When news outlets misrepresent people, they can create a biased narrative that’s difficult to change.
We chose to include former Google engineer James Damore in our “With Prejudice” series for two reasons: 1) there are significant discrepancies between the memo he wrote and how it was covered in the media, and 2) the media’s misrepresentations when the memo was first leaked have continued and can be seen in the coverage of more recent events — such as a Feb. 15 National Labor Relations Board memo and a Feb. 17 protest at Portland State.
In our analysis, we returned to the origin of the story. Our team analyzed Damore’s memo itself as well as three news articles that covered his firing last August. We found that the coverage repeatedly mischaracterized Damore’s arguments through faulty reasoning, subjective language and a lack of multiple viewpoints.
We quantified these distortions: CNN, The New York Times and The Washington Post received overall integrity ratings of 12 percent, 17 percent and 20 percent, respectively (lower scores mean less objectivity). On the other hand, we found Damore’s memo used more nuanced arguments and varied perspectives, and it received a rating of 74 percent.
We also interviewed Damore about how the media has reported on his memo, and he discussed several ways in which he believes the coverage misrepresented him. “Having open discussions without partially judging people as bigots — to do that seems important for society in general to survive,” Damore said. “If we can’t have these discussions, then the only way to solve our problems is through force and violence.” Read the full interview here.
Distortions in the media coverage
The following sections break down the distortions in the media coverage and the memo in terms of spin, slant and logic.
Note: The Knife’s analysis process was designed to assess the objectivity and distortion of news articles and other communications that are presented as factual. Damore’s memo differs in that it was intended to be internal company feedback. We analyzed it with the same standards that we use for news analysis, and this may account for some of the memo’s distortion rating.
The CNN, Times and Post articles we analyzed were 69, 77 and 79 percent spun, respectively (the higher the score, the more spun the article). To illustrate how spin works, consider the following sentences (spin words bolded):
“Storm at Google over engineer’s anti-diversity manifesto” (CNN)
“Google has fired the employee behind that controversial diversity manifesto” (The Washington Post)
Words like “anti-diversity” aren’t data-based or objective, and they frame the memo in a negative light. It’s misleading to call it an “anti-diversity” memo when in its first sentence, Damore says, “I value diversity and inclusion, am not denying that sexism exists, and don’t endorse using stereotypes.”
The word “manifesto” is also misleading as it can have negative connotations and it doesn’t accurately describe the memo. A “manifesto” is a statement that “publicly” expresses the author’s “intentions, motives, or views,” whereas the memo was meant to be an internal company document.
When we remove the spin words and replace them with more objective language, these sentences become something like:
Google has fired the employee who authored a memo about diversity
Not as emotionally charged, right?
Damore’s memo also contained some subjective language. For example, he said policies mandating equal representation of men and women in tech could have “disastrous” consequences and that political correctness is “complacent to the extremely sensitive PC-authoritarians.” However, there was significantly less subjective language in Damore’s memo overall, which is why it got a lower spin rating.
Unlike most internal documents, the news, in theory, is meant to be objective. It’s worth noting that an internal company document scored higher than the news outlets.
When a news outlet gives more weight to one perspective than others, we call it slanted. Here’s the main viewpoint that the media supported: Damore’s memo perpetuated harmful gender stereotypes, sexism and anti-diversity.
Here are two ways the news outlets slanted their coverage:
1. Cherry-picking information
The outlets selected parts of the memo that, when taken out of context, may suggest Damore was promoting stereotypes. For example, CNN says Damore wrote there are “higher rates of anxiety disorders among women.” This isn’t exactly what he wrote and it lacks context.
Citing a cross-cultural study about personality trait differences between the sexes, Damore wrote that women, on average, have more “neuroticism,” which is a psychological term for a particular personality trait. Individuals who score higher on this trait are more likely than average to experience anxiety, among other negative emotions.
Also, the articles left out most, if not all, of the suggestions Damore gave for how the gender gap could be reduced at Google. Had they included this information, the predominant view that the memo was “anti-diversity” may have made less sense.
2. Quoting critics and not supporters
While the memo had critics, there were others who supported it. The outlets didn’t cite supporters, however, which may give the impression that the public reaction has been mostly negative. This may further the view that Damore was wrong to write what he wrote.
Given this emphasis and cherry-picking, as well as other slant issues, the CNN, Post and Times articles rated 79, 79 and 64 percent slanted, respectively.
In comparison, Damore’s memo was only 22 percent slanted. Why was it less so? For one, Damore’s viewpoint was more nuanced than that of the media coverage. He said men and women differ “in part” due to biological reasons and that differences between the sexes “may in part” explain why there is a gender gap. The wording here is important because it leaves open the possibility that these differences may not explain the gap, and that there could be other factors as well.
He also acknowledges other points of view. For example, he said, “I hope it’s clear that I’m not saying that diversity is bad, that Google or society is 100% fair, that we shouldn’t try to correct for existing biases, or that minorities have the same experience of those in the majority.”
There were additional viewpoints that Damore did not explicitly acknowledge in the memo, and we explore these later in the analysis.
Note: The news outlets were also missing other key information that would have helped provide a more comprehensive understanding of the story, and they received lower overall integrity ratings due to these omissions. For example, they did not mention that Damore said Google asked employees for feedback after a diversity training, and that he submitted the memo as a response to this request. See the Data section of our Technical Sheet for a full breakdown of these missing points.
A fallacy is a flaw in reasoning that may appear logical, but it’s not based on valid reasoning. A straw man fallacy is when someone’s argument is misrepresented in a way that makes it easier to refute.
CNN and The Times both had straw man fallacies in their coverage. CNN said Damore argued that “women aren’t suited for tech jobs for ‘biological’ reasons.” According to The Times, Damore argued that the lower percentage of women at Google was a “result of biological differences instead of discrimination.”
It would be fairly straightforward to refute these claims if Damore had made them, but he didn’t.
What he wrote, for example, was that the differences in “the distribution of preferences and abilities of men and women differ in part due to biological causes and that these differences may explain why we don’t see equal representation of women in tech and leadership.”
That’s a different argument than what the outlets presented. He’s not saying women aren’t biologically “suited” for tech or that biology explains the gap “instead” of discrimination. He’s saying men and women differ “in part” due to biology and these differences “may” explain the gender gap. Furthermore, he’s talking about averages based on population level statistics, not all men or all women. In other words, counter to what CNN and The Times suggest, he’s not saying biology is the only factor, and he implicitly allows for the possibility that discrimination plays a role or that the differences he describes may not explain the gap.
More recent coverage
On Feb. 15, the National Labor Relations Board released an “Advice Memorandum” written by Jayme L. Sophir, an associate general counsel at the NLRB, concluding that Google didn’t violate labor laws when firing Damore. It said that Damore’s comments in the memo, such as “men’s prevalence at the top of the IQ distribution” were “discriminatory and constituted sexual harassment, notwithstanding effort to cloak comments with ‘scientific’ references and analysis, and notwithstanding ‘not all women’ disclaimers.” Parts of Sophir’s memo itself mischaracterized Damore’s memo — e.g., it does not refer to “men’s prevalence at the top of the IQ distribution.”
A few days later, Damore participated on a panel at Portland State University at which protesters damaged sound equipment and pushed a student volunteer. Prior to the talk, local paper Willamette Week called him a “Tech Bro” that said “Women Are Biologically Unfit to Be Engineers.” This is a misrepresentation; as discussed above, Damore’s memo does not say this.
Recapping Damore’s claims
Before we look at some other perspectives that aren’t included in the memo (or in the news coverage), let’s summarize some of Damore’s central claims:
- There are factors other than sexism or discrimination that could in part explain why Google does not have 50 percent female representation.
- There are differences between men and women on average, based on population level statistics. (He qualifies this by noting a number of these differences are small and there is significant overlap between the genders.)
- These differences may in part explain the gender gap in tech.
- Women and men may differ partly because of biological reasons.
Note that while the perspectives that follow were not explicitly stated in the memo, the memo does not deny or exclude them. Nor do they contradict the above claims. The purpose of exploring these other views is not to suggest that one perspective is absolutely right or wrong, but rather to promote viewpoint diversity and spur further thought, exploration and critical thinking.
Scientific opposition to the memo
Heterodox Academy (HxA) is described as a “politically diverse group of 1,700 professors and graduate students” that aims to increase viewpoint diversity in universities. Two of its directors, Sean Stevens and Jonathan Haidt, authored a series of articles reviewing academic literature that both supported and in some ways opposed some of the claims in Damore’s memo.
After a detailed and nuanced review, the authors concluded that “Damore is correct that there are ‘population level differences in distributions’ of traits that are likely to be relevant for understanding gender gaps at Google and other tech firms.” But it also pointed out some criticism of the memo. Perhaps the main criticism was that Damore did not explicitly acknowledge that “most researchers studying [gender differences] assume that biology, childhood socialization, and current context interact in complex ways, and most psychologists know that pointing to a biological contribution (such as a genetic or hormonal influence) does not mean that an effect is ‘hard wired,’ unmalleable, or immune to contextual variables.” Read the authors’ full review here.
The above HxA article also lists a number of experts that have written about the memo who both support and critique it. This New York magazine article also provides a summary of rebuttals from experts. One the main criticisms is what HxA pointed out above – that Damore didn’t adequately acknowledge the complex interactions between nature and nurture. Furthermore, some of these experts argue that nurture (socialization, discrimination, environment, sexism, culture, etc.) may play a larger role than biological differences when it comes to certain differences between the sexes (such as interests) and work-related outcomes (such as the representation of women in tech).
Damore addressed some of the criticism he received in our interview with him. Read the full interview here.
Limitations and considerations in science
One of the main tenets of science is falsifiability. A scientist makes a hypothesis and then tries to prove it wrong, and get others to prove it wrong, and the more they are unable to prove it wrong, the more confidence scientists have in the hypothesis. In the pursuit of falsifying a claim, theories that scientists thought were true turn out to be incorrect or incomplete, and are revised or rejected. When considering scientific knowledge, we should also allow for the chance that what isn’t considered a scientific possibility today may become a possibility in the future, and this could happen in the case of differences between men and women, and human traits and performance in general.
Also, the existence of differences between men and women doesn’t necessarily mean they can’t be changed in the future, even some biological ones. How much of what we’ve inherited – biologically, psychologically or socially – is outdated and malleable?
There is an ongoing nature-nurture debate when it comes to understanding differences between the genders. Let’s focus on nurture for a moment. How much do the differences between how we raise boys and girls contribute to the differences we see in outcomes between men and women? How much of it is due to media? Carefully designed studies can indicate how much a trait, ability or interest is based on nature or nurture, but it’s not always clear.
Finally, science doesn’t make aesthetic judgments about what we should or shouldn’t value. For example, right now, Western culture, broadly speaking, measures success based on largely material standards: how much money you make, what car you own or how big your house is. It similarly rewards more material results, such as working more hours and improving efficiency. What it generally doesn’t do, arguably, is measure or value compassion, or long-term ecological thinking — look at how we treat our environment, for example. So while it may be the case that differences between men and women — biological or otherwise — in part explain average differences in income between men and women, perhaps if societal values and measures of success changed, some of these differences could too.
Given all these different points of view, how are we to navigate complex and sensitive social issues? We believe the answer, in part, is civil discourse – in the media and elsewhere. It’s important to expose and rectify communication that goes against civil discourse, such as ad hominem attacks, misrepresentations, emotional outbursts, divisiveness or, in some cases, even violence.
Musa Al-Gharbi, a sociologist at Columbia university, suggests a few strategies to practice this on a personal level:
- Create an environment where your peers understand they don’t need to give up their entire worldview to consider your viewpoint or agree with you on one point.
- Instead of expecting people to adopt your language or values, try to understand where others are coming from.
- Be open to the possibility you may be wrong.
Here’s some suggested homework: find someone with whom you disagree and ask them for their permission to have civil discourse about a topic on which you disagree. Try to gain clarity on what parts of your argument are fact and which parts are opinion, and try reaching consensus on this. Explore as many viewpoints as possible and practice the above three strategies. This is one way we can change the world.
The Raw Data
Unspun and unbiased. These are the facts.
NLRB says Google did not violate labor laws by firing Damore
On Feb. 15, the U.S. National Labor Relations Board (NLRB) released an “Advice Memorandum” saying Google had not violated labor laws when it fired former Google software engineer James Damore last August. Before he was dismissed, Damore had filed a complaint with the NLRB alleging Google violated his rights as an employee.
Damore explained during an interview with podcast host Dave Rubin that, prior to being fired, he had attended a number of diversity meetings and programs at Google in which participants were asked for feedback. Damore responded to these requests by writing a memo discussing his views on Google’s diversity policies, saying he “wanted to have a one-on-one conversation with the diversity programs and really discuss the issues but it never really came to that.”
Damore emailed the memo to organizers of these programs in early July. After not hearing back from them, he sent the memo to individuals and email lists within the company and getting feedback from his co-workers. He said he then sent the memo to an internal email list called “Skeptics” which he thought would be “able to point out holes in [his] argument.” After sending it to the list, he said he received personal attacks such as being called a “misogynist” and a “terrible person” and some people sent emails to upper management and human resources asking that Damore be fired. Sometime after that, the memo was leaked to the public.
After the memo was released in early August, it was discussed and criticized online and in national media outlets. Google subsequently fired Damore on Aug. 7 for violating the company’s Code of Conduct, which prohibits “advancing harmful gender stereotypes in [Google’s] workplace.”
In January, Damore and another former employee filed a class action lawsuit against Google, which includes allegations such as “Google employs illegal hiring quotas to fill its desired percentages of women and favored minority candidates, and openly shames managers of business units who fail to meet their quotas — in the process, openly denigrating male and Caucasian employees as less favored than others.” It also alleges that “Google employees who expressed views deviating from the majority view at Google on political subjects … were/are singled out, mistreated, and systematically punished and terminated from Google.”
Damore’s 10-page memo, titled “Google’s Ideological Echo Chamber: How bias clouds our thinking about diversity and inclusion,” discusses what he calls “moral biases” and prejudices at Google, and comments on the unequal representation of women in software engineering and leadership.
The first sentence of the memo states, “I value diversity and inclusion, am not denying that sexism exists, and don’t endorse using stereotypes.”
With respect to bias, Damore says that Google “lean[s] left” and that this political bias and prejudice “maintains its hold by shaming dissenters into silence.” He says neither the right nor left is “100 percent correct and both viewpoints are necessary for a functioning society or, in this case, company.”
Regarding unequal representation of women in software engineering, the memo says:
- There exists a less than 50 percent representation of women in tech and leadership.
- Population level differences in distributions of preferences and abilities between men and women may in part explain unequal representation.
- A number of these differences are “small” and the distributions overlap, “so you can’t say anything about an individual given these population level distributions.”
- The differences in preferences and abilities of men and women may be in part due to biological causes.
Damore says in the memo, “We need to stop assuming that gender gaps imply sexism” and proposes options for reducing the gender gap in “non-discriminatory ways.” He says that he believes in gender and racial diversity and thinks “we should strive for more.” He also includes an evaluation of Google’s diversity programs.
See how the articles rate in spin, slant and logic when held against objective standards.
Total Integrity: 74%
Total Integrity: 20%
Total Integrity: 17%
Total Integrity: 12%