Psychology has a diversity problem

Lack of diversity in psychological research is a serious problem that needs to be addressed, according to an Australian academic.

A study led by Associate Professor Mark Nielsen from The University of Queensland School of Psychology found an overwhelming sampling bias in developmental psychology towards research conducted in Western cultures.

“Over the last decade there has been an increasing call for psychological studies to move away from what we call a WEIRD-centric approach – that is, using research participants from a Western, educated, industrial, rich, and democratic background,” he said.

To investigate the extent of this problem, Professor Nielsen and colleagues, Daniel Haun (University of Leipzig), Joscha Kärtner (University of Münster) and Cristine Legare (The University of Texas at Austin) surveyed articles published between 2006 and 2010.

They focused on the journals Child Development, Developmental Psychology, and Developmental Science, and analysed the geographical region of the participants, where data were collected, and the affiliation of researchers.

“What we documented was alarming – 90 per cent of papers published relied on data drawn only from WEIRD participants,” Dr Nielsen said.

“Countries in Central and South America, Africa, Asia, the Middle East and Israel contain around 85 per cent of the world’s population, yet contributed less than three per cent of participants.

“This is a real problem, as quite often researchers will draw conclusions about their results stating ‘This study shows that children of X age will…’ without clarifying that their findings might not apply to all children or indeed anything more than a minority of the world’s children.

“Taking data collected only among children from privileged WEIRD backgrounds and assuming it represents the psychology of children growing up in any environment would be like a biologist studying a domesticated tabby cat and suggesting it tells them everything about a wild lion – it may reveal some things, but a lot will be missing.

“It is important to ensure that findings are replicated with participants of different cultures in order to explore whether results can be generalised, as these outcomes are not just used by scientists, but by parents, educators, and policy makers too.”

The research team analysed developmental psychology journals, but Professor Nielsen suspected that biased sampling would be found in other sub-disciplines of psychology.

“The onus is now on other sub-disciplines to demonstrate that they are different in their sampling strategies or to acknowledge that their conclusions and theories must be treated as tenuous until tested among less homogenous and globally unrepresentative participants,” he said.

Professor Nielsen said that although there is widespread acknowledgment of this problem within the psychological research community, there is little evidence of procedural change being implemented.

“Psychology, as a field, has in recent years become rightfully self-critical with crises of replication and statistical inadequacy.

“To continue ignoring the possible impact of environment and culture is both neglectful and bad science, and should be seen as great as these other crises which the profession is taking efforts to combat.”

The research is published in the Journal of Experimental Child Psychology.

Media: Associate Professor Mark Nielsen, m.nielsen@psy.uq.edu.au, +617 3365 6414; Dani Nash, UQ Communications, dani.nash@uq.edu.au, +61 3365 7436.