Recent scandals relating to care failings within the NHS have led the UK government to recommend that providers examine the recruitment methods for healthcare professional education programmes and initiate better screening of those entering the professions (Francis, 2013). The School of Healthcare Sciences at Cardiff University has committed to interviewing all applicants prior to enrolment and have instigated a multiple mini-interview (MMI) structure to do so.
In MMIs candidates have many opportunities to make a first impression, meeting different assessors at each station, suggesting the process is fairer and more consistent when compared to traditional panel interviews (Eva et al, 2004). However, if MMIs are designed to select for specific attributes and personalities, do they result in a homogenous student population and thus reduce the diversity of experiences, thoughts and behaviours within? Is the process which is thought to be 'fair' actually fraught with bias?
This project aimed to investigate bias within the MMI structure for Physiotherapy recruitment at Cardiff University. It considers the design and scoring of interview stations and their inclusivity, through the monitoring of performance at each station by applicants with differing characteristics.
A retrospective review of 306 applicants interviewed during the 2016 admissions cycle mapped characteristics against scores at each station for each applicant, with ethical approval granted via Cardiff University. Gender, age, disability, previous educational level and socio-economic backgrounds were analysed against interview scores to evaluate for any significant differences between characteristics and performance.
Detailed scrutiny performed using Mann Whitney U, Kruskal-Wallis tests and pairwise comparisons revealed the following key findings:
- 26-29 year olds out-perform 18-21 year olds
- Those with and without a disability performed equally well.
- Applicants who had already attended higher education scored significantly higher than those who had been schooled in further education, comprehensive, independent and academy environments
- With the exception of Grammar schools, the male candidates performed consistently less well compared to their female counterparts from the same educational background provider.
Cost and savings
Running multiple mini-interviews has facilitated efficiency savings in terms of time. 6 candidates can be interviewed within a 45 minute timeslot, whereas previous panel-style interviews meant each candidate utilised 20-30 minutes of time. The interviews are completed within 10 days, providing much earlier feedback and decisions for candidates.
This study contributes evidence to suggest that the MMI process is not inherently biased however further work is needed to facilitate improved performance from male candidates. The design of individual MMI stations should promote inclusivity, and ultimately encourage the physiotherapy profession to explore issues of equality and diversity within its ranks.
Within the Physiotherapy profession there remains an under-representation of males, black and minority ethnic (BME) backgrounds and those from socially deprived areas. Whilst this study did not examine ethnicity of applicants it has shown the male applicants perform less well in recruitment processes. An absence of exposure to this diversity in the profession itself may lead to a lack of empathy and a failure to meet the needs of those particular groups as both patients in the clinical arena but also within recruitment to the profession in higher education. Further work is required on promoting and facilitating access to the profession in these under-represented groups.
Top three learning points
- Multiple Mini Interviews do not appear to be adversely affected by bias within the process
- Male applicants to physiotherapy do not perform as well as female applicants
- Underrepresentation from some groups remains within the profession and more needs to be done to recruit these
This work was not funded.
This work was presented at Physiotherapy UK 2017.
For further information about this work, please contact Jill Morgan at: MorganJ63@cardiff.ac.uk