Saturday 26 May 2018

Can we trust social science?

Patricia Casey
Patricia Casey

Patricia Casey

'Science is science and facts are facts" according to dictum. However, a lack of political diversity in those undertaking research in the realms of social science, is now causing acute concern in academic circles.

Authors of a recent scientific paper argue that the social sciences once valued diversity of viewpoints as a means of enhancing creativity, problem solving and research. They believe that this diversity has now been eroded and that the political leaning of those working in research in these areas has become monochrome, restricting studies and viewpoints on major social issues as a result.

The issues affected are likely to vary - from society's response to substance misuse to the effect of marital breakdown on children or the role of poverty in crime.

The paper was written by a number of authors, including the renowned social and psychological scientists Jose Duarte and Jonathan Haidt. It was published online in the journal Behavioural and Brain Science. The title 'Political diversity will improve social psychological science' speaks for itself.

The five authors have been working together since 2011 to promote scientific diversity in what has become known as the Heterodoxy Academy. World-famous evolutionary biologist Steven Pinker has described their paper as one of the most important in the recent history of the social sciences.

The authors start by pointing out that researchers in the social sciences who describe themselves as "liberal" outnumber those who say they are "conservative" by a ratio of more than 12:1, and this has been the culmination of a gradual shift of several decades.

The authors argue that lack of political diversity can undermine the validity of social and psychological research by embedding liberal-leaning values into the research questions and methods.

This can create confirmation bias and the authors, themselves of "liberal leaning", believe that the absence of diversity among the peer reviewers of these papers limits challenges to methodological weaknesses and biases that emerge in the course of data analysis.

The inclusion of researchers with other opinions would strengthen the methodology, according to the writers. As an example, they describe the liberal-leaning researchers' rejection of the idea of stereotypes, which they believe increases prejudice. Yet, subsequent studies by one of the few conservative researchers in this area show that some stereotypes are accurate and this is now a robust finding.

The authors describe the plight of researcher Robert Putnam. Himself a liberal, he was at the wrong end of political correctness when he reached the conclusion that indigenous populations with high migrant populations tended to become withdrawn and isolated from their communities.

The authors also mention an anthropologist, Napoleon Chagnon, who was embroiled in a bitter row when he challenged the notion that some tribal peoples were not peace-loving, as previously portrayed, but often violent.

By coincidence, I came across a link on Twitter to a fascinating article in the Guardian newspaper on June 3 by Will Storr, the author of 'Selfie: How did we become so self-obsessed and what it's doing to us', in which he discussed the politics of self-esteem.

Storr records that this began with the work of a Californian politician John "Vasco" Vasconcellos, who was convinced that low self-esteem was the cause of a multiple social problems, including unemployment, educational failure, child abuse, domestic violence, gangland disputes and homelessness.

He persuaded the state of California to finance a task-force to promote self-esteem and to recruit the world's best researchers to explore it.

His ideas were initially pilloried by politicians, educationalists and cartoonists. But by 1987, he had become a national treasure and was lauded on major media outlets. A task force began hearing testimonies from the public and he organised teams to travel to schools to combat drug misuse using the mantra of self-esteem.

In 1988, the task-force and Neil Smelser, a retired sociologist, went public and announced that "the correlational findings" between self-esteem and social illness "are very positive and compelling". State after state instituted self-esteem courses and Vasco became the hero of moment based on the correlational findings.

Now, some of the task-force members have been interviewed by Storr and they claim that contrary to the findings reported, no such positive statistics were available. Instead, the results were mixed, but the legislature in California could not be presented with mixed results and the public was fed a falsehood.

According to those Storr interviewed, Vasco, as a politician, also had power within the university. "Scholarly gobbledygook" is how one of those working with Vasco described the findings.

These salutary papers and books now emerging are casting a troubling light on the power of uncritical, politically driven, social research.

How many other received wisdoms now accepted in public life are "scholarly gobbledygook"?

Health & Living

Life Newsletter

Our digest of the week's juiciest lifestyle titbits.

Editors Choice

Also in Life