Flawed selection criteria for redundancies

Management at the University of Liverpool informed the UCU Branch on the 25th January that it intended to make 47 people redundant on the application of 2 research metrics. Those were: mean grant income (applied using REF Units of Assessment) and a Field Weighted Citation index (FWXI) score.

Evidence subsequently submitted to the University by UCU showed that the redundancy scheme was based on the manipulation of REF data and breached the University’s REF 2021 Code of Practice. Further evidence submitted by UCU showed that FWCI could not be meaningfully applied to individuals.

The authors of the Hong Kong Principles  and The Leiden Manifesto endorsed UCU’s view in public, as did the Universities UK body, the Forum on Responsible Research Metrics. Similar concerns were presented to Senate in a paper that had wide support across the University. The Vice Chancellor refused calls for a vote on the proposals by several members of Senate.

As a result of this evidence and pressure, the University published a new redundancy selection criteria on the 5th May. This new criteria identified 32 people at risk of redundancy.

This new criteria retains research grant income as the primary redundancy selection element for selecting staff. It also adds what it calls a “rounded assessment of individual contribution.” The same “mitigation elements” that were outlined in the previous business case have been applied.

Grant income

Instead of linking the mean grant income figure to REF Unit of Assessment, the University used the codes assigned to staff members for the purposes of national data collection by the Higher Education Statistics Agency (HESA). UCU has produced a report to show that this criteria is based on data that is not fit for purpose and contains fundamental statistical errors. In summary, this report demonstrates:

  • The University has used inappropriate grant income thresholds, which our models demonstrate >50% of Russell Group staff in matched areas do not meet.
  • HESA cost centres are designed for HESA reporting purposes only, not for the purpose of setting redundancy thresholds.
  • Over half of those selected for redundancy have been measured by thresholds that were artificially inflated because inaccurate HESA cost centres were used.
  • The University has applied HESA codes in a completely inconsistent manner, using different thresholds for staff working in the same research field; in one example, the threshold for some staff is 5 times lower than for others working in precisely the same research subject.

Read the full report here.

The “rounded assessment of individual contribution”

The new selection criteria purported to introduce a ‘qualitative assessment’ that supplemented the research grant income targets. This “rounded assessment of individual contribution” was introduced to counter some of the criticism outlined above. However, it is UCU’s view that this is a sham qualitative assessment, for the following reasons.

  • Not one person from the original group of 47 at risk of redundancy was removed when the qualitative criteria were applied; all we removed because of changes to the grant income criteria
  • None of the staff selected for redundancy have been given any information about the way those post hoc qualitative criteria were applied or calculated.
  • The “qualitative” targets are set at an unfeasibly and unreasonably high level that allowing managers maximum discretion over who they fire.

Read the full report here.