Abstract

Lack of diversity is a significant issue in academic STEM, whether from a social justice, economic, or scientific quality perspective. In this scoping review, we synthesise peer-reviewed evaluations of diversity interventions in academic STEM from high-income countries between 2011 and 2023. One hundred and eighty-four intervention evaluations are identified through database searches (Scopus, Web of Science Core Collection, and IBSS) and forwards and backwards citation searches of included articles. Interventions are predominantly conducted in the US and targeted at women and ethnic or racial minorities. Other dimensions of diversity such as disability, socio-economic background, and sexual orientation are rarely addressed. The most common type of intervention is one that provides support-related services, followed by diversity training such as bias awareness workshops. Those seeking policy-level change are seen less frequently. Evaluations are heterogeneous in outcomes studied and designs. Interventions were overwhelmingly positioned as successful, with considerations of failures or unintended consequences rare. The findings from this review emphasise the need for holistic evaluations of diversity interventions that provide transparency between intervention and evaluation goals and processes. If institutional commitment to diversity is sincere, it is vital that resources and incentives for such evaluations and their dissemination are reliably available and that measures are taken to discourage evaluations acting as ‘any change is good’ box-ticking exercises.
Original languageEnglish
Number of pages38
JournalStudies in Higher Education
Early online date6 Jan 2025
DOIs
Publication statusE-pub ahead of print - 6 Jan 2025

Funding

This research was conducted as part of the Reimagining Recruitment Project (Ref: EP/S012168/1), which was funded by the EPSRC Inclusion Matters call.

Fingerprint

Dive into the research topics of 'Increasing Diversity in STEM Academia: A Scoping Review of Intervention Evaluations'. Together they form a unique fingerprint.

Cite this