Advances in Health Sciences Education

, Volume 13, Issue 2, pp 181-192

First online:

mini-PAT (Peer Assessment Tool): A Valid Component of a National Assessment Programme in the UK?

  • Julian ArcherAffiliated withMedical Education Research Fellow to the Foundation Assessment Programme, University of Sheffield Email author 
  • , John NorciniAffiliated withFoundation for the Advancement of International Medical Education Research (FAIMER)
  • , Lesley SouthgateAffiliated withSt George’s Hospital Medical School, Medical and Healthcare Education
  • , Shelley HeardAffiliated withLondon Deanery
  • , Helena DaviesAffiliated withAcademic Unit of Child Health, University of Sheffield

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access



To design, implement and evaluate a multisource feedback instrument to assess Foundation trainees across the UK.


mini-PAT (Peer Assessment Tool) was modified from SPRAT (Sheffield Peer Review Assessment Tool), an established multisource feedback (360°) instrument to assess more senior doctors, as part of a blueprinting exercise of instruments suitable for assessment in Foundation programmes (first 2 years postgraduation). mini-PAT’s content validity was assured by a mapping exercise against the Foundation Curriculum. Trainees’ clinical performance was then assessed using 16 questions rated against a six-point scale on two occasions in the pilot period. Responses were analysed to determine internal structure, potential sources of bias and measurement characteristics.


Six hundred and ninety-three mini-PAT assessments were undertaken for 553 trainees across 12 Deaneries in England, Wales and Northern Ireland. Two hundred and nineteen trainees were F1s or PRHOs and 334 were F2s. Trainees identified 5544 assessors of whom 67% responded. The mean score for F2 trainees was 4.61 (SD = 0.43) and for F1s was 4.44 (SD = 0.56). An independent t test showed that the mean scores of these 2 groups were significantly different (t = −4.59, df 390, p < 0.001). 43 F1s (19.6%) and 19 F2s (5.6%) were assessed as being below expectations for F2 completion. The factor analysis produced 2 main factors, one concerned clinical performance, the other humanistic qualities. Seventy-four percent of F2 trainees could have been assessed by as few as 8 assessors (95% CI ±0.6) as they either scored an overall mean of 4.4 or above or 3.6 and below. Fifty-three percent of F1 trainees could have been assessed by as few as 8 assessors (95% CI ±0.5) as they scored an overall mean of 4.5 or above or 3.5 and below. The hierarchical regression when controlling for the grade of trainee showed that bias related to the length of the working relationship, occupation of the assessor and the working environment explained 7% of the variation in mean scores when controlling for the year of the Foundation Programme (R squared change = 0.06, F change = 8.5, significant F change <0.001).


As part of an assessment programme, mini-PAT appears to provide a valid way of collating colleague opinions to help reliably assess Foundation trainees.


Foundation programme multisource feedback reliability validity work based assessment