What do scientists think about the biological claims made in the document about diversity written by a Google employee in August 2017?

I’ve analyzed Mr. Damore’s memo, read and watched Google’s responses, watched a video of one of Google’s diversity training courses, and read about a dozen media reactions: pro, con, and neutral. I’ve also familiarized myself with the research on the biological differences between males and females, implicit bias, and stereotype threat. Both Google and James Damore have reacted foolishly toward the current conflicts resulting from the politicization of work place gender representation. The fact that Mr. Damore’s document did not become an issue inside of Google until it became public, and that Google was contending with discrimination law suits when it was released, leads me to suspect that had it not been leaked or had it been leaked while Google was not being sued, Mr. Damore would not have been fired.

As for the science of his memo, it contains 32 references: 8 Wikipedia articles, 4 pop-science articles, 8 news articles, 1 each message group and blog posts, and 10 serious research journals or books. Excluding his claims about Google’s internal practices, he makes 38 claims pertaining to gender, diversity, bias, and/or work place practices. Of these, he supports 7 with authoritative literature, 10 with non-authoritative literature, and 21 have no support at all.

Page 6 expresses his fears and insecurities about his status, and page 7 consists of political statements to protect his status. A majority of his suggestions are politicized programs and alienating, which he suggested that Google not do, and he fails to consider the costs of his policy suggestions as he says that Google should do. Lastly, as others have pointed out here, while he gets most of the facts right from the standpoint of biological and evolutionary psychology, he fails to properly treat the full body of research on gendered personality influences that more comprehensively account for the nuances of epigenetic, genetic and environmental influences [1].

Having analyzed one of Google’s diversity training videos, I agree with Mr. Damore that at least this part of their diversity program does politicize the science [2]. Its weakness arises from its reliance on the Implicit Association Test (IAT) and a racial names resume study (Lakisha Test), each of which are weak studies that overly optimistic diversity researchers and promoters prematurely latched onto. That’s not to say that tolerance training is futile. When done well, it invites participants to become cooperative joint problem solvers, to reduce the effects of bias. Diversity training though, often done poorly, gives rise to anger and provokes a visceral response and promotes a toxic work environment [3]. Google’s diversity training video exemplifies a diversity program done poorly.

The IAT was published in 1998, promoted by Harvard, and thought to be able to detect unconscious biases [4]. That the IAT has low repeat-ability and is easily falsified is not new news (Google search left to the reader). A META analysis of 499 studies over 20 years involving 80,859 participants conducted in May 2016 by the University of Wisconsin at Madison, Harvard, and the University of Virginia, concluded that even if people do harbor unconscious biases, there is no evidence that these translate into discriminatory behavior. The IAT does however support a multi-million dollar industry for diversity consultants, and provides a brow beating cudgel for racial activists:

‘the correlation between implicit bias and discriminatory behavior appears weaker than previously thought [. . . and ] there is very little evidence that changes in implicit bias have anything to do with changes in a person’s behavior’

The “Lakisha Test” is likewise research of questionable validity and unknown reliability. It sought to detect hiring biases by measuring responses to resumes submitted with “white”and “black” sounding names. The research itself was riddled with poor controls, a small field of study, and lurking variables: 1) it was conducted in only two cities; 2) it only responded to newspaper ads; 3) the template resumes were subjectively padded by researchers who have no direct experience or training in the target career fields; 4) it failed to control for how the resumes were filtered - by humans or machines, how they were sorted, or for how many were received after the job(s) had been filled; 5) it had no data collection for the race, sex, or recruiting experience of the reviewers; 6) it was conducted almost 15 years ago (2004); 7) it was not a longitudinal cohort study; and 8) IT DOES NOT PREDICT WHICH APPLICANTS WOULD BE HIRED OR THEIR STARTING PAY. [5]

Both Google and James Damore, driven by fear and insecurity, fell into fools’ traps set by left wing and right wing political actors. Google, in response to work place discrimination lawsuits, let itself become infiltrated by diversity activists and “specialists”. Now embedded, Google will be hard pressed to root them out. Damore, by informing his political opinions with right wing rhetoric, instead of doing careful research and thinking for himself, lost his promising career and now earns speaking fees touring the country to tell his story. What a tragic fool’s tale for both parties.

[1] The Google Memo: What Does the Research Really Say About Gender Differences?, https://heterodoxacademy.org/the...

[2]

Unconscious Bias @ Work | Google Ventures

[3] Diversity training was supposed to reduce bias at Google . . . it backfired, http://www.latimes.com/business/...

[4] A Meta-Analysis of Change in Implicit Bias, https://www.researchgate.net/pub...

[5] Are Emily and Greg More Employable Than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination, http://cos.gatech.edu/facultyres...