Ian Mannix
Cosmos newsroom
A readily available statistical tool has revealed the extent of gender and racial bias among leadership teams in Australia’s leading universities.
In fact by one measure the outcome shows “extreme” racial discrimination, and the authors say there is “negligible statistical likelihood of achieving racially unbiased outcomes in the future.”
The research, which is based on 2018 data from Australia’s most influential 8 universities (Group of 8, or Go8), is described in the US scientific journal PLOS One.
Professor Peter Robinson and complex systems researcher Cliff Kerr at the School of Physics at Sydney University, found “significant sex discrimination” in the appointments of university chief executives (Vice Chancellors) who were in office in 2018.
“However,” the authors say, “… at the same dates, extreme racial discrimination was implicit in the selection procedures for both Vice Chancellors and senior appointments in all these universities.
“The University of Sydney’s senior appointments were found to have had the most racially biased outcomes among the country’s eight main research universities. Significantly, there is negligible statistical likelihood of achieving racially unbiased outcomes in the future in any of the contexts considered, unless the selection procedures are significantly modified.”
The University of Sydney (USyd) welcomed the analysis.
A USyd spokesperson said: “We welcome evidence-based research that addresses issues of bias and discrimination in the workplace.
“We’re determined to be a place that promotes cultural diversity and combats racism, and our 10 year strategy reaffirms this commitment.
“Work is ongoing to increase the diversity of our staff at all levels in a considered and effective way. This includes our Aboriginal and Torres Strait Islander employment plan, as well as our ongoing support for culturally diverse and racialised staff and students.
“While this paper considers gender and race, we also have initiatives in place to widen access to the University for staff and students across the board.
“We’re currently working on additional ways we can support the removal of systemic barriers for culturally diverse people at the University.”
The Go8 Chief Executive Vicki Thomson says all its members are committed to cultural diversity and inclusion and have strategies in place to promote this.
“Increasing diversity is an ongoing mission for higher education institutions. This commitment includes addressing not only gender and race but improving access for members of the community with a disability and all underrepresented groups.”
How did the report come about?
The data analysis was done after an article in The Australian newspaper reported that “Of the 40 VCs, it noted that only 13 were female and just one was non-European in ancestry.”
The researchers read the article and concluded “no objective analysis was done” to determine just how unlikely such an outcome would be and to what extent it indicates bias in VC selection processes and panels.
The purpose of this study was not to reveal selection outcomes, although it did that, but to show how standard statistical methods based on the binomial distribution can assist discussions of anti-discrimination “by providing objective measures of the likelihood of observed appointment and selection outcomes, and estimates of any selection biases that may underlie such outcomes.
More: Discrimination hurts
“The aim is to provide objective measures that can be used to illuminate discussions of policy and strategy, and to measure progress,” the authors say in their paper.
“Many discussions of discrimination are hampered by the absence of easily accessible methods to analyse just how discriminatory an appointment process might be in the context of the available pool of qualified applicants.
“Such methods are well established, but are often not familiar to those working on antidiscrimination and related areas of public policy because training in these fields seldom includes the relevant mathematics.”
The research outlines in detail how to use the analysis to reach considered outcomes. “Details…are intended for readers with at least some undergraduate statistics, but are not needed to follow the discussion in the main document.
The webapp that computes the measures is publicly available. Users need only enter three numbers to use it: the total number of appointments, the expected number or fraction of people who are in a target group of interest (eg: women), and the actual number of appointees who belong to the target group.
Using the app and explaining the data assumptions, the writers conclude the following about sex discromination:
“If women constitute 50% of the qualified pool of applicants for Vice Chancellorships and the sex of applicants is irrelevant to the appointment process… we expect on average 20 women to be appointed. However, just as with coin tosses, a perfectly fair process can result in some random deviations from this number. So, the question is whether the actual outcome of na = 13 female VCs implies bias and, if so, how much?”
They concluded there is less than a 2% chance of this outcome in an unbiased selection process.
here is no evidence of strong sex discrimination in the country’s eight main research universities for senior appointments (i.e., Faculty Deans and members of their governing Boards or Senates) for those in office as of 2021.
In applying the tool to racial discrimination, where only one non-European VC level appointment was made among 40 roles, the authors conclude that the chances of only one non-European appointment being made were between one-in-a million and one-in-27-billion.
The tool can be applied to analyse biases in any group of appointments and for other target groups.
