Fast forward to this afternoon.
Twitter is buzzing about an article by Satoshi Kanazawa entitled "Why Are Black Women Less Physically Attractive Than Other Women?" published on Psychology Today. Say what now? Black doesn't crack! Didn't anybody teach him that? But I can be objective. Before jumping on the bandwagon to defend my race, I decided to read his article. Psychology Today removed the article but you can read it here.
Let's break it down.
At the end of each interview, the interviewer rates the physical attractiveness of the respondent objectively on the following five-point scale: 1 = very unattractive, 2 = unattractive, 3 = about average, 4 = attractive, 5 = very attractive.The physical attractiveness of each Add Health respondent is measured three times by three different interviewers over seven years.
OBJECTIVE? Excuse me? How is that an objective measure of attractiveness? Perhaps he needs to sign up for a basic research methods class and then relearn terms such as objective and subjective. Does he really think that little of his readers? Does he really think that that would fly as an objective measure?
And who are these 3 interviewers? What are their races? Attractiveness is a cultural construct and hence, highly subjective. An Asian male may be viewed as attractive by fellow Asians but merely average by a Caucasian rater. Kanazawa has conveniently ignored this. He has a history though. (Yes, I know that's a Wiikipedia link and not the best source but you can follow the references to learn more about Kanazawa).
At this point we don't even need to read the article any further. If your operational definition and measurements are flawed, your results will be flawed. But let's read a few more sentences.
From these three scores, I can compute the latent "physical attractiveness factor" by a statistical procedure called factor analysis. Factor analysis has the added advantage of eliminating all random measurement errors that are inherent in any scientific measurement.
Factor analysis does what? Maybe I need to retake statistics because my stats professors never told me that factor analysis eliminates all random measurement errors. Wow. Factor analysis is better than I thought! I am actually thinking about grabbing my stats books and relearning factor analysis. I've been missing out. Who is Kanazawa writing this stuff for? I've seen from his website that he has several peer reviewed articles. Blogs on Psychology Today are not peer reviewed so bloggers can post a wide range of articles from scientific to unnscientific. I would hope that in his peer reviewed articles he pays close attention to basic research methodology and statistics.
Wait, what was that? He doesn't?
At the end of the article, he proceeds to explain his findings as a function of testosterone levels. Blacks have higher levels of testosterone than other races. Considering that his measurements are flawed, it isn't even worth addressing the testosterone issue. Additionally, I would need to do research on testosterone levels in various races. I simply cannot take his word for it.