Monday, June 8, 2015

Social Science: an (unintentional) outlet for progressive politics

On Friday, the Wall Street Journal ran an editorial decrying social science as a liberal conspiracy. Their accusation is based on the recent discovery that a graduate student at UCLA fabricated data for a study investigating conservative views about gay marriage.

The editorial irritated a lot of people in science media, but I thought Jesse Singal's take on the matter over at Science of Us was particularly interesting. Singal effectively rebutted the Journal's editorial, though his answer to the bigger question about the credibility of social science wasn't as convincing.

So is social science a liberal conspiracy? My reflexive answer to that question is yes. After putting a little more thought into the question, my answer is still yes, however it needs to be qualified. Social science doesn't exist to promote a certain political platform, but that's what happens regardless of intentions, and that's what has happened for the last 50 years.

Before anyone complains, let me point out that I'm not accusing anyone of dishonesty; I'm merely pointing out an unfortunate characteristic of human nature: people believe what they want to believe, and academics are no exception. This is why research funded by cereal companies consistently finds that breakfast is the most important meal of the day, to cite the dumbest example I can think of. The same goes for any study produced by an organization with a vested interest in the subject. Anti-smoking groups, soda companies, big tobacco and conservative think tanks all prove my point nicely.

It's a serious charge to accuse an entire field of promoting a political agenda, even if unintentionally. But I'm not alone in making such an accusation. To plagiarize myself from a post last year on the subject of rape culture, it's no secret that
the social sciences ... are regularly criticized for not being rigorous enough. The results of research in these fields are often highly subjective and difficult to quantify, to the point that researchers in other fields want the National Science Foundation to stop funding social science research. In addition, social scientists have been accused of being ideologically biased and engaging in group think--by other social scientists.
Singal and others would no doubt protest this assessment of the situation. Indeed, in response to the Journal's version of this argument, Singal cites NYU psychologist Jonathan Haidt, who has made a name for himself in part by highlighting just how politically slanted social science is as a field. Because of Haidt's work, Singal argues, social scientists are now having an internal discussion about how politics is affecting their research, so they can't be "mindlessly promoting liberal causes."

It's certainly true that Haidt has brought an important issue to everyone's attention, but the fact that one researcher had to come forward and blow the whistle on his colleagues is not a sign of academic rigor in the field, just very clear evidence that social science has a credibility issue. To cite the same paper Singal cites, Haidt's argument can be summarized like this:

[A] lack of political diversity can undermine the validity of social psychological science via ... the embedding of liberal values into research questions and methods, steering researchers away from important but politically unpalatable research topics, and producing conclusions that mischaracterize liberals and conservatives alike ...

This isn't to say that social scientists are incapable of conducting sound research, because we've learned a lot about a variety of important subjects from their work, human sexuality and personality psychology being two of my favorite examples. Nonetheless, with the good comes an awful lot of bad: useless studies based on unreliable surveys, conclusions that can't be quantified and results that have been outright fabricated, and the examples of this shoddy research go on for days and outmatch anything you might see in the physical sciences.

So if anything, I'd say the Journal reached the right conclusion for the wrong reasons. It isn't the one example from UCLA that illustrates how severe of a credibility issue we're talking about, but several decades worth of politically biased, methodologically dubious social science research.


  1. It might be more accurate to say that social "science" always serves the needs of the current aristocracy. If you look at books written 100 years ago, you'll find sociologists and psychologists "researching" and "finding" "results" that agreed with official wisdom in 1910. The "results" are wildly different from current "results" because current aristocrats have a wildly different set of taboos and requirements. But the corruption is consistent.

  2. I don't agree entirely. There's a lot of research that defies the accepted wisdom, yet it still gets published. It's often buried in obscure journals and usually isn't given much media coverage, but it does exist. Much of what I write here is based on those papers. See the two linked examples in the second to last paragraph of this post for examples.