How feminism has made me a better scientist

Feminism is not a branch of science. It is not a set of testable propositions about the observable world, nor is it any single research method. From my own perspective, feminism is a political movement associated with successes such as votes for women, setbacks such as the failed Equal Rights Amendment, and continuing struggles in areas ranging from reproductive freedom to equality in the workplace. Feminism is also a way of looking at the world through awareness of the social and political struggles of women, historically and in the present. And others will define feminism in other ways.

How is this relevant to the practice of science? As a researcher in statistical methods and applications, and I have found feminism to help me do better science. I make this claim based on various experiences in my work.

To start with, statistics is all about learning from data. Statisticians from Florence Nightingale and Francis Galton through Hans Rosling and Nate Silver have discovered unexpected patterns using mathematical modeling and data visualization. What does that have to do with feminism? Feminism is about a willingness to question the dominant, tacitly accepted ideology. This is essential for science. What is labeled “non-ideological” basically means the dominant ideology is accepted without thought. As Angela Davis said in her lecture, Feminism and Abolition, “feminist methodologies impel us to explore connections that are not always apparent. And they drive us to inhabit contradictions and discover what is productive in these contradictions. Feminism insists on methods of thought and action that urge us to think about things together that appear to be separate, and to disaggregate things that appear to naturally belong together.”

Along with questioning the dominant, tacitly accepted ideology is the need to recognize this ideology. This is related to the idea that a valuable characteristic of a scientist is a willingness to be disturbed. We learn from anomalies (see discussion here), which requires recognizing how a given observation or story is anomalous (that is, anomalous with respect to what expectations, exactly?), which in turn is more effective if one can identify the substrates of theories that determine our expectations. An example from our statistical work is our research using the Xbox survey to reveal that apparent swings in public opinion can largely be explained by variations in nonresponse.

On a more specific level, feminism can make us aware of work where the male perspective is unthinkingly taken as a baseline. For example, a paper was released a few years ago presenting survey data from the United States showing that parents of girls were more likely to support the Republican party, compared to parents of boys. I’m not completely sure what to make of this finding—for one thing, an analysis a few years ago of data from Britain found the opposite pattern—but here I want to focus on the reception of this research claim. There’s something oddly asymmetrical about how these results were presented, both by the authors and in the media. Consider the following headlines: “The Effect of Daughters on Partisanship and Social Attitudes Toward Women,” “Does Having Daughters Make You More Republican?”, “Parents With Daughters Are More Likely To Be Republicans, Says New Study,” “Parents Of Daughters Lean Republican, Study Shows,” “The Daughter Theory: Does raising girls make parents conservative?” Here’s our question: Why is it all about “the effect of daughters”? Why not, Does having sons make you support the Democrats? This is not just semantics: the male-as-baseline perspective affects the explanations that are given for this pattern: Lots of discussion about how you, as a parent, might change your views of the world if you have a girl. But not so much about how you might change your views if you have a boy.

The fallacy here was that people were reasoning unidirectionally. In this case, the benefit of a feminist perspective is not political but rather just a recognition of multiple perspectives and social biases, recognizing that in this case the boy baby is considered a default. As the saying goes, the greatest trick the default ever pulled was convincing the world it didn’t exist.

To put it another way: it is not that feminism is some sort of superpower that allows one to consider alternatives to the existing defaults, it’s more that these alternatives are obvious and can only not be seen if you don’t allow yourself to look. Feminism is, for this purpose, not a new way of looking at the world; rather, it is a simple removal of blinders. But uncovering blind spots isn’t that simple, and can be quite powerful.

A broader point is that it’s hard to do good social science if you don’t understand the community you’re studying. The lesson from feminism is not just to avoid taking the male perspective for granted but more generally to recognize the perspective of outgroups. An example came up recently with the so-called gaydar study, a much-publicized paper demonstrating the ability of a machine learning algorithm to distinguish gay and straight faces based on photographs from dating sites. This study was hyped beyond any reasonable level. From a statistical perspective, it’s no surprise at all that two groups of people selected from different populations will differ from each other, and if you have samples from two different populations and a large number of cases, then you should be able to train an algorithm to distinguish them at some level of accuracy. The authors of the paper in question went way beyond this, though, claiming that their results “provide strong support for the PHT [prenatal hormone theory], which argues that same-gender sexual orientation stems from the underexposure of male fetuses and overexposure of female fetuses to prenatal androgens responsible for the sexual differentiation of faces, preferences, and behavior.” Also some goofy stuff about the fact that gay men in this particular sample are less likely to have beards. Beyond the purely statistical problems here is a conceptual error, which is to think of “gay faces” as some sort of innate property of gay people, rather than as cues that gays are deliberately sending to each other. The distinctive and noticeable characteristics of the subpopulation are the result of active choices by members of that group, not (as assumed in the paper under discussion) essential attributes derived from “nature” or “nurture.”

Looking from a different direction, feminism can make us suspicious of simplistic gender-essentialist ideas such as expressed in various papers that make use of schoolyard evolutionary biology—the idea that, because of evolution, all people are equivalent to all other people, except that all boys are different from all girls. It’s the attitude I remember from the grade school playground, in which any attribute of a person, whether it be how you walked or how you laughed or even how you held your arms when you were asked to look at your fingernails (really) were gender-typed. It’s gender and race essentialism. And when you combine it with what Tversky and Kahneman called “the law of small numbers” (the naive but common attitude that any underlying pattern should reproduce in any small sample) has led to endless chasing of noise in data analyses. In short, if you believe this sort of essentialism, you can find it just about anywhere you look, hence the value of a feminist perspective which reminds us of the history and fallacies of gender essentialism. Of course there are lots of systematic differences between boys and girls, and between men and women, that are not directly sex-linked. To be a feminist is not to deny these differences; rather, placing these differences within a larger context, and relating them to past and existing power imbalances, is part of what feminism is about.

Many examples of schoolyard evolutionary biology in published science will be familiar to regular readers of this blog: there’s the beauty-and-sex-ratio study, the ovulation-and-clothing study, the fat-arms-and-political attitudes study (a rare example that objectified men instead of women), the ovulation-and-voting study, and various others. Just to be clear: I’m not saying that the claims in those studies have to be wrong. All these claims, to my eyes, look crudely gender-essentialist, but that doesn’t mean they’re false, or that it was a bad idea to study them. No, all those studies had problems in their statistics (in the sense that poor understanding of statistical methods led researchers and observers to wrongly think that those data presented strong evidence in favor of those claims) and in their measurement (in that the collected data were too sparse and noisy to really have a chance of supplying the desired evidence).

A feminist perspective was helpful to me in unraveling these studies for several reasons. To start with, feminism gave me the starting point of skepticism regarding naive gender essentialism—or, I could say, it help keep me from being intimidated by weak theorizing coming from that direction. Second, feminism made me aware of multiple perspectives: if someone can hypothesize that prettier parents are more likely to produce girls, I can imagine an opposite just-so story that makes just as much sense. And, indeed, both stories could be true at different times and different places, which brings me to the third thing I bring from feminism: an awareness of variation. There’s no reason to think that a hypothesized effect will be consistent in magnitude or even sign for different people and under different conditions. Understanding this point is a start toward moving away from naive, one might say “scientistic,” views of what can be learned or proved from simple surveys or social experiments.

I consider myself a feminist but I understand that others have different political views, and I’m not trying to say that being a feminist is necessary for doing science. Of course not. Rather, my point is that I think the political and historical insights of feminism have made me a better statistician and scientist.

As I wrote a few years ago:

At some level, in this post I’m making the unremarkable point that each of us has a political perspective which informs our research in positive and negative ways. The reason that this particular example of the feminist statistician is interesting is that it’s my impression that feminism, like religion, is generally viewed as a generally anti-scientific stance. I think some of this attitude comes from some feminists themselves who are skeptical of science in that is a generally male-dominated institution that is in part used to continue male dominance of society, and it also comes from people who might say that reality has an anti-feminist bias.

We can try to step back and account for our own political and ethical positions when doing science, but at the same time we should be honest about the ways that our positions and experiences shape the questions we ask and produce insights.

Feminism, like religion or other social identifications, can be competitive with science or it can be collaborative. See, for example, the blog of Echidne for a collaborative approach. To the extent that feminism represents a set of tenets are opposed to reality, it could get in the way of scientific thinking, in the same way that religion would get in the way of scientific thinking if, for example, you tried to apply faith healing principles to do medical research. If you’re serious about science, though, I think of feminism (or, I imagine, Christianity, for example) as a framework rather than a theory—that is, as a way of interpreting the world, not as a set of positive statements. This is in the same way that I earlier wrote that racism is a framework, not a theory. Not all frameworks are equal; my point here is just that, if we’re used to thinking of feminism, or religion, as anti-scientific, it can be useful to consider ways in which these perspectives can help one’s scientific work.

All of this is just one perspective. I did get several useful comments and references from Shira Mitchell and Dan Simpson when preparing this post, but the particular stories and emphases are mine. One could imagine a whole series of such posts—“How feminism made me a worse scientist,” “How science has made me a better feminist,” “How Christianity has made me a better scientist,” and so forth—all written by different people. And each one could be valid.

I was impelled to write the above post after reflecting upon all those many pseudo-scientific stories of cavemen (as Rebecca Solnit put it, “the five-million-year-old suburb”) and reflecting on the difficulties we so often have in communicating with one another; see for example here (where psychologist Steven Pinker, who describes himself as a feminist, gives a list of topics that he feels are “taboo” and cannot be openly discussed among educated Americans; an example is the statement, “Do men have an innate tendency to rape?”) and here (where social critic Charles Murray, who I assume would not call himself a feminist, argues that educated Americans are too “nonjudgmental” and not willing enough to condemn others, for example by saying “that it is morally wrong for a woman to bring a baby into the world knowing that it will not have a father”).

When doing social science, we have to accept that different people have sharply different views. Awareness of multiple perspectives is to me a key step, both in understanding social behavior and also in making sense of the social science we read. I do not think that calling oneself a feminist makes someone a better person, nor do I claim that feminism represents some higher state of virtue. All I’m saying here is that feminism, beyond its political context, happens to be a perspective that can help some of us be better scientists.

123 thoughts on “How feminism has made me a better scientist

  1. “feminism can make us suspicious of simplistic gender-essentialist ideas such as expressed in various papers that make use of schoolyard evolutionary biology—the idea that, because of evolution, all people are equivalent to all other people, except that all boys are different from all girls.”

    Here’s a recent blogpost of mine on this topic:

    I would say this perspective is pretty mainstream in ev psych.

  2. Dear Andrew:

    A very interesting post. And also quite timely, given the metoo movement and the sad role some in our profession have played in it. This is why I love your blog. Your posts are always well-written, well-thought and obviously you put alot of time into it. For very few sites on the internet, would I say the same thing.



  3. “Consider the following headlines: “The Effect of Daughters on Partisanship and Social Attitudes Toward Women,” “Does Having Daughters Make You More Republican?”, “Parents With Daughters Are More Likely To Be Republicans, Says New Study,” “Parents Of Daughters Lean Republican, Study Shows,” “The Daughter Theory: Does raising girls make parents conservative?” Here’s our question: Why is it all about “the effect of daughters”? Why not, Does having sons make you support the Democrats?”

    This seems difficult to disentangle from the alternate hypothesis that journalists are writing from a Democrat-as-baseline perspective. Had the study results been inverted, would we be more likely to have seen “Does Having Sons Make You More Republican?” or “Does Having Daughters Make You More Democrat?”?

    • I’d like to tweak Airman’s comment by positing the alternative hypothesis that journalists suspect that readers are more likely to click on an article with “daughters” in the title than “sons”. Rather than assuming that the media is treating men/boys as the default, my guess is that it’s all based on advertising revenue.

      That said, a better click-bait title would probably be “You’ll never guess which political party parents with daughters are more likely to vote for!” Put that up with a picture of Hillary and Trump and you’d be golden.

      • The situation is completely symmetric, it’s not like there’s a third or fourth sex that’s common. If the female sex results in more X compared to male, then the male sex results in more “not X” compared to female…

        • False. What the study presumably showed is that there is no correlation between having a so and political orientation, whereas having a daughter may make make you more likely to vote republican imparted with having no children and make you more likely to vote republican even if you have a son. I haven’t read the paper but I assume that this is the result and there is no gender bias there. Incredibly weak reasoning in this section of the post.

        • David:

          Instead of discussing what the study “presumably showed” and what you “assume is the result,” why don’t you look at the paper, or at least follow the links to where I discussed it. The comparison was having a daughter vs. having a son; it was not a comparison to no children. When comparing childless people to parents, there are obvious confounding factors such as the desire to have children in the first place. All the studies I’ve seen on this topic compare people who have daughters to people who have sons.

        • In defense of some of these articles (I am thinking especially of Glynn and Sen’s 2015 AJPS), at least one the mechanisms proposed is that parents becomes liberal on certain issues after learning from their daughter about her experiences at school, work etc. So it’s a little less awkward to write “daughters make parents more liberal b/c they learn from their daughters” compared to “sons make parents more conservative b/c they fail to learn from their sons what they would from daughters.” Kind of thing.

  4. More pointedly, it’s a lot about being able to establish accurate base rates. Fermizing I think it’s called in the book Superforcasting by Prof. Philip Tetlock. Man it can be tedious process, as I am now discovering. It fits my way of thinking, to a great extent.

  5. > Feminism is about a willingness to question the dominant, tacitly accepted ideology.

    If that is the definition of “feminism” that you want to use, then I guarantee that Charles Murray — and every professor I have ever met — consider themselves to be feminist.

    But, sadly, any time your definition of word X results in you including every person as an X, you need a better definition . . .

    • D:

      When I wrote, “Feminism is about a willingness to X,” I was not meaning to imply that everyone who is willing to do X is a feminist. And I certainly wasn’t meaning to imply that everyone who considers himself as willing to do X is a feminist.

      Just by analogy, one might write, “Conservatism is about a willingness to respect the wisdom of existing institutions,” without implying that everyone who respects the wisdom of existing institutions is a conservative.

  6. I appreciate the most all the qualifications/clarifications you put in the last 3 paragraphs. For myself, thinking about the whole mess of middle east (Israel/Palestine conflict, war on terror, nation building) made me become a moral relativist (and hence a worse person?) and better able to see things from multiple perspectives.

  7. There’s a lot of unintentionally irony in this self-congratulatory post. For all the talk about “questioning the dominant, tacitly accepted ideology”, “moving away from naive…views”, and feminism being a “simple removal of blinders”, the article itself parrots many of the biases of academia’s dominant, tacitly accepted social liberalism: construing “reproductive freedom” as a positive goal, as if allowing women to kill their children constitutes moral “progress” in favor of womanhood, even when abortion is more often targeted at girls and racial minorities; implicitly presenting “equality in the workplace” as a goal not yet achieved due to inequality of outcome despite equality of opportunity; refusing to consider the media’s Democrat-as-baseline view but rather casting blame on a “misogynistic” industry–one which now routinely gets people fired for even speaking against feminism; declaring that you’re opening yourself to be “aware of multiple perspectives” when you don’t discuss a single reason why rational people might disagree with your view or a single way feminism might make you worse as a scientist, merely gesturing that such articles could be written–there is no critical self-reflection apparent beyond promoting the dominant ideology in academia and on the left.

    • Cameron:

      I don’t see where in the above post I’m congratulating myself. Also, I thought I made it clear that feminism is just one perspective; for example, I wrote, “I consider myself a feminist but I understand that others have different political views. . . . I do not think that calling oneself a feminist makes someone a better person, nor do I claim that feminism represents some higher state of virtue.”

      Regarding reproductive freedom: I recognize this is a value that some people have but other people don’t. Another example, in the U.S. context, is the freedom to own guns: some people think this is important, other people don’t. Or various aspects of freedom of the press.

      Regarding the last half of your comment: All I can say is that I was writing from my own personal perspective, and you are welcome to write your own post arguing why feminism has made me a worse scientist. Sometimes a personal perspective is valuable, and you should feel free to share yours.

      In addition, your comment has the problem that you put things in quotes (“progress,” “misogynistic”) that I never wrote, which suggests to me that you’re arguing not with me but with someone else.

      • I don’t think saying feminism is just one perspective is very nice, my civil rights should not be subject to anyone’s opinion for validation. Therefore, being a feminist does make you a better person. Not being a feminist deprives half the population of rights.

        • It’s too general a claim that ‘feminist makes you a better person’. It’s in daily lives [particular contexts] that such a claim has to be evaluated. On practice.

          I think we underestimate how many men also joined in promoting civil rights.

          The politics of feminism are fascinating. I think that Amy Chua’s new book Political Tribes frames eve feminist politics well. Particularly at this time. And as I mentioned before, in the lead up to the 2016 election.

        • In the US feminism goes significantly further than “civil rights” traditionally understood, modern US feminists are not fighting for political equality, – instead, they are mostly fighting for subsidies and affirmative privileges. Just take a look at the history “wag gap” hoax and “diversity in tech” debates. Modern western feminism is a rent-seeking movement.

    • Cameron:

      I want to emphasize this point: I was writing from my own personal perspective. As I wrote in the above post, I could just as well imagine reading an article by someone else about how Christianity has made them a better scientist, and that could make sense too. There are many frameworks which, although not themselves part of science, can affect how we do science, in a positive way. Or, positive in some settings and negative in others. It makes complete sense that I can see the positive aspects, and maybe it takes someone else, such as you, to point out the negative effects.

      • “One of these guys”? You mean, someone who… _disagrees_ with you? If the only way you can think of to respond to a contrary opinion is to point and sneer at it from a different comment chain, Andrew’s blogpost sailed right over your head. I suggest you re-read it, and try to be better.

  8. >>>All I’m saying here is that feminism, beyond its political context, happens to be a perspective that can help some of us be better scientists.<<<

    You could replace "feminism" by a lot of things here?

    Is there *any* perspective, that does not help, at least *some* of us, be better scientists?

    • Anon:

      Could be. What’s interesting is how it works. A post that says “X makes me a better scientist” isn’t so interesting; a post that says “X makes me a better scientist because A, B, C” could be interesting.

        • The fun starts when different commentators construe your impressions instead to be claims of what they want you to be claiming. :)

        • That’s why I prefer to be a fermiser than a feminist. I’m use to challenging guys to a debate because I had spent a good portion of my life debating my father, an all India 1st prize college debater and I understand joined the debating club at Cambridge University. Not sure how he did there.

          Then too, the Internet is a wild wild west of communication. One has to be prepared to wrangle with pretty prickly people.

        • Rahul,

          The posters at Andrew’s blog are implicitly and expressly making claims about feminism, in different contexts. Each has his or her conception of it. In fact these definitions have been contested among feminists. Just review Anonymous and Andrew’s posts.

          For some time there was polarization in the feminist community. Camille Paglia, for example, has been a strong critic of other feminists, particularly during the lead up to the 2016 election.

          An ‘impression’ might be characterized as a ‘weak’ or ‘weaker’ claim. Moreover, I don’t think what is being expressed here are simply ‘impressions’, given the comparison that you have drawn between ‘impressions’ and ‘claims’. Rather I think the opinions about feminism contain ‘assumptions’.

          In short you can excavate a ‘claim’ or an ‘assumption’ in an impression.

  9. I enjoy and am sympathetic with the post overall, but I would say that Rebecca Solnit article you cite is awfully weak. I think it shows how Feminism might have made her a worse scientist.

    (1) Feminism prevents her from reading Morris with an open mind, but only as an enemy. (He would agree that social structure is malleable, an that the past doesn’t justify the present. She would have be wise to listen to his implicit caution that we do not just change our minds with an act of will.)

    (2) Feminism leads her to the fallacy that a single example from anthropology contradicts the existence of a larger general tendency. (And hey, if patriarchy is not a general phenomenon then what is Feminism about, anyway?)

    She’d be better off arguing that 50’s suburban household is myth. Except, Feminism makes so much hay out of that story, I don’t think she could.

    Overall, I’d suggest the Self is the Trap. Andrew’s encounter with Feminism led to reflective self criticism; for Solnit, at least here, self-justification.

    • I think that temperament and learning disposition are lynchpins to doing good science. Not one’s gender. More fundamentally, it’s matter of fluid and crystalized intelligence. I believe that our education & cultures stunt both because we think in binary terms. In dichotomies in the analytical domains.

      • Certainly there is no question anymore about women being able to excel in science.

        However, the Feminist point of view does not brook treating temperment and gender as independent variables. Rather, for the Feminist, one’s temperment is a function of the operation of social structures with effects that differ by gender. A Feminist model and a naive “biological gender diffence” model do not generally make different predictions about data, to the first order.

        • Dzhaughn,

          Thank you. Can you explain what you mean by ‘A Feminist model and a naive “biological gender diffence” model do not generally make different predictions about data, to the first order.’

          An example if feasible, would be great.

        • Dzhaughn said, “… the Feminist point of view does not brook treating temperment and gender as independent variables”

          I’m dubious that there is a single “Feminist” point of view.

          @Andrew: It might be helpful if you gave your definition of “feminism” to clarify what you mean in your title and discussion in this post.

    • I think anthropological evidence can be useful in falsifying some claims about “human nature”, where human nature is assumed to be universal. If the claim is a strong one about something that is supposedly universal, then it seems we only need one piece of (good, independently verifiable) evidence to falsify it. Of course, the claim only needs to be softened slightly to survive the challenge, e.g. you can instead claim that humans have a “strong tendency” to do X. At that point though, the strength of the tendency and the situations where it does or doesn’t apply becomes up for debate, so I think the anthropological evidence has served its purpose.

    • The points [1] and [2] are far more about simply ‘learning disposition’ and ‘temperament’ than about an ‘ism’, like Feminism. They can be best characterized as ‘habits of mind’. I know a substantial number of men and women who are, in different degrees, closed minded, contingent upon the context. For example, I review some Tweets by women and men that show that, despite their appeals to fairness and freedom, just can’t take any criticism. They then clearly show impatience, anxiety, etc.
      Observation in Point 2 has applied to nearly all of us. Even exceptionally smart academics, including lawyers, philosophers, statisticians, physicians. Feminism, Chauvinism, etc are perhaps characterizations that imply that they are more extreme in their views. But it’s really the adversarialness among feminists that have led to polarized stances. I would argue that these habits of minds are a reflection of intelligence levels. John Rawls and I once, in Harvard Square, discussed what constitutes intelligence. And we concurred that it does demand self-reflection. Rawls approach to self-reflection was not based in psychology as such. I had an interesting rapport with Rawls, on whom I practiced my humor. He would laugh actually.
      I think he understood my lampoon side.

      I’m suggesting, therefore, as I did earlier, learning disposition and temperament, genetic in part, are what we are really talking about. Some individuals are simply are predisposed to addressing cognitive dissonance, the mother of many of interactions.

  10. One of the core pillars of modern day feminism is that women get paid $0.77 for every $1 a man gets paid. I hear this often repeated everywhere from the feminists I know personally to the politicians that write and pass bills into law. This $0.77 result is based on a linear regression model with a single factor with two levels (male, female). Based on this one-factor model, grandiose claims of patriarchal oppression are summoned as the explanation for the difference in pay. Feminism makes me a better scientist because their way is an example of what not to do when trying to understand a phenomenon, and instead that one should dig deeper into the data before proclaiming victory in understanding the causes.

    • Now, just because politicians repeat something incessantly does not make that something fundamental.

      Nevertheless, the assertion led economists to notice the correlation of number of years worked and income: most of the “gender gap” in income results from a “gender gap” in the number of years worked. The dependence of income on number of years worked is potentially important in a lot of interventions that have nothing to do with gender.

      And it still does not show that society is not oppressive, although it ought to move one some distance away from a “hiring managers are pigs” story toward something like the “work of raising children is differentially placed on women” story.

      • A topic close to my heart is also how *taxes* are placed differentially on women. The way this works is that married couples can take tax breaks whether the “second” earner works or not. The second earner (one with even slightly lower salary potential) then can choose to go to work and give up 8+ hours a day of their life, or not. If the primary earner is making something like median income or higher (say 60k/yr or more) then the average tax on the second earner’s income is somewhere between about 40 and 50% of each dollar. Whereas the average tax on the first earner is something like 25%. If men have slightly higher salary history even just because they’re a few years older or there’s a small (say 1%) bias towards paying men more… then it’s consistently the women who are in the “second” earning category.

        Combine this taxation with needing to hire a child-care/house maintenance person (nanny) out of mostly post-tax money, and you can’t break even after taxes if the second earner earns less than somewhere close to 100k/yr (say you need to hire 30k/yr in nanny/childcare and increase other expenses by an additional 10k (convenience foods, transportation, car insurance etc), this is ~80k of salary before taxes if your tax rate is close to 50% on the second earner.

        Women without children don’t need to hire that child-care/house maintenance help and so that changes their decision making quite a bit. It can be absolutely financial suicide to take even an upper middle class job under this kind of taxation in the presence of children.

        • For a family income of $100k wouldn’t the marginal tax rate for the addition of $50k be something like 25% to 35% depending on the state? If I understand how taxes work in the US (maybe I don’t), that’s not somewhere between 40% and 50%.

        • First comes something like 8% from social security and Medicare, 15% for self employed. Then 25 or 28% federal, then state something like 6-10%, it winds up between 40 and 50% depending on specifics.

        • Thanks, I was missing the FICA part (not exactly a tax as it also generates entitlements, but it’s true that it may not make a difference).

          Anyway, that’s progressive taxation working as intended. Mo’ money, mo’ taxes and then some.

        • Except it’s not working as intended because the result is less workforce participation and less tax revenue. Originally in the 30s when the whole thing was designed there was very rarely two earners, so more money meant higher wages for the one earner, not twice as much time being sold.

        • Also considering the long-term value of investment in Early Childhood Health and education the appropriate discount rate for a new mother or father age 30 makes any future entitlement payment 30 years out worth pretty much zero in Net Present Value, particularly when including risk assessment

        • It seems you would like the system to pursue conflicting goals: workforce participation and parental investment in early childhood health and education.

        • I don’t want to get too far off the main topic here, but my main concerns are that the system of taxation is

          1) Unfair: it treats the second earner very badly giving this person very little opportunity to make decisions (the decision are made for them).

          2) Inefficient: It leads to highly educated people doing jobs that could more efficiently be done by less educated people, such as college students or immigrants seeking opportunities or the like, thereby harming both the middle-class second earner, and the potential nanny/home care person.

          3) Unnecessary: a universal basic income and a flat tax is in fact *more* “progressive” in that it gives money to everyone, including low income people rather than just reducing their taxes, and it has vastly reduced administrative overhead, and it provides a much better tool for monetary policy.

          Eliminating this problem has potential to vastly increase the GDP of our country while improving the lives of millions and giving more *choice* in the matter of how their children are taken care of to families.

        • If the second earners have very little opportunity to make decisions, what about first earners? Aren’t their options equally limited?

          I agree that a flat tax rate is more fair, but I don’t think it will ever get popular support. A really flat federal income tax would be 20% and at least 90% of the people would end up paying more (several times as much for those in the bottom half). The US has one of the most progressive tax systems in the OCDE.

          (I don’t coment on the UBI idea, it means different things for different people and in my admittedly limited knowlege of the subject none of those things are well defined.)

        • Carlos: flat tax can never work in the US without the UBI, but with the UBI it works, and works well. I define a UBI as basically any fixed payment given to all citizens whose quantity depends only on factors other than income, such as health status, age, etc. In other words your take home income is:

          TH = UBI + (1-r) * W

          where TH is take home, UBI is some fixed dollar amount paid by govt, r is the tax rate, and W is wages/earned income.

          I’ve done a fair amount of analysis on this idea, and it’s a very viable thing numerically, politically I don’t know. I think it’s off the radar of too many people to be politically viable at the moment, and I think much of the ideas surrounding it are misunderstood.

          With the above situation, UBI set to some amount similar to around $500/mo per adult, and r set to the amount it would need to be set to in order to keep the deficit constant (around 30% +-), the only families who would in the end pay more are those making more than about 150k. However with this policy in place far more people could go to work and hire home help, many many families would have opportunity to increase their earnings, and while their taxes would increase, on net their take home would increase. Furthermore literally tens of millions of jobs could come into existence in both high level areas (100k+) and in home care type jobs. Furthermore hiring people would be vastly simplified: pay each person (1-r)*W and send r*W to the govt in their name each month… not to mention eliminating the complexities of a horrific tax code that crushes productivity in other ways.

          But now I think we’re fully off the topic. If you want to talk more about it I’d be happy to set up a post on my blog and we can carry on there.

        • I would be curious to see some detailed numbers (how the to pay for that and where those tens of millions of jobs would be coming from). Maybe high-income households will be willing to increase workforce participation given a lower marginal rate, despite having less need for additional money if they have now a higher post-tax income. But in low-income households both effects would tend to decrease workforce participation: the windfall would be higher (relative to income) and the increase in marginal tax rates would make working less attractive. There would be upwards pressure on wages, I’m not sure that would make much easier for the highly educated people to outsource undesirable tasks to the less educated people.

        • For many low income families receiving benefits the effective marginal tax rate is more than 100%, earning more money means losing eligibility for benefits worth more than the extra money, particularly housing and childcare and food and healthcare subsidies. Eliminating this problem alone could dramatically increase employment.

          I spent a bunch of time trying to do detailed analysis with limited data and a model that was hard to fit in Stan due to the size of the model (spatial time series across the whole country). I wasn’t getting paid to do it. Ultimately I had to back burner it. But the basic results are sound I think, I had graphs that showed two distinct non employed populations, one a group of second earners married to upper middle class spouses, and one a group of parents with children and no income at all, who must be living off subsidies. Both have policy based reasons why they really can’t get out of their situation. Both groups would have dramatically better outcomes with simpler more fair welfare and tax policies (Ubi + flat tax) the number of people involved was not small.

        • Many low income families also get currently benefits above the $12k per adult that you proposed, so if you’re going to remove any other subsidies they won’t be very happy.

          As for the second (non)earners married to upper middle class spouses making $120k, I don’t think they will rush into the workforce because the marginal tax rate is reduced from 37% to 30% but of course I’ve not run any model so I can’t tell for sure.

          Finally, it’s not so clear that “second earners married to upper middle class spouses” are penalized. One could also say that the one who is penalized is the one who has to work. And if the “second earner” works and the taxes paid by the household go up you cannot say that the taxes paid from the “old” earner remain constant and the increment is paid entirely by the “new” earner. You could also make the calculation in the other sense, calculating first the taxes for the “low” earner income alone. It would be more reasonable to apply to each of them the effective rate for the couple. Or of course they could file separately and in that case it would be clear that if both work the “high” earner pays more taxes (in dollars and in percentage) than the “low” earner.

        • 2018 is a new tax law, but in the old law the 25% tax bracket hit at around $76k which is like around 60th percentile it’s not just people with 120k incomes affected. And after $135k or so FICA no longer applies marginally, so you can see in the data where two surgeons who each have earnings potential in the $250k range are in fact both working… The problem is pretty big for society in terms of lost productivity and reduced economic welfare. It’s not a problem for the very rich, but any family with less than 90th percentile income would likely wind up better off with achievable changes.

    • The single factor regression model does tell you something important: that there is an overall pay gap. More factors might help pin down why it occurs, but does not change anything about woman being paid less.

      All the others factors you might include in the model (education, type of work, years of experience) are all influenced by your gender, so you can hardly argue that it somehow proves the pay gap is not real or not due to discrimination.

      I think type of work is especially interesting – it leads one to consider why “typical woman jobs” are paid less well than “typical male jobs”. In fact, there are cases where once a profession was opened to women (examples include secretaries and school teachers) both wages relative to other jobs and respect for the profession dropped. I know some freshwater economists would say that wages are determined entirely by the free market, but I believe that the perception of society also matters. And there are still government jobs as well, where it applies as well.

      And years of experiencing are directly tied to the expectation that the mother and not the father should take primary care of the children.

  11. If you really want to “question the dominant, tacitly accepted ideology”, I suggest you write an essay on how feminism has _not_ made you a better scientist. We’ll look for the remains of your career in the resulting blast crater…

      • So you realize it’s an ideology and STILL not even allow but welcome its effects on your scientific research. That, good sir, is very troubling and tells a lot about your idea of science.

      • My point is that in 2018, the idea that feminism makes you a better scientist is a completely orthodox belief. You had _better_ believe your proposition, or your career will suffer severely for your apostasy.

        Feminism _is_ the dominant ideology in academia, and I believe that science is suffering because of our unwillingness to question that dominant, tacitly accepted ideology.

  12. Labels are funny things. At best they are a form of shorthand that make communication via spoken word or text less tedious; perhaps even possible. At worst, especially when self-applied, they are terrible classifiers as likely to land one in the wrong bin as grant you entry into the tribe whose fellowship you seek. Adorning oneself with labels is also a form of surrender as it allow others to define you, rather than you taking the time to define yourself. Ideas on the other hand are sturdier things. Liberty for all and equality before the law require only that you be a member of humanity and in reasonably good standing. They are aspirational it’s true; just like science. And so they often fail of their promise, in the short run, to deliver what we demand; and yet in the long run they have delivered far more than all the labels people have peeled off and stuck on their foreheads in hopes of changing the world, or at least themselves. Exhibit “A”, from this evening’s NYTimes:

    I hope this doesn’t come across as harsh. In the past I sponsored (successfully) at prior firms the first first gay partner, the first black female partner and the first female partner who wanted to work part time in order to spend time with her three children. I reached those decisions based not on any labels that they (or I) wore but rather by judging them irrespective of labels.

    • Thanatos:

      Point taken, that actions are more important than words. I don’t delude myself that the above post is going to change the world or to “deliver what we demand.” I’m just sharing my impressions. More generally, when blogging sometimes I offer directly actionable advice (for example, when I posted a few years back on zero-avoiding priors) and other times I share thoughts that I think are interesting, even if there’s nothing much you can do with them (for example, my reviews of Nassim Taleb’s books). I think the latter sorts of posts can be indirectly useful in improving our understanding of the scientific process. The point of the above post was not to “seek entry into a tribe” or to stick a label on anybody’s forehead but to share some introspections about science.

    • Thanatos: With regard to the link, when I was a student I had two of my female tutors (in different years) try to be romantic with me (no where near as serious as in the link) but it made me feel very uncomfortable and I find it weird that others claim to not understand this. Also, I feel that I have to add that they were very physically attractive to prevent a misunderstanding of the reasons for discomfort.

      Andrew: I do wonder if there is some confounding at work here – some academics who are very aware and critical of implicit reasoning who label themselves as feminist researchers rather than the feminist mindset/worldview itself. I feel that among “labelled people” most are not very aware and critical of implicit reasoning but rather mostly caught up in the group think of the “label”. Thanatos’ link does suggest that.

    • Thanatos

      Great comments. I think we accept eclectic thinkers more than we let on. Academia, for example, seeks out people who think differently. It’s that their views are overgeneralized and may be applied to contexts that are inappropriate.

      I especially like your point about ‘letting others define you’. This is what I, gleefully refer to as ’emotional blackmail’ culture. I see this tendency in a good number of individuals. And sometimes it’s worth challenging and sometimes just letting it go.

  13. Thanks for the extremely clear description. Most of what you said resonated with my personal experience as a scientist and feminist.

    Regarding the framing asymmetry: Peter Hegarty ( is a very interesting and compelling social psychologist, who has gone some way towards documenting and quantifying this effect within academic psychology and scientific thinking, e.g., here, and here, and many others.

    In some cases, Hegarty’s data could be stronger; others are quite compellingly large effects. Just thought these might be interesting to some readers!

  14. I am deeply confused about how waving the flag of the dominant ideology (dominant in media and politics) has any positive effects on critical thinking.

    You mention data. Well, central feminist concepts such as the gender pay gap, rape culture or even patriarchal structures in western countries have still to be proven using scientific means by, well, feminist scientists like yourself.

    If feminism has something going for it it’s exactly NOT its scientificness. It appeals to our hearts, not minds. That’s why even self declared feminist outlets like The Guardian and Der Standard have had to admit that things like the gender pay gap are NOT based on data in the way they and other feminists have claimed for decades.

    The main point of this article is like saying: my belief in God makes me a better scientist. The opposite is true.

    • Skythe:

      It’s ok that you’re deeply confused about all this. The general topic—the ways in which non-scientific ideas affect our scientific practice—is confusing. That’s one reason I blogged about this: I think it’s good to be open about our confusion, also we should be open about variation. For example, you say that your belief in God does not make you a better scientist. That’s fine. But it would be a mistake to exclude the possibility that, for various other people, belief in God does make them better scientists. There are lots of reasons why this could be true.

  15. I really like the idea that feminism can make a person a better scientist by offering additional ways of looking at the world. Feminism does not tell you what to think or say; it just offers some perspectives. No “-ism” worth its salt takes the place of critical thought.

    In contrast, I remember teacher training sessions where, at the end, we were asked to fill out a T chart whose heading on the left read, “I Used to Think…” and the one on the right, “Now I Know…”–as though the whole point had been to *replace* our existing ideas (and, moreover, to supplant uncertainty with certainty). Unfortunately that’s an all-too-familiar trope.

    Granted, it’s possible, in a training session, to learn that you were wrong, or at least uninformed, about something. Even so, rather than jump to the “new” conclusions right away, I would rather mull over the literature on my own, ask some questions, etc.

    • >>>I really like the idea that feminism can make a person a better scientist by offering additional ways of looking at the world.<<<

      Would we say this is true for the other "-ism"'s too? e.g. Take racism for example.

      Almost any such labelled ideology must offer some new ways of looking at the world?

      e.g. Could we say that racism could make someone a good scientist?

      • Some “-isms” open up perspectives, some close them off, and some do both. If we define racism, briefly, as the belief, or actions based on the belief, that some races are superior to others (intellectually, morally, or otherwise), then it counts as an “-ism” that closes off perspectives. Communism does both; it offers a (sometimes) compelling critique of capitalism but insists dogmatically on a particular solution. Sometimes it depends on the variety of the particular “-ism.” Some kinds of feminism do close off or limit perspectives. But I think Andrew is talking about a different kind, and so am I.

        • I like to think of science & ideology as orthogonal axes.

          Science is what helps me evaluate the “-ism”‘s like feminism, racism, communism etc. for their content (good & bad), impact etc.

          But the moment you say “X-ism helps me do better science” you lose that neutrality of vantage point.

          I prefer to keep apart the tools inherent to practicing good science (logic, reasoning, math, empiricism etc.) from the ideologies i.e. “-ism’s” which rightly ought to be the subject matter what science studies and evaluates.

          I’d mistrust the scientific output of a social-scientist who’s a self proclaimed racist. And so also for the other “-ism’s”

        • I’d mistrust the output of any social scientist who can’t identify and take into account their own ideological perspectives.

        • Rahul: To give a variant of Box’s oft-quoted maxim: “All models are wrong; some are sometimes useful and sometimes misleading.”

          I think your “science & ideology as orthogonal axes” models is sometimes useful and sometime misleading. For example, some people’s concepts of “science” are very much like ideology; and sometimes ideology contains perspectives that can contribute to doing good science.

        • Martha:

          I often see examples where “some people’s concepts of “science” are very much like ideology”.

          But I struggle to find convincing cases where “ideology contains perspectives that can contribute to doing good science.” Any examples you can offer?

        • I’m not Martha but here is a link where a pretty pro- and eminent academic talks about a few situations in which a specific ideology made specific perspectives more noticeable to him:

          Other than that: There is in many definitions of ideology no such thing as a human doing ideology-neutral things. The network and framework of ideas, assumptions and convictions through which someone views the world are there, whether they acknowledge them conciously or not. This gets easier to grasp on the more applied end of science where you might decide to research the copper uptake mechanisms of rice (or whatever, I’m far away from applications and also from plant sciences) because you heard of rice farmers in some places having trouble with water that contains too much copper and you kinda care about those farmers and think it sounds intriguing so you research that instead of, I dunno, corn reproduction. In this fictional case the mere choice of research topic is inspired by your ideological views, including e.g. that those rice farmers’ issues matter. Even if you don’t do it to help them, your knowing about this phenomenon is informed by your view of the world and what in it matters.

        • Rahul:

          First, I am going to agree with anon that you can’t start inquiry from anywhere but where you find yourself, but I don’t think that is your point anyway.

          Ideology might be too vague – here values and ethics are separated from logic of inquiry (scientific reasoning) It’s one perspective.

          On the other hand, if you mean creed – I agree with you – those are always obstructive.

          Now I am wondering if what Thanatos has quote is more of a creed than an meta-epistomology…

          Though I strongly disagree with part of what’s quoted – “Ways of knowing informed by the motive of caring for everyone’s needs will produce more valuable representations than ways of knowing informed by the interests of the dominant.”

          As argued in my post I linked to above, “Ways of knowing motivated be desire to better grasp reality as being reasonable (making sense of the world in ways that we hope are connected to that reality we have no direct access to) will produce more valuable representations than ways of knowing informed by any other interests.”

        • Rahul said: “But I struggle to find convincing cases where “ideology contains perspectives that can contribute to doing good science.” Any examples you can offer?”

          Possibly we are using “ideology” in different ways. I looked it up on the web just now, and found a variety of definitions.

          However, looking back at your comment that I replied to, I see that you wrote, “I prefer to keep apart the tools inherent to practicing good science (logic, reasoning, math, empiricism etc.) from the ideologies i.e. “-ism’s” which rightly ought to be the subject matter what science studies and evaluates.”

          If I am reading this correctly, you seem to be defining “ideologies” as “isms” — and you also include “empiricism” as one of the “tools inherent to practicing good science.” So “empiricism” would be an example of an “ideology that contains perspectives that can contribute to doing good science.”

        • Unfortunately, Science and Math are not some secret souse you can add anywhere to make it better.

          The results of scientific inquiry depends on how you pose a question (as was shown many time on this blog), and how you pose a question may depend on your Ideology.

        • Almost all -ism’s open up some perspectives and close off others. Racism opens up the perspective that there might be important differences between races. Arguably anti-racism closes off this perspective. We could question the utility of research motivated by racism and should be very wary about its veracity (I would be wary of any research motivated by any -ism), but is it impossible that valid scientific discovery can be made even under such a framework?

    • >Feminism does not tell you what to think or say; it just offers some perspectives.

      You might want to let the feminists in on that, because they don’t seem to have gotten the memo. Pretty much every time I’ve encountered someone self-identifying as a feminist lately they were barking orders.

  16. Thanks for the accessible rehash of feminist standpoint theory / parts of feminist science studies! We need more of that in this community. It would be great if you added cites to that, or at least suggested readers google the people who have done so much work on that if they want to know more.

    Also, I’m a bit confused how you reconcile being a feminist with not thinking others should also be feminists? I know your definition of it is different, but just about any other one I have seen from feminists involves some aspect of normative claims making (e.g. that sexism and such are bad). Feminists challenge dominant views because they do harm to people. Is there any normative aspect to your feminism? If so, do you then believe that others ought to share those values? In the extreme, can a feminist be okay with others opposing things like women’s suffrage?

    • Quant:

      1. I am not up on this literature myself so please feel free to add these links in a comment.

      2. There is a normative aspect to my feminism and I would prefer others to agree with me, particularly on political issues that affect others. But I recognize that people disagree.

      • Warning: extreme grumpiness ahead (apologies in advance)

        Unless you have wholly renounced quantitative methods you would do well to read: and links.

        Thus is highlighted the problem of adopting labels instead of ideas: “The masculine cognitive style is abstract, theoretical, disembodied, emotionally detached, analytical, deductive, quantitative, atomistic, and oriented toward values of control or domination. The feminine cognitive style is concrete, practical, embodied, emotionally engaged, synthetic, intuitive, qualitative, relational, and oriented toward values of care. These cognitive styles are reinforced through the distinctive types of labor assigned to men and women—men having a near monopoly on the theoretical sciences, warmaking, and on positions of political and economic power calling for detachment and control; and women being assigned to hands-on emotional care for others. The feminine cognitive style is said to be epistemically superior because it overcomes the dichotomy between the subject and object of knowing and because an ethic of care is superior to an ethic of domination. Ways of knowing informed by the motive of caring for everyone’s needs will produce more valuable representations than ways of knowing informed by the interests of the dominant.”

        Really? Is that what you now believe? Not me. I believe women CAN and DO (whenever trained as boys are trained) reason abstractly. If that makes me a non-feminist, well, so be it.

        P.S. Yes, I know some feminist epistemologists have come around to the view that quantitative methods might be legit in some cases, but if you have the stomach to go down the rabbit hole and read what most of them write … well caterpillars, pass the hookah pipe to your new friend as he passes by. He’s gonna need it.

        *The truth is discovered, not created; otherwise the scientific method is meaningless* – T.S.

        • Thanatos,

          I lean to your views. First of all, each of the characteristics or traits listed are not less starkly exhibited in men and women, based on my own experiences. That is to say, women have been put into positions in which they are having to make the same decisions as men have traditionally. I spent nearly 4 years with national security communities as well as those who were forging 21st century force planning structure, a project that was under discussion in both the Clinton and Bush administrations. So that requires being trained in fields in which men have dominated.

          Moreover, I find many men who have cognitive styles that are generally associated with women. Granted physiologically men and women are different. For example, many younger educated fathers with their children, display very nurturing traits. I gather that they may have spent more time with a nurturing figures, like their Moms perhaps.

          I’ve come across women that are controlling too Nor would one associate masculine traits with them. I think we draw characterizations that are stark.

          My aunt, a combinatorial mathematician could be held to having the cognitive styles associated with men. My uncle, a physicist did most of the housework, including cooking. My uncle’s father took care of 5 children, as his wife died when my uncle was only 5 or so.

        • At its core, I think Feminism is about caring about the societal outcomes of women and how cultural artifacts and power structures affect those outcomes, and intentionally trying to avoid those historical biases to the extent that they harm women or men or both.

          If that is a good definition I am all for it.

          But I think plenty of people label themselves as Feminist and then speak for “true” feminism or whatever out of their own personal beliefs that place a blanket over all women in an inappropriate way. I think that’s what’s going on in your quoted text. That “women’s ways of learning” stuff has been commented on several times by Martha Smith here at the blog. She’s been pretty vocal about how those labels didn’t fit her own thinking as a member of the group of academic mathematicians, and she’s a woman. So I think anything that tries to generalize “how women are” and fails to consider variation and the opinions of actual women is more or less just political power-play and can be usually ignored.

          My impression is that the stuff you’re quoting was a popular academic topic among feminist theorists in the 1980’s and has largely fallen out of favor, but it’s not like I follow this topic closely.

        • Daniel,

          I was about to make the same point. Some of the discussion among feminist theorists was popular in the 80’s and has been put on the back burner I’d say. I think the immediate concerns are economic in nature. With the increasing reliance on technology, women and men are having to retrain to learn relevant tools and skills.

        • But that’s a bit of the No-true-Scotsman thing, isn’t it?

          If you surveyed a bunch of people what’s feminism at its core, I feel we would get very divergent answers. So it’s a bit frustrating discussing “feminism” because everyone’s conception of it varies wildly.

        • It may be. My impression was that this core concern was pretty widely accepted, but that lots of additional things are bolted on by varying people. I doubt very many people who label themselves a feminist would disagree with the idea “Feminism is about caring about the societal outcomes of women and how cultural artifacts and power structures affect those outcomes, and intentionally trying to avoid those historical biases to the extent that they harm women or men or both.” at least in some mostly equivalent rewording perhaps. But each person might add in “and also…” with a lot of additional baggage.

          I agree that the wide variety and vehemence of belief for the “and also” stuff can be frustrating. because of that I tend to not use the label “feminist” even though I do agree with what I perceive as the core idea described above.

          For example what Camille Paglia thinks feminism is obviously is pretty different from what say Gloria Steinem thinks feminism is. Participating in that is not particularly of interest to me, but I don’t think either one of them would say that they don’t care about the societal outcomes of women, or that we can freely ignore historical biases that harm women (or men). I think they have a lot to say about what the appropriate interpretations and actions are though

  17. Citing the list of article titles that were all centered on the daughter as if they were all independently derived feels a bit misleading. The first one, “The Effect of Daughters on Partisanship and Social Attitudes Toward Women”, is the title of the paper. The rest of them are articles summarizing the study; no surprise that they copy the source’s framing.

    Why did the study’s itself use this framing? That appears to not be as much because of a male-as-baseline assumption as because they’re responding to pre-existing research in the area. Washington (2005) takes a similar daughter-focused title about politician’s voting records, but it itself is a response to Warner 1991 “Does the Sex of Your Children Matter? Support for Feminism among Women and Men in the United States and Canada”. That study looks specifically at feminist attitudes; it makes sense to frame that as “having a girl makes parents more concerned with women’s issues”. It’s reasonable to think that having a female child would affect one’s perspective on women’s issues more than having a son would, though of course a comparison of parents of daughters against parents of sons would not distinguish between the two. The most recent survey does a similar thing; they test the impact of children’s sex not just on partisan identification, but also on traditional views of women, abortion, and teen sex. They didn’t end up finding strong evidence that the latter three changed, so they just chose the one that was significant and made that the headline (hooray!). Regardless, their study was also explicitly studying impact on women’s issues, so it makes sense to think of daughters as the more directly relevant condition.

    Paper in question:
    Washington 2005 pdf link:
    Warner 1991, can’t find a pdf link, but jstor:

    (Side note, looking at these studies makes me agree with your initial skepticism at the results. They point in different directions, and are filled with multiple hypothesis testing. The most recent study tested four results, found no significant change on three of them and an effect in the opposite direction from expected on the other one, so they reported that as their headline without, as far as I can tell, doing any correction for multiple comparisons or the lack of agreement with their initial hypothesis. The 1991 study found “Using data from the Detroit and Toronto metropolitan areas, the analysis shows that for women in both settings and for Canadian men, having female children is associated with more egalitarian views. The finding that American men’s views are unaffected by the sex of children is interpreted as being related to different cultural and political climates in the United States and Canada.” That also seems like a serious just-so story and a sign for concern about how they’re splitting up their data to find statistically significant subpopulations.)

    This also just feels like the kind of thing that could be spun either way. The authors highlight the impact of daughters in the title, and you go “Why the assumption that sons are the baseline?” If the headlines had instead been “Does Having Sons Make You More Democrat?”, I can easily imagine critics going “Why are males framed as active and changing the world, while daughters are passive and backgrounded?”.

    Here’s the flipside of the advantage you cite: yes, feminism makes you more resistant to hypotheses from other schools like evo-bio, but this isn’t purely a benefit. There are many other cases where an ideology will cause someone to make up a just-so story to support it. Pushing yourself into a different area of ideology-space makes you more resistant to bad ideas from competing ideologies, yes, but it also makes you more resistant to good ideas from competing ideologies and more susceptible to bad theories from your own ideology. This just collapses down to an on-the-merits debate about whether evo-biology or feminist skepticism is right about a list of hypotheses; you basically define evo-biology as the extreme position on one side, then say that feminism is not the opposite extreme, but the entire rest of the space. So defined, feminism will almost definitely bring one closer to the truth, but one could just as easily say the opposite: that feminism is the dismissal of any sex- or gender- based differences, that evolutionary biology is not about saying that sex defines everything but that there are important differences between the sexes that research will bear out. By that definition, evolutionary biology is almost definitely correct. But all we’ve done is acquire territory through re-definition, without any reference to how these ideologies play out in the real world, or whether our definitions actually agree with other people’s.

  18. Thanks! Some classics (and good reads) include:

    Sandra Harding’s work on “Strong Objectivity”,
    Donna Haraway’s work on “Situated Knowledges”,
    Patricia Hill Collins’ work on the “Outsider Within”.

    Each makes a strong epistemological case for what you’re arguing in this post in different ways. There’s a whole literature on feminist epistemology within feminist science studies, full of great work and thorough debates, much of it written by people who were trained as scientists. I really think readers would benefit from even a passing mention of this in the main post (“If you want to learn more, look into these keywords…”). It gets much more readership than comments.

    Plus, acknowledging the feminists who have made these arguments before is a matter of justice. You’re doing great work to promote the ideas, but you don’t need credit for inventing them. As you say so well in your papr about the “gayface” study, reading and citing the prior work on a substantive topic makes for much better research!

    • “Strong claims”? More like “a lot of vague whiny verbiage and very little empirical evidence”

      What you are describing is re-packaged Marxism with “being a proletarian makes you seethings better, even though when I try to explain why I am right I cant do it but it is not because I am stupid or wrong, it’s because explanation is a white man/ capitalist game and and I am not playing it so I will just be whining a lot and loudly, and when the time comes I will murder you”

      • Curio:

        “Makes you seethings better . . .” I like that! Sounds like the kinda thing Gollum would say.

        I haven’t read anything by Harding, Haraway, or Collins, but now I’m curious about where in their books they get around to saying that they will murder people.

  19. If you create strawmen this stupid, you are an idiot yourself:

    “Looking from a different direction, feminism can make us suspicious of simplistic gender-essentialist ideas such as expressed in various papers that make use of schoolyard evolutionary biology—the idea that, because of evolution, all people are equivalent to all other people, except that all boys are different from all girls.”

    Who fucking thinks ALL boys are different from ALL girls? Does he realize that there can be group differences without every member of the group displaying those differences? Or does he know this and chose to use this stupid formulation anyway?

    • Jim:

      Let me clarify.

      1. In the actual schoolyard, you will see pure deterministic gender essentialism, of the form that all boys do X and all girls do Y, and if you’re a boy who does Y, or a girl who does X, there’s something wrong with you and you’re not a real boy or girl.

      2. Some social researchers really do seem to think in terms of universals, and even when recognizing exceptions seem to think that each exception can be explained away; see here, for example.

      3. More generally, lots of researchers have published papers expressing an attitude that gender differences are so large and persistent that they will show up even in small samples of noisy data. An example would be that claim that beautiful parents are 8 percentage points more likely to have girls. Yes, this is an average difference, there’s no “all” here—but any realistic difference would have to be on the order of 0.1%. The researchers who draw strong conclusions from this sort of data pattern are following the fallacy that Tversky and Kahneman called “the law of small numbers”: the supposition that average patterns should show up in small, noisy samples. Similarly with that study that claimed that women were 20 percentage points more likely to support Barack Obama during certain times of the month. Again, the mistake is not for these people to suppose that there are some average differences between whatever groups are being compared, but for them to think that these differences are large enough that they will show up in these noisy samples.

  20. Thank you for your interesting perspective. It’s also interesting to see many male names as commenters and their nuanced opinions.

    However, as a feminist statistician myself, I think you missed a few big important pieces that really predicate today’s practice of statistics and causal inference, especially in my field of public health. You talk about practical, modern considerations, but seem to miss this historical issue of statistics really being a philosophy as well as a practice. I will build the case here and I think you will see what I am talking about.

    1. Des Cartes, “I think therefore I am” could be argued as the start of the concept of causal inference. This was at a time when women were essentially not allowed to be philosophers and weigh in on the philosophical discussion.
    2. The scientific method – which we all now use for causal inference in statistics in medicine – was developed in Greece at a time of extreme misogyny in the culture. While they were also opining about “democracy”, a study of history will show women did not participate in this so-called democracy. We don’t even have much left of women statues or art to even know what women wore at the time. They were a slave class then.
    3. SAS is our big statistical software many of us use today. It was originally based on punch cards. These were developed as part of the Jacquard loom. When this technology was developed, Jacquard himself ran the loom, and women just threw the weft. Again, women were in the slave class.
    4. When SAS formed as a company, there was one founding woman, but she was immediately bought out, and the remaining founders got rich by essentially linking themselves directly to the government. Today, this patriachal infrastructure still dominates. Women are not at the top of SAS. There are no procs named after women statisticians. Modern history of SAS and statistics shows women doing punchcards (“keypunch”), and men doing the analysis. I challenge you to find any historical pictures of a man making punchcards.

    Please reflect on this result I will state here. Through all of this history I have reviewed, predominantly women have been the data collectors and recorders, and predominantly men have been the data analyzers. Setting aside for a moment how this unequally distributes power, it creates a practical problem in that those who are familiar with the measurement – how it was done, an estimate of its error – are completely segregated from those who are analyzing the measurements. The 80% of healthcare workers in the US who are women are paid very low; the 20% of men tend to be in the higher echelons. Yet almost all of our production healthcare data in the US come from data entry from women, and almost all of the analysis is done by men who simply receive the data, but have not had any hand in its collection and recording.

    I feel this misogynistic stratification in statistics has led to some profound mistakes in today’s practice of statistics. For example, if you read how Hosmer & Lemeshow recommend how to do survival analysis in their classic book, you will see that they inherently trust EACH DATA POINT to have equal, random, and sufficiently low measurement error. As a person who has done more data collection and entry in her life then most men (and women) – especially statisticians – I can attest to you that this assumption is a huge mistake. It leads to the overfitting of models (especially the inappropriate addition of interaction terms into regression models), the misuse of data due to misuderstandings about their meaning, and ultimately, the misinterpretation of the answer.

    This subjugation of women as the data collectors – the measurers – who simply serve a class of statistician men has not really done us any favors in terms of seeking the right answer to our question about causal inference. It’s part of why we often get the wrong answer over and over. Example: Observational studies showed that hormone replacement therapy was cardioprotective, but Women’s Health Initiative – a huge clinical trial on the topic – found that it actually causes heart attacks. I think all that misinterpretation might have been avoided if we did not have this stratification of statistical practice by gender. I’m not saying there are no women statisticians – I’m one, so obviously that’s not true. I’m just saying most of them did not enter a lot of data and avoided “secretarial” work so they could deliberately enter the “male” stratum of statistics. I have more than one female colleague who insists she won’t enter data or do anything secretarial lest she be denigrated as a non-analyst. I, a research secretary and fashion designer who evolved into a epidemiologist and data scientist, am an outlier. And from my perspective, lot of knowledge is being lost as a result of this feature of the patriarchy that expresses itself in modern statistical practice.

    Although everything you said is important, I think this historical misogyny in the field of statistical practice actually negatively impacts our practice of statistics today, and this is a big problem if we want to get the right answer when we analyze our data. Even a misogynist should be able to see this practical problem, I think.

    You start your post by saying the feminism is not a branch of science. I agree, it is a branch of philosophy, as is statistics. I’m surprised we are not seeing more discussion of how these philosophies align as well as compete. Thank you for your thought-provoking perspective.

    • Monika:

      Interesting points. I’ve thought a lot about the importance of data collection and measurement and how they are neglected within statistics, but I hadn’t made this connection to sex roles, the idea that measurement and data collection are considered to be a lesser topic of research.

    • I do believe that very widely things like data entry are tasks for the minuends to do as its tedious and unimportant. It is tedious but it is almost always critical. Hence, something _women_ can be given to do. I have always tried to have some involvement in data collection, entry and verification.

      However, I side more with weak rather than strong linguistic relativity

      Thoughtful and well meaning people can get over it with reasonable effort.

      Should look up CS Peirce scolding Bertrand Russel for dismissing Vicoria Welby’s contribution to philosophy on language in the early 1900s…

      • Keith said,
        “I do believe that very widely things like data entry are tasks for the minuends to do as its tedious and unimportant.”

        I think you meant “minions”, not “minuends”? (Perhaps an autocorrect corruption?)

    • Although I am in agreement in with you that all of Mathematical/Statistical reasoning is engendered by Philosophy (and “is” Philosophy), I don’t quite understand why you say that causal inference started with Descartes.

      I think the scientists and philosophers before Descartes also used such modes of thinking.

      Descartes was the Philosopher who showed through logic that he could only be certain of his existence. In other words, I can only certainly say that I exist, but not you. I can doubt all knowledge, all of existence, but in some form I exist. From this foundation, he reconstructed the real world. This could be thought of as an “axiom” of all Philosophy. It has to be true, at the least for me, as I write this on the blog.

      Nevertheless, engaging in causal reasoning is not to say that the person started this mode of thinking. Perhaps, I am mistaken, and you may elucidate further.

      As for feminism, it is an “umbrella” term, meaning many things to many different persons. Fundamentally, each issue is separate and can be compartmentalized so that it can be judged based on its merit and demerits. Personally, I am not in agreement in terms of “clustered” thinking, in which many social issues are glued together under an umbrella term, and people are either for or against this “label”.

  21. “measurement and data collection are considered to be a lesser topic of research”

    A large part of the admiration of both Charles Darwin and Marie Curie stem from their seminal theories being the result of endless hours of painstaking measurement work. I admire my own scientific hero, Richard Goldschmidt, not only because of his bold theories, but also because those theories were grounded in decades of lab research. With very few exceptions, landmark scientific theories began as simple – but novel – 1 + 1 combinations of insights gained from lab work.

    But I can see where this might be more of an issue in statistics. If you believe that an algorithm can only be meaningfully tested against synthetic data, what value do we gain from measurement and what worth do we ascribe to those who do it?

  22. Dear Andrew,

    I will love you to invite you to bring your point of view to our conference. Registration is open and free.
    Looking forward to see you there!





Leave a Reply

Your email address will not be published. Required fields are marked *