The New York Times recently ran a long piece exploring the history of women in STEM fields and attempting to explain the ever-present difference between men and women in performance and participation in these fields. The article begins by citing research on perceptions of female aptitude in math and science:
“Researchers at Yale published a study proving that physicists, chemists and biologists are likely to view a young male scientist more favorably than a woman with the same qualifications. Presented with identical summaries of the accomplishments of two imaginary applicants, professors at six major research institutions were significantly more willing to offer the man a job. “
She shares an anecdote that is supposed to display the prejudice of professors against females in the field, but instead illustrates one valid reason for the bias displayed by the Yale study:
“Other women chimed in to say that their teachers were the ones who teased them the most. In one physics class, the teacher announced that the boys would be graded on the “boy curve,” while the one girl would be graded on the “girl curve”; when asked why, the teacher explained that he couldn’t reasonably expect a girl to compete in physics on equal terms with boys.”
Enter Bayes’ Theorem
Bayes’s theorem is a foundational principle of statistics and probability that allows us to update our estimations about the trueness of a fact based on new evidence. The math of Bayes’ theorem is simple and elegant, and the overarching idea is powerful — we can use evidence in a formalized manner to change the probability that something may be true, and this can often have non-intuitive results.
The classic example of Bayes in action is medical tests — for example, if 1% of women have breast cancer, and a mammogram detects the cancer 80% of the time with a 10% false positive rate, what is the probability that a positive result means the woman has cancer? If a mammogram is positive, the chance of cancer is less than 8% due to the presence of false positives, as well as the low baseline population rate of cancer.
What does this have to do with women and STEM fields? Readers of this site are familiar with the allure that even a plain looking girl can have at the height of her availability and youth. This isn’t just a factor when getting free drinks at the bar – it extends to the classroom, hiring for jobs, treatment in everyday life, and many other areas. Girls in primary and secondary school are judged to be better students, despite boys showing a significant advantage in standardized tests starting around middle school. The article highlights the ways that women are supposedly discouraged by the system, but makes no mention of the advantages they enjoy.
Put simply, women are more likely to be handed accomplishments without having to work for them, both due to the power of their sexuality and as unconscious overcorrection for their supposed disadvantages in opportunity. Given an applicant with a certain pedigree – a Ph.D, say, from a top graduate program —we will have a certain estimation of that person’s intelligence and aptitude. However, the “false positive” rate on those qualifications identifying extremely high aptitude is likely to be much lower for a man, who has not enjoyed the advantages of a feminized education system, catch-up programs, and the hint of his sexuality influencing the evaluations of his superiors.
The bias against hiring a woman whose qualifications are equal to a man, and their subsequent lower salary offer, is simply a use of Bayesian inference. It accounts for the implicit probability that the female will not be as good as her résumé suggests, to say nothing of the chance that she will leave her job to begin a family and leave her employer empty-handed at some point in the future. If, as the example above states, both men and women implicitly behave as if men are superior in math and science, we must give some consideration that this is a possibility.
If Men Are Better At Math/Science — What’s The Big Deal?
The media is encouraged to sing the praises of women where they excel compared to men, and females indeed show demonstrated advantages in many cognitive areas. They are better at language acquisition, picking up on non-verbal cues, and we are all familiar with their evolved capacity for psychological manipulation. Many would suggest that women have better organizational skills. They are incarcerated for violent crimes less often, are less prone to risky behavior, and are more resilient to psychological trauma such as PTSD.
But when it comes to exploring why men have long-demonstrated advantages in certain disciplines, the media scrabbles to ascribe the boogeyman of injustice perpetrated on the protected class. The article is quick to dismiss the repeatable and longitudinal difference between males in females in standardized testing, a long-standing form of evaluation that every college and grad school uses to give out valuable admissions spots. It also does not mention the lack of female representation in technology entrepreneurship, a field that is less dependent on credentials and more on individual drive, creativity, and aptitude.
It could certainly be true that women are discriminated against AND that they are simply less common at the far right of the aptitude bell curve necessary for competitive positions in academia. But I challenge you to find this idea entertained in any mainstream publication despite the mountains of circumstantial evidence. Larry Summers was tarred and feathered for even mentioning research on population dynamics as a potential driver of this difference. The lesson here is that, when you begin an “inquiry” by presupposing the conclusion, you will end up with a politically correct and eminently intellectually dishonest worldview.
Read More: The Anti-Male Commercial
