Wednesday, May 25, 2011

The dangers of pseudoscience

- by Stacey Prince

Last week Psychology Today published an article claiming that Black women are less attractive than women of other ethnic groups. The author, Satoshi Kanazawa, has a history of disseminating racist and sexist conclusions based on weak pseudoscience. After a widespread public outcry, Psychology Today removed the article from their website. However, the damage had already been done, and many readers may not have been aware of the on-line retraction. By publishing Kanazawa’s article in the first place the magazine gave validation to a racially biased finding and its underlying message that Black women are less valued than women of other ethnicities. It’s shameful that Psychology Today gave a platform to this author, who has a history of promoting other equally biased and discriminatory beliefs (click here for a summary). It’s also shameful that this widely read magazine, which puts psychological findings in an accessible, non-academic format and is read by millions, chose to reference conclusions that are so scientifically flawed. In addition to the fact that “Black” is not a definitive category in the first place and that he did not consider the confounding effect of social bias, Kanazawa's conclusions were drawn from another, separate study on adolescent health and behavior that did not even purport to measure the construct he was trying to draw conclusions about. You can read a thorough critique of his findings here.

The problem is, many people reading Psychology Today will not necessarily have the critical analytic skills to understand that Kanazawa’s conclusions are unfounded, and will take them at face value. Those of us who went to graduate school or took undergrad scientific methodology classes have some valuable skills for evaluating research findings, whether they appear in trade or academic venues. For example, we know that correlation does not imply causation, that selecting an appropriate, representative sample is key, and that tiny statistical findings, even when they have significant ‘p values,’ do not necessarily translate to concrete real-world differences. Even if you haven’t had formal training in evaluating research findings, there are some basic things to look for that are “red flags” warning of pseudoscience. This recent article highlights a few such red flags, including being wary of studies that purport to generalize findings from a small and/or non-representative group of participants to everyone, those that try to quantify something that is highly subjective, personal and culture-bound (such as physical attractiveness), and those that don’t reference any conflicting data or possible limitations to their findings.

Unfortunately there is a long history in Psychology (as well as other sciences) of pseudoscience being used to support inaccurate and harmful racist, sexist and homophobic conclusions. For example, the claim that Black people are less intelligent than individuals of other ethnicities based on IQ testing has been roundly debunked, due to the extremely contextual nature of the skills that are actually assessed by such tests, skills that are largely impacted by social and cultural factors including affluence, access to quality education, and nutrition. Despite this, supposed differences in IQ based on race (which itself is a socially constructed concept) have been used to justify slavery and eugenics. Such pseudoscientific findings also uphold the myth of meritocracy - the belief that life presents a level playing field to individuals of all social groups and that everyone can get ahead if they just work hard enough, thereby blaming the individual when success is not forthcoming and ignoring systemic inequities. You can find a very useful summary of scientific racism here. Similarly, conversion therapy (efforts to change an individual's sexual orientation from gay or lesbian to heterosexual) is promoted, often with conservative religious motivations, based on faulty and flawed studies, despite the fact that it has been shown to have significant negative mental health effects for many clients (click here for an excellent review).

Of course, it is also true that empirical evidence can be incredibly useful in advancing social justice – take, for example, the use of data on children of gay and lesbian parents in testimony and amicus briefs to support legislation legalizing adoption by same-sex couples. On the other hand, it can be harmful. The dissemination of findings like Kanazawa's, for example, has the potential to encourage poor self-esteem, eating disorders, and internalized oppression in Black girls and women, let alone biased treatment by others. It was irresponsible of Kanazawa to write the article in the first place, but it was even more irresponsible of Psychology Today to publish it. So, the take home message for readers and consumers of research is this: don’t believe everything you read, and try to develop and utilize critical thinking skills when digesting research findings, whether in mainstream media or professional journals. If you are a therapist and your client brings up research such as this, you can be of great assistance by helping them to analyze the validity of the findings. If you are a mentor or research supervisor, you can help your students understand the social context in which knowledge is accumulated, uncover researchers' underlying assumptions, and when necessary be able to challenge conclusions that are held up as "knowledge" by the powers that be in their community.

Saturday, May 28, 2011

Here's an interesting update: In a landmark case in the UK, a psychologist has been found guilty of malpractice and her license suspended for "treating" a gay man for homosexuality. While this was only an internal sanction of a licensing body rather than a legal ruling, it is still a huge step. In this case the real science regarding conversion therapy - including the fact that it is harmful to more than half the clients treated, causing depression and suicidality, and lack of sound scientific evidence that sexual orientation can be changed - overcame the pseudoscience offered by proponents of conversion therapy, often to veil their underlying religious and values-based motivations.

Here's another update: Psychology today has published an apology on their website, and say they are taking measures to ensure that something like this does not occur again. I would love for them to publish or discuss those measures as I think it would be helpful to all of their readers to understand how to watch out for inappropriate and offensive misuse of science in the media.

No comments:

Post a Comment