.