Skip to content
 

Association for Psychological Science claims that they can “add our voice and expertise to bring about positive change and to stand against injustice and racism in all forms” . . . but I’m skeptical.

David Leonhardt writes:

Close to 10 percent of black men in their 30s are behind bars on any given day, according to the Sentencing Project. . . .

When the government last counted how many black men had ever spent time in state or federal prison — in 2001 — the share was 17 percent. Today, it’s likely closer to 20 percent (and this number doesn’t include people who’ve spent time in jail without being sentenced to prison). The comparable number for white men is about 3 percent.

Even 3% seems like too much. But 20% . . . I can’t even.

In the midst of all this, the Association for Psychological Science issued a statement:

The United States is once again confronting its history of racial discrimination and inequity. . . .

From its inception, the field of psychological science has studied the causes and harmful impacts of stereotypes, prejudice, and disparity. . . . our profession now has a sizable body of research that can be brought to bear to understand the persistent racism and subsequent confrontations we are witnessing in cities and communities across our nation and throughout the world. . . .

We stand ready to add our voice and expertise to bring about positive change and to stand against injustice and racism in all forms. . . .

As a diverse, international organization of more than 30,000 scientists and researchers around the world working to understand all aspects of mind and behavior, APS profoundly respects the essential worth of all people and cultures, regardless of race, gender, and ethnicity.

They also feature this page, which links to 18 past APS articles about racism, on topics ranging from health risks to the implicit association test.

That’s all fine—I mean, no, it’s not all fine, there are lots of problems with the implicit association test (see here and here) which I don’t see mentioned on that APS page—but it is what it is.

But that’s not the only way that the field of psychological science has studied the causes and harmful impacts of racial disparity. There’s also to this recent paper, which features some bold claims:

The prescriptive values of highly educated groups (such as secularism, but also libertarianism, criminal justice reform, and unrestricted sociosexuality, among others) may work for groups that are highly cognitively sophisticated and self-controlled, but they may be injurious to groups with lower self-control and cognitive ability.

The above-cited paper has a politically right-wing take on racial politics, writing that “religion would have greater utility for regulating violent behavior among societies with relatively lower average IQs than among societies with relatively more cognitively gifted citizens,” which is academic-speak for “People with African descent need religion to stop them from criming.” I’m not at all convinced by their argument, but it did appear in the flagship journal of the Association for Psychological Science, and one of its authors is Roy Baumeister, a William James Fellow, which is pretty much the highest award given by the Association for Psychological Science (oblig kvell).
Indeed, if you want to understand “persistent racism,” what better place to start than with the psychology literature, which has a rich history of racial essentialism, past and present? Indeed, so much of academic psychology is left-wing, that it’s only fair that they occasionally publish something from the right.

Just to be clear, nothing special about psychology. I’m a statistician, and racism is central to the history of statistics. Galton, anyone? Racism is one framework that people have to understand the social world, and a popular framework it is.

No easy answers

There are no easy answers here. As noted above, the APS is a diverse organization of more than 30,000 people, and these people have diverse views. Some of them are all into the implicit association test, others are into the idea that secularism, criminal justice reform, and “unrestricted sociosexuality” don’t work for countries full of black people.

If the Association for Psychological Science wants to make a statement about racism, I think they should acknowledge the diversity of views on racial politics as evidenced in the papers they publish.

When they say, “We stand ready to add our voice and expertise . . .”, I don’t know what they’re talking about. What is their “voice”? Is Roy “The large institutions have almost all been created by men. The notion that women were deliberately oppressed by being excluded from these institutions requires an artful, selective, and motivated way of looking at them” Baumeister part of this voice? Do his statistical methods count as part of their “expertise”?

Look. My point is not to “cancel” Baumeister and his colleagues, or drop their work down the memory hole, or whatever. It is what it is. It was published by Psychological Science. Baumeister won the APS’s top award. I’m not a member of the APS, nor am I even a psychologist. Who am I do judge what they publish in their publications (except when they print flat-out lies; that does bother me). My point is that Baumeister’s right-wing perspective on sex roles and his right-wing perspective on race relations are part of otherwise-mostly-left-wing academic psychology—not just part of its prehistory but part of it right now.

Assuming psychology wants to keep this research in the fold, I think they should acknowledge it, and not pretend that the only things psychology has to say about race are coming from the left side of the political spectrum. Psychology is also continuing to offer old-school racial and gender-essentialist explanations of social patterns.

Summary (including a mangled quote from John Tukey)

As the statistics quoted at the top of this post indicate, there are serious disparities in American society. But I’m not convinced that the profession of psychological science “has a sizable body of research that can be brought to bear to understand the persistent racism and subsequent confrontations we are witnessing in cities and communities across our nation and throughout the world.” I’m not convinced that the implicit association test and things like it have anything useful to add, and I’m not convinced that traditional racial explanations of social and economic disparities have anything to add.

I understand that, as individuals and as an organization, the members and leaders of the Association for Psychological Science would like to be helpful in these difficult times. As individuals, they can help in various ways: they can peacefully protest, they can join neighborhood watch groups to protect their communities, they can be good parents, friends, and neighbors, they can vote and write their congressmembers etc. They can be productive members of society and coach Little League and support local businesses, they can minimize their ecological footprint and bring a smile to work every day. They can give Ted talks, they can blog, they can even go on twitter. Lots of ways to make yourself useful. But, to paraphrase John Tukey, the combination of an archive of Psychological Science articles and an aching desire to be helpful does not ensure that reasonable help can be extracted from a given archive of articles.

This is similar to my feelings about social science and coronavirus. I get that we all want to feel useful, but maybe promoting old journal articles isn’t actually useful at all.

121 Comments

  1. Keith O'Rourke says:

    Nice post – especially the Tukeyesq quote.

    > There are no easy answers here.
    The current Ottawa police chief (who happens to be of African descent) essentially said the same thing – no one has the answers right now.

  2. Jordan says:

    As a social psychologist, I have almost the exact same attitude toward statements about what “the field” has to offer for addressing problems like Covid and racism. There’s little reason to think that what might be presented as the consensus within the field will be useful insight. Your view may come from the outside as a statistician, but it matches my own view from the inside having seen how much of the sausage is made.

    However, the APS being unlikely to provide reliable insight isn’t the same thing as the archive of psychological science lacking useful information. Might it be possible that there are some diamonds buried in the dungheap of psych studies? Or, might any diamonds be impossible to identify because so few published studies provide the information needed to evaluate them properly (e.g., data & code not made available)?

    • Andrew says:

      Jordan:

      Yes, some psychological science should be relevant here. For example, I think there is research on how people categorize others. While this won’t be directly relevant to fixing the New Jim Crow, etc., it should provide some insight. The problem is that, institutionally, the APS is set up to promote its journals’ work indiscriminately. Well, not quite indiscriminately, as they don’t seem to want to draw too much attention to the racial and gender essentialist work of Baumeister etc.. But, yeah, there’s gotta be some relevant social psychology work outside the “implicit association” and “embodied cognition” paradigms. and outside the “draw individual inferences from aggregate statistics” methodology.

  3. Travis says:

    science and politics are entirely different realms

    science contaminated with subjective political advocacy (no matter how noble) is no longer science

    • Paolo Inglese says:

      They were able to put into also religion, because why not?
      Soup is more delicious when you add ingredients, until… well, you know what happens.

    • Jacob Alter says:

      Science is definitely political. This blog is filled with examples of how biases compound to determine what constitutes “worthy” science. What topics are worth studying? Whose research is worth funding? What methods are valid? Whose research is elevated and whose is relegated to the background? What are the boundaries of scientific ethics? These are all inherently political questions and how we answer them is cemented in the politics of academic establishment. No amount of perceived “objectivity” and “detachment” is going to remove the inherent politicization of these questions: it’s the name of the game. We make science work by interrogating and working through the politics of what we do; ignoring this fact not only stagnates the culture of science, but stifles scientific inquiry and the search for truth.

      Furthermore, the simple act of telling the truth and reporting facts (without even adding commentary) can be viewed as “political advocacy” (e.g., the rise of “alternative facts”). Most anything can be viewed as political advocacy depending on who’s saying it and who’s listening. The boundary of science and politics isn’t anywhere as clear-cut as you claim, nor do I think it’s possible or desirable to make that boundary neat.

      • The God of Boundaries says:

        The other side of the coin of “everything is political” is that it dilutes the meaning of “political”. Sure, in a trivial sense everything IS political, since anything we do is somehow tied to political institutions, maybe through omission, as you say, but I don’t find that to be insightful at all. We DO draw lines between what’s political and what isn’t: I don’t think many people would agree with me if I were to start calling myself politician, or if I’d try registering my stamp collecting club as a political organization.

        Indeed, that is important to recognise here that when we are using words such as “political” or maybe “religious” we are not referring to the impotent fact that, yes, everything is somehow related to politics or religion. Rather we are referring to (active) participation in certain socially defined institutions.

        (I’m referring here to the widespread misunderstanding that atheism is religion too, since it takes a stance on a religious question).

        I can, of course, agree with you that “The boundary of science and politics isn’t anywhere as clear-cut as you claim, nor do I think it’s possible or desirable to make that boundary neat”, but… forgive me my sarcasm, but making the observation that questions regarding demarcation criteria are not clear cut is blisteringly obvious, and at worst can lead to nonsense such as claiming those boundaries don’t exist at all and somehow everything is political, religious or whatever.

        • Jacob Alter says:

          I see what you mean, but when I say that science is political, I don’t mean it in any “impotent” or “trivial” or “obvious” sense. I don’t mean it in the sense that science “has to do” with politics in some vague, circumspect manner. I mean that politics are FORMATIVE to science; I mean that science IS a politics. I mean that what we consider “science” is defined through and functions through systems of political power. And who wields that power, who benefits from it, and who is hurt by it is a result of capital-P Political factors. And the results have capital-P political consequences. This is true in one part because (1) scientific research is funded in large part by governments; (2) its results are interpreted and weaponized within the political arena. For many researchers, to do science is to — somewhat obviously — actively participate in these institutions in the narrowest sense. On the other hand, our means for producing, disseminating, and consuming science also are shaped by social attitudes and Political forces and they replicate and reinforce these systems. This affects what research is given power, validity, and funding. I don’t believe it waters down the meaning of “political” to call science political in that sense. Politics is inextricable. Just because it’s not readily apparent to every researcher all the time doesn’t diminish that.

          • Alex Gamma says:

            > I mean that what we consider “science” is defined through and functions through systems of political power.

            Science *defined* through systems of political power? No, that’s too much. Critical theory is a nice idea, but you can push it too far. Of course, science as practiced today is influenced heavily by political ideology and other extrinsic factors, but *defined*? No. The principles of science are defined as principles of observation and inference and it is widely understood that politics & co are a source of bias. Hence this blog.

      • Paolo Inglese says:

        What we see here is a bit more extreme than a natural subjectiveness.
        It’s not an unavoidable pressure by our minds about what we do, we say, and we research.
        Here, we are talking of using science as a mean to justify a belief based on prejudice. Here, we are talking of such a strong bias that makes all the mistakes disappear in front of the authors’ eyes.

  4. Garnett says:

    In my younger days as an anthropologist I saw a lot of hand-wringing about promoting anthropological relevance. It was usually accomplished by vague slogans like “fighting oppression” and “empowering marginalized voices”.

    I studied cooperative hunting behavior among modern hunter-gatherers, and every day I struggled to find a reason why studying this topic could make the world a better place. In the end, I think my most relevant accomplishment was to make people laugh and cringe at the weird food I ate.

  5. Kaiser says:

    As with many of Andrew’s posts, it pays to read through to the end!

    I have always enjoyed most those journal articles which have discussions and rejoinder. Would something along those lines improve journals? I appreciate this takes resources. Even an editor’s note on each published piece explaining why it is important, and what are its limitations could help separate publishing from complete endorsement.

  6. Jonathan (another one) says:

    In these troubled times, proclamations of care are clearly superior to everything else. You don’t have to actually help… you just have to stand ready to help, because you care. And standing ready with a body of literature that you’ve already published, well, it’s hard to see how you can do any better than that. You already helped before the times were troubled! Now you just have to reiterate how much you care, and you’re done, until it’s time for another press release because people have forgotten how much you care, or you need to explain that there’s some new thing you care about.

  7. yyw says:

    Psychology needs to demonstrate basic competency in analytical thinking first before worrying about the society at large.

    • Martha (Smith) says:

      Good point. So maybe routinely having online discussions of papers as they are published might be a way to demonstrate the state of basic competency (for better or for. worse) of the field in analytical thinking.

  8. paul alper says:

    Andrew began by quoting a part of an article by David Leonhardt in the NYT. Some other startling things Leonhardt alleges:

    “For most white Americans, interactions with the police happen rarely, and they’re often respectful or even friendly. Many white people don’t know a single person who’s currently behind bars.”

    “Incarceration rates for black men are about twice as high as those of Hispanic men, five times higher than those of white men and at least 25 times higher than those of black women, Hispanic women or white women.”

    • Anonymous says:

      “Incarceration rates for black men are about twice as high as those of Hispanic men, five times higher than those of white men and at least 25 times higher than those of black women, Hispanic women or white women.”

      What should the ratios be?

      • Andrew says:

        Anon:

        I don’t know what the ratios should be, but as noted in my above post, I think 3% is too much. 20% is an absolute disaster.

        • Steve Sailer says:

          What are the crime rates relative to the imprisonment rates? A 2011 Obama Administration report found:

          “In 2008, the off ending rate for blacks (24.7 offenders per 100,000) was 7 times higher than the rate for whites (3.4 offenders per 100,000).

          p. 11 of https://www.bjs.gov/content/pub/pdf/htus8008.pdf

          The basic problem is that blacks commit a vastly disproportionate share of murders in the United States. When you cite the government statistics, however, many people get very hot under the collar about it.

          • Joshua says:

            Steve –

            > The basic problem is that blacks commit a vastly disproportionate share of murders in the United States.

            Is that the basic problem, or is it an outcome of a more basic problem?

          • AJ says:

            Crimes committed is a function of how crime is defined, how the area is policed and how the laws are enforced. There are racial biases at every single step of that process. By the time the data makes it crime committed database, it is already very biased and flawed.

            We have no good measure of actual crimes being committed by race. What we do have crime recorded. This is a non-trivial distinction

            • Steve Sailer says:

              That would have been a persuasive-sounding argument over a half-century ago, but huge amounts of effort and money have been spent on solving it since then. As James Q. Wilson explained in 2002:

              “Estimating the crime rates of racial groups is, of course, difficult because we only know the arrest rate. If police are more (or less) likely to arrest a criminal of a given race, the arrest rate will overstate (or understate) the true crime rate. To examine this problem, researchers have compared the rate at which criminal victims report (in the National Crime Victimization Survey, or NCVS) the racial identity of whoever robbed or assaulted them with the rate at which the police arrest robbers or assaulters of different races. Regardless of whether the victim is black or white, there are no significant differences between victim reports and police arrests. This suggests that, though racism may exist in policing (as in all other aspects of American life), racism cannot explain the overall black arrest rate. The arrest rate, thus, is a reasonably good proxy for the crime rate.

              “Black men commit murders at a rate about eight times greater than that for white men. This disparity is not new; it has existed for well over a century. When historian Roger Lane studied murder rates in Philadelphia, he found that since 1839 the black rate has been much higher than the white rate. This gap existed long before the invention of television, the wide distribution of hand guns, or access to dangerous drugs (except for alcohol).”

              https://media.hoover.org/sites/default/files/documents/0817998721_115.pdf

              • James Whanger says:

                History matters. Historical racism matters. Historical economic exclusion matters. Reason matters. James Q. Wilson type reasoning does harm. It stops short of a full understanding and a delineation of the actual problem. It IS the problem encapsulated in sloppy critical thinking of a famous social scientist.

              • Martha (Smith) says:

                We need to be careful to think about what statements such as “Black men commit murders at a rate about eight times greater than that for white men” are really telling us. A statement like this is often interpreted as saying, ” There is something inherent about black men that makes them commit murder at a rate higher than that for white men.”
                Compare this with the statement, “Children with larger shoe sizes are better readers than children with smaller shoe sizes.” This may seems absurd on the face of it, until we consider the confounding factor of age: A child’s feet get larger as the child gets older; and children’s reading skill improves as they get older. In other words, the statement, “Black men commit murders at a rate about eight times greater than that for white men” is a statement of correlation, not one of causality — but people often interpret it as a statement of causality.

        • Anonymous says:

          That is not the ratios I was asking about.

          paul alper said: “Incarceration rates for black men are about twice as high as those of Hispanic men, five times higher than those of white men and at least 25 times higher than those of black women, Hispanic women or white women.”

          To take a concrete example, what should the ratio of the incarceration rates of hispanic men to asian women be?

        • Anonymous says:

          Figuring out what the ratios should be is obviously a very difficult question.

          Can you at least tell us what some of the covariates of the ratios should be?

          • Andrew says:

            Anonymous:

            I don’t know what you mean by “the covariates of the ratios,” but my point is simply that a rate of 3% of men having been in prison sounds to me like too large a number, and 20% seems like a disaster. As a society, something is going terribly wrong if this is happening. Some international comparisons are here.

            • Steve Sailer says:

              Are you objecting to the absolute size of the 20% to 3% ratio or to the relative size?

              You are aware that all of these arguments that academics are bringing up today were enormously persuasive-sounding to academics and influential upon political leaders and judges in the 1960s?

              Pretty quickly, New York City went from looking like Breakfast at Tiffany’s to looking like Taxi Driver.

              • James Whanger says:

                I would like to retweet this as support for my argument that you need to engage in more critical thinking.

      • James Whanger says:

        What Anonymous is saying and evidently doesn’t want to say explicitly is that he wants to know the causes for this increased rate — to put forth the argument that the rate of crime justifies the rate of incarceration, which is such an oversimplification that it is suggestive of problematic analysis.

        The point he misses is that these rates are not some KPI to be compared, but simply a crude estimate that reveals a widespread systematic problem with our society that has many causes. Importantly, one would have to be rather blinded by intentionally contrarian beliefs to not recognize that historical systematic economic exclusion combined with socially normative racism to be a primary cause.

    • Steve Sailer says:

      You’ll notice that no figures are offered for the imprisonment rate of Asians. It’s hard to reconcile the dogma of White Supremacy with how the criminal justice system imprisons a lower rate of Asians than of whites.

      • Terry says:

        Steve:

        A recent commenter on this site made accusations of anti-Asian bias because Asians were characterized as non-human “pawns”. I think all of us on this site have an obligation to be more inclusive of Asian lives whenever the intersection of race and crime is interrogated here. The implicit marginalization of Asian bodies must not be normalized! Indeed, are we not endangering Asian bodies by depersonalizing them in this way?

  9. Wonks Anonymous says:

    The above-cited paper has a politically right-wing take on racial politics, writing that “religion would have greater utility for regulating violent behavior among societies with relatively lower average IQs than among societies with relatively more cognitively gifted citizens,” which is academic-speak for “People with African descent need religion to stop them from criming.”

    Don’t put your words in other people’s mouths. Does the effect they write about go away if you remove African countries from their analysis? Does one even need to believe in the existence of “race” for their theory (they don’t mention it in their abstract)? Would it not apply if some people have lower IQs for purely environmental reasons like lead pollution?

    • Andrew says:

      Wonks:

      Indeed, they do not use the word “race” in their paper. But I think that’s what they’re talking about when they refer to country-average IQ and “groups of different mean cognitive ability.”

      To put it another way: if you think the claims in that paper are supported by their data and analysis (which I don’t), then (a) it has a lot of relevance to current debates about racial discrimination etc., indeed a lot more so than the implicit association test and other things on that APS webpage, and (b) it is consistent with racist attitudes that are what you’d expect to hear at your local country club. So, if the APS is going to write about their relevance to racial strife, I think they should be open about the fact that they’re publishing research which is consistent with racist attitudes. I suppose they stand by this research—after all, they published it in their top journal!—and they should be open about this position.

      • Alex Gamma says:

        Andrew

        But there *is* a way to study the hypothesis they formulate *without* being racist? If not, that would be a problem, wouldn’t it?

      • Steve Sailer says:

        “Racist attitudes” = “empirical attitudes” about facts that you’d prefer people would shut up about.

        • James Whanger says:

          Reporting data without it being filtered through critical thinking about the underlying premises, assumptions, and causes is propaganda. It is not some brave person reporting uncomfortable facts. It is revealing of the source trying to pull off this rhetorical slight of hand snd not enlightening of underlying reality.

          • Steve Sailer says:

            Speaking of “critical thinking,” do you really see yourself as engaged in “critical thinking” when you concoct excuses for not knowing the basic statistical facts about homicide in America?

            It sounds more like what Orwell called “crimestop.”

            • James Whanger says:

              The statistical facts do not exist in isolation. Your argument suggests that you believe they do. This suggests that you should engage in a bit more critical thinking about this topic that moves beyond the statistics and into the causal reasoning.

            • James Whanger says:

              I do recognize that this does not seem to be your area of strength and so I will understand if you leave it up to those of us capable of such.

      • Wonks Anonymous says:

        All of physics is “consistent with racist attitudes”, as that’s orthogonal to race. I’d expect that the vast majority of science is. The IAT has been used to directly analyze racial differences, the reason you don’t consider it “relevant” is because you (and many others, including me) don’t think those studies are very good. And if this study is as badly done, that would be another reason for it being irrelevant.

  10. Anonymous says:

    “I’m a statistician, and racism is central to the history of statistics. Galton, anyone?”

    Not just Galton. There’s a movement by a prominent Biostatistician to get some prize named after Fisher change because Fisher was a Eugenist. See here:

    https://twitter.com/daniela_witten/status/1268392722311700481

    There seems to be strong support for the notion that naming a prize after Fisher is ongoing institutional racism in statistics. If so, then naming so many statistical artifacts in stats books after Fisher exposes minorities to institutionalized racism during almost every stats course (one minory academic at Harvard reported in the twitter thread experiencing PTSD from having these kinds of arguments).

    I don’t see how Fisher doesn’t get scrubbed from Stats books and will be interested to see if Gelman’s texts all get scrubbed.

    • Andrew says:

      Anonymous:

      We have the First Amendment in this country! I don’t see how my texts are gonna “get scrubbed.” Let’s not get paranoid here.

    • somebody says:

      > There seems to be strong support for the notion that naming a prize after Fisher is ongoing institutional racism in statistics. If so, then naming so many statistical artifacts in stats books after Fisher exposes minorities to institutionalized racism during almost every stats course (one minory academic at Harvard reported in the twitter thread experiencing PTSD from having these kinds of arguments).

      There’s a clear difference between these two things. The prize was named in honor of Fisher well after the fact, without his involvement. The things that he invented are named after him because he invented it. Changing the former doesn’t actually scrub meaningful information while the latter does.

      Not coming down on one side or the other, just the slope doesn’t seem all that slippery to me.

      • Anonymous says:

        It didn’t take people to long in the twitter thread to make the leap from “Fisher prize honor bad” to “naming other stat things after Fisher bad”. Fisher’s getting canceled for sure. Your subtle distinctions wont make a difference.

        • somebody says:

          Idk man, I’m scrolling through it now and I haven’t seen a single person calling for what you’re asserting is definitely going to happen. There’s a couple of person who suggest it as a reductio ad absurdum. The closest is one person here
          https://twitter.com/rwcorty/status/1268557483158712321
          who suggests not using eponymous terminology at all because it reflect egocentrism, with zero hearts or retweets.

          It seems like you really want people to be calling for something you disagree with just so that you can feel better than them. Also, I don’t really see the purpose of lying about something that’s right in front of me. Are you hoping that people aren’t going to check and will just assume you’re right?

          • Anonymous says:

            Huh?

            The real point was Andrew mentioned Galton, but that was 150 years ago or whatever. There are similar issues much closer to home/now (for example Fisher).

            As for the rest, the way the winds are shifting *is* interesting. Those twitter threads are already so long and branchy it’s hard to go back and find specific ones, but I’ve seen ones suggesting renaming, ones suggesting teaching is the real arena rather than obscure prizes most haven’t heard of, ones talking about removing Galton and Fisher from R packages, and so on. There’s like a 100 added every hour.

            Do you doubting this is where it was going to go? How can Fisher be so evil you can’t name a prize after him (they were evening hoping past winners would renounce their prize as a kind of protest), but have his name slathered all over every intro to stat courses for social science majors? How can a prize name be institutionalize racism so extreme it gives harvard academics PTSD, but seeing his name throughout stat courses is no problem?

            • Andrew says:

              Anon:

              I don’t know if this helps, but I think it’s easy to teach an intro to stat course and not mention Fisher at all. I just checked Regression and Other Stories and it mentions Fisher zero times. This is not out of political correctness or whatever; it’s just that his work is so foundational that we don’t always give him credit by name. For example, we mention maximum likelihood and the logistic transformation. Fisher did important work on both these things in 1911 and 1925, or something like that, but we don’t bother citing him. Similarly, we don’t cite Laplace for all the things we do with the normal distribution. We do use the name “Bayes,” but that’s just because Bayesian has become the official name for certain statistical methods.

              I think the only way that I’d bring up Fisher’s name in an intro stat course is if I wanted to give historical background, but in general I’m not a fan of that sort of textbook thing where they talk about great figures of the past. I’d rather just get to the methods and the applications.

              • Michael Nelson says:

                Possibly you would talk about Fisher because you would like to preempt others’ efforts to portray the study and use of statistics as having been innocent or virtuous “from its inception.” Same reason I think it would be a good idea to talk about Galton in an intro biology course, or John Watson of “Little Albert” fame in intro psych. I once had a class of high school seniors react with incredulity that making one research subject think they had killed another subject would raise ethical problems, because they do that kind of thing all the time on reality TV. But I digress.

            • Anonymous says:

              Here’s just some examples I found quickly:

              A guy talking about removing Fisher from some R package:
              https://twitter.com/sbagley/status/1267857712391454721

              A guy currently writing a stats book and wanting to know how to handle Fisher:
              https://twitter.com/EricSchles/status/1268639467625156609

              A guy talking about renaming other stats stuff that were associate with Eugenicists:
              https://twitter.com/RoyAFrye1/status/1268545562514784256

              Again, I’m bit shocked that the idea this wasn’t going to stop at Fisher and this one prize is controversial.

              • somebody says:

                Here’s your first example

                https://twitter.com/sbagley/status/1268215158767144963?s=20

                Not “scrubbing from books”, providing additional context and replacing the example dataset which was an actual eugenics dataset.

                Your second example is quite plainly not calling for his named to be scrubbed from textbooks but rather asking for a sensible approach to the subject, also providing another example of contextualizing his views.

                Your third example is the reduction ad absurdum I talked about. That’s why he says this

                > Should we rename stuff after Lysenkoist “scientists”?

                Lysenkoism being a Soviet pseudoscientific campaign against genetics.

                Quite clearly, you aren’t actually reading the sources you’re linking.

                Despite not disputing the merit of changing the name of an award to avoid honoring a eugenicist, you’re already afraid of the slippery slope into censorship that you think may result, before that name change has actually happened, even though you acknowledged to me earlier that the censorship doesn’t follow logically from the name change of the award. And you’re so worried about it that it doesn’t seem to matter to you whether or not the examples you cite actually demonstrate the behavior, because it’s already so obvious to you what’s going to happen. If it’s so obvious to you, why bother with linking these bullshit examples you didn’t read at all? Just say it’s obvious to you, and I can agree or disagree? It’s not like you’re hoping other people also won’t read them, right?

            • somebody says:

              Your evidence for people calling for Fisher’s name to be scrubbed from the history books was a specific claim about what was happening on a link you posted, and that claim is false. My only explanations are

              1. You lied and hoped nobody would check.
              2. You never actually checked to see if what you insist is happening was actually happening, instead just assuming it was playing out that way.

              Both suggest a kind of motivated reasoning where you hope other people are doing something you find silly so that you can worry about it.

              Maybe the desire to scrub Fisher’s name from all the technical terminology is a general twitter trend, but I have had no exposure to it and you haven’t shown any evidence of real momentum behind it. That being the case, it’s not so obvious that I can’t “doubting this is where it was going to go”.

              • Anonymous says:

                Nothing gets past you “somebody”! I lied about the phenomena of Fisher being canceled by citing people who are literally removing Fisher from multiple R packes and then cleverly covered it up by providing links to the whole discussion.

                Clearly you’re hard to fool.

              • somebody says:

                You claimed “ I don’t see how Fisher doesn’t get scrubbed from Stats books”, then as evidence repeatedly linked to people not calling for or personally doing that.

                Not that I care about this issue very much, but this online argument strategy of buttressing claims with sources that one hasn’t read and don’t say what one thinks they say is deleterious to productive discussions.

              • Anonymous says:

                I’m willing to bet that it will happen, I’m also willing to bet that when it does people like your good self will pivot to deciding it’s entirely appropriate that it should happen, and will forget your previous scepticism.

              • Joshua says:

                somebody –

                Just the language in itself is interesting:

                > There’s a movement by a prominent Biostatistician to get some prize named after Fisher change because Fisher was a Eugenist.

                Which describes a “movement” and then notes that the “movement” is “by a [singular] prominent biotatistian.” Is it just me, or does a “movement” usually refer to a large group of people by definition, as in the “civil rights movement.”

              • Anonymous says:

                To Somebody, other Anonymous, and Joshua,

                I’ve made zero statement for or against removing Fisher’s name and don’t have as strong viewpoint on it. The whole issue isn’t driven by reason, it’s driven by power and feelings. I have none of the of the former and little of the latter.

                I stand by my prediction that based on the logic, assumptions and mood of that discussion that Fisher will get the cancel treatment. And note, it’s literally already started. One of the respondents talked about actively (currently) removing Fisher from multiple R packages used in the teaching of statistics.

              • somebody says:

                Let me quote myself here

                > Not coming down on one side or the other, just the slope doesn’t seem all that slippery to me.

                What I really care about here is the practice of repeatedly backing up your claim with links where the thing you say is happening, people calling for eugencists’ names to be scrubbed from textbooks, is not happening. The closest thing you brought up is someone who is explicitly not “scrubbing them” from an R module used in their math course, but replacing their datasets and providing additional context around their names.

                > I stand by my prediction that based on the logic, assumptions and mood of that discussion that Fisher will get the cancel treatment.

                And that’s fine! It’s fine to say “I feel like his name will eventually be removed from more and more things.” Just don’t lie! Doing so makes it hard for me to believe that you have no hard feelings on the subject.

    • Steve Sailer says:

      Here’s a thread by string theorist Ed Witten’s daughter Daniela Witten. She is on the warpath against Ronald Fisher
      this week:

      https://twitter.com/daniela_witten/status/1268392722311700481

      She is the dorothy gilford endowed chair and professor of statistics and biostatistics at the U. of Washington.

  11. Jonathan says:

    I have what may be an objection, which is that you cant rigorously get from psychology to people in prison without going through the legal field, and yet that appears to be the pretension. The legal field has wrestled with the issues of equity for over 50 years. That wrestling isn’t just a bunch of bullshit articles published in journals, nor is it a bunch of Harvard types preaching about what others should do. There is rampant ignorance about how the legal system works, and that perpetuates through educated classes who presume they know how it should work and thus how it works and can works, though they, in legal terms, dont know shit from shinola.

    There have been a slew of model legal codes developed. Want to know how far they’ve gone: to the point where any crime could be considered, for exculpatory purposes, to be a product of the mental conditions of poverty. That is, they extended the doctrine of necessity by coupling it with the conception that society can create mens rea through the imposition of unfair conditions. There is a lot of subtllety in that: mens rea means a form of idealized intent which is evaluated cirumstantially, but which occurs in an individual, so they took the crowd aspects, like when a riot is incited that is a separate and worse charge than just rioting, and said the state itself takes on the guilty role of incitement. You can see why this code failed to be enacted. But the ideas were gone over many, many times, and have been implemented in a number of ways. People dont even grasp the history of things like 3 srike rules: the point was to incorporate personal responsibility as a series of warnings where repeated offenses carried higher punishments. You know, learning. It became a problem because the system applies it unevenly. Why? Because no system is perfect, and because there are incentives which tilt results. One incentive is pressure from communities to get repeat offenders out of their communities. But then you get obviosuly unfair cases, so people pressure the opposite way. Except the communities still want repeat offenders out. They just want the system to be better at that, which is tough.

    I apologize for being on a soapbox, but the gall of presumption rankles. There are many hundreds of years of top minds working in law. You think maybe there’s been interest in equity along the way? Like, I dont know, maybe the Constitution? Like how the concept of equities was perverted by the Crown and how that appears in our lega system? Like why Miranda rights are required? Like why you are entitled to the assistance of counsel? Like why the prosecution must show you its case?

    It’s like when people say justice will only be served if the police are all convicted. In other words, throw out the guarantees of a trial and say you believe so therefore there is no reasonable doubt no matter what they argue. The application of the law is to prevent lynchings, not to enable the ones you want. That was the problem in the past: they allowed lynchings the people wanted, so black men were murdered with the connivance of the law. The response should not be to do the same.

    Since a post like this may draw fire for race, let me note the issue with Mr. Floyd’s murder is that the police treat black perceived common criminals differenly than they do white perceived common criminals. They are far more willing to use violence, and they approach encounters with black people with a higher expectation that force will be required. One way this expresses is what black people see as a hair trigger where the police suddenly become violent even when the victim is trying to comply: police expectations lead them to ‘handle’ the situation and to preserve their own safety by acting with violence first. And often with overwhelming violence because assaulted people may resist even if passively. Most people dont understand that an officer perceives even stiffness as resistance, if that officer is acting in the mindset that this encounter requires force. It isnt just deaths but the fear that any encounter with a police officer will result in a violent assault.

    The taint comes from the bottom, meaning that black perceived common criminals are treated with much higher violence. That taint spreads across the class. That is a legal phrasing, but it’s also mathematical; there is a class, call it whatever is inside a tent, and there is violence in the form of a camel, and the camel sticks its nose inside the tent, which means then the camel is in the tent. The intimation of violence becomes violence when the camel gets its nose inside the tent. That legal argument is hundreds of years old.

    I also want to note that police reform should be a conservative issue because black perceived common criminals deserve to have the same priviliges and immunities as white perceived common criminals. I say it this way because that’s a misunderstood legacy of the Movement: you cant just stop saying ‘boy’ to black men, cant just stop refusing service to well dressed black women, but you also have to treat the black people you perceive to be criminals the same as you do the white people you perceive to be criminals. That is the obligation of the state under the 14th Amendment.

    But please, the day when pschologists tell the legal system what to do is the day when all hope is lost.

    • Jonathan says:

      As a note, though I love My Cousin Vinnie, the idea that he went to law school at all and didnt know there is no surprise is not tenable. There is no surprise. There is no surprise. It’s a mantra when learning criminal law.

      • Never heard of Baumeister until recently.

        There is a lot of virtue signaling going on all across the Internet. I suppose that is better no virtue signaling. It seems to me that some percent of social scientists look upon others as research subjects which had driven me away from reading social science research. It has reflected in their conversations sometimes.

        As a couple of scientists I know say, ‘I’d rather converse with a fiction writer. So much of social science is boring.

        • Phil says:

          “There is a lot of virtue signaling going on all across the Internet. I suppose that is better no virtue signaling.”

          Yes. When people say ‘virtue signaling’ they almost always use the term negatively. And indeed it can be a bad thing, e.g. if people try to get credit for virtues they don’t have. But people are heavily influenced by what they perceive to be normal in society, and ‘virtue signaling’ can help virtues be seen as normal. To give an example I’ve used before: I never used to wear the little ‘I donated’ sticker after I donated blood — I almost felt like that would be a bad thing to do, as if the reason I were donating were to brag about it to other people. But I’ve completely changed my mind on that. I want people to think that it’s normal to donate blood — certainly I wish more than 8% of the populace did so — so now I wear the sticker and I wish other blood donors did too.

          Of course, one person’s virtue signaling is another person’s vice signaling, and vice versa. Those people who fly confederate flags or wear swastikas or whatever presumably feel like they are virtue signaling.

        • Dzhaughn says:

          A virtue signal’s value must be weighed against its cost, including that it counter-signals the virtue of humility.

    • Terry says:

      “the gall of presumption rankles”

      This wins the Joseph Ducreux award of the day.

      http://www.quickmeme.com/Joseph-Ducreux/

      All fun aside, good post!

    • AJ says:

      >Since a post like this may draw fire for race, let me note the issue with Mr. Floyd’s murder is that the police treat black perceived common criminals differenly than they do white perceived common criminals. They are far more willing to use violence, and they approach encounters with black people with a higher expectation that force will be required. One way this expresses is what black people see as a hair trigger where the police suddenly become violent even when the victim is trying to comply: police expectations lead them to ‘handle’ the situation and to preserve their own safety by acting with violence first. And often with overwhelming violence because assaulted people may resist even if passively. Most people dont understand that an officer perceives even stiffness as resistance, if that officer is acting in the mindset that this encounter requires force. It isnt just deaths but the fear that any encounter with a police officer will result in a violent assault.

      As someone who has had the luck to be black all my life, you put it better than I ever could. My lived experience is perfectly explained by your last sentence

  12. Peter Dorman says:

    I think the problem begins with what Andrew is saying and goes further. Consider implicit bias testing. The test results are unstable, and worse, low implicit bias as tested does not correspond to less biased behavior. Yet — and this is my point — for ideological reasons the implicit bias framework has become hegemonic in institutional trainings and “reforms”. I wouldn’t be surprised if all four of those Minneapolis cops had had implicit bias trainings.

    Anecdote: My former institution, Evergreen State College, had a big blowup a few years ago that made national news. The precipitating issue (not well reported) was the demand for mandatory bias training for faculty. Faculty voted against the mandatory part, so advocates drew the conclusion that faculty were irredeemable racist and needed to be forcibly retrained. It was a horrible time to be there. But the irony is that bias training, despite its noble intentions (which I don’t dispute) has a poor track record, as innumerable police department outcomes demonstrate.

    How responsible is the psychology profession for this state of affairs? I think the larger responsibility falls on the leftward side of the political spectrum, which has elevated consciousness change as a strategy at the expense of altering institutional and economic structures that limit the ability of abused people to challenge their abuse. (At Evergreen, despite the turmoil, the option of including language in the faculty’s collective bargaining agreement to make equitable treatment of students a condition of employment was not placed on the table.) But psychology is culpable for taking advantage of the opportunity for greater influence, rather than cautioning the consumers of its studies that evidence for the transformative power of consciousness changing interventions is weak.

  13. Psyoskeptic says:

    I’m reminded of the Amicus brief from the APA that was cited in the Brown v. Board of Education verdict from the supreme court making segregation illegal. I’m reminded because the APA successfully assisted in bringing about a good outcome even though the science from Clark & Clark that was used to bolster it was far from ideal and Andrew would tear it apart if it came out today (they essentially made the conclusion they wanted to make from results that were the opposite of what they expected).

    So, you never know, they might end up doing something good in spite of themselves. If they do it won’t be any good in the long run for psychology because they’ll be insufferable afterwards.

  14. Michael Nelson says:

    The APA makes a strong and positive contribution to mental health, and to society in general, via the DSM. The product is flawed, and the process is flawed, but it does at least attempt to capture “the voice” of the field. Yeah, the DSM used to say homosexuality is a disorder, but that’s because that was the (bigoted) consensus. It is both the product of, and a driver of, social change.

    The DSM can have this influence because it’s intended to be used by psychologists acting in their traditional roles. If the APS wants to be similarly influential, and effective, it needs to put a lot more thought into its model. It needs an inclusive process for determining what our “voice” is, and it needs to fit the product to our traditional roles. There are all kinds of challenges to doing that, but that’s what a sincere effort would look like. It would not look like a vague call to its members to “transform this knowledge into actions.”

    • Harold Duke says:

      @ Michael Nelson

      I am all for healthy discussion and dissenting voices.

      Here is a starting point for the carnage that has been unleashed by the APA and the DSM and “black box” questionairres. And notice also the support from the Center For Mental Health Policy. Then remember that the NIMH Directors Advisory Council had the same folks “whispering” policy in their ear. Not long after, the NIH Directors Advisory Council had the same folks “whispering” in their ear.

      Psychometric Properties of the Vanderbilt ADHD Diagnostic Parent Rating Scale in a Referred Population Mark L. Wolraich,1 MD, Warren Lambert,2 PD, Melissa A. Doffing,1 MA, Leonard Bickman,2 PD, Tonya Simmons,2 BS, and Kim Worley,2 MD 1Child Study Center, University of Oklahoma Health Sciences Center, and 2Center for Mental Health Policy, Vanderbilt University

      I am happy to provide a complete timeline since then as well. DSM5 has been roundly panned. The industry influence and secrecy surrounding its “revamp” is well documented.

      Or I could have simply said to see link. https://m.youtube.com/watch?v=BgMD66_yBu0

    • yyw says:

      “The APA makes a strong and positive contribution to mental health, and to society in general, via the DSM.”

      What’s the basis of such a strong statement?

    • Andrew says:

      Terry:

      I think it’s worth studying comparisons across groups, and I also think the overall statistics, as quoted by Leonhardt at the top of this post, are relevant too. In statistical terms, the conditional distributions and the marginal distributions are both relevant to our understanding.

      • Joshua says:

        Andrew –

        > I think it’s worth studying comparisons across groups, and I also think the overall statistics, as quoted by Leonhardt at the top of this post, are relevant too.

        I can’t get past the paywall, but I’ve read other articles she has written. As such, I think that it is important to foreground the low quality of the data. Often the data we have are simply a product of the reports of the people involved, without any kind of independent confirmation. And the categories are vaguely defined and applied. and not validated As such, the data for comparisons across groups have limited value in giving us context for video evidence of situations such as the video evidence of what happened in Minny.

        • Steve Sailer says:

          While statistics on petty crime vary much in quality from jurisdiction to jurisdiction, homicide statistics are pretty carefully collected in just about all jurisdictions in America, because a body with a hole in it demands attention.

          For calculating the ratio of black and white homicide ratios, which the Obama Administration estimated as blacks being about 7.6 times as homicidal as whites, here are a still a couple of sizable problems with the FBI’s aggregation of local jurisdiction data.

          First, unlike in lesser violent crimes, the chief witness, the victim is no longer around to be interviewed. So the federal government’s statistics on the race of homicide offenders must leave out the large fraction of “uncleared homicides.” How does this bias the Racial Ratio? Well, homicide clearance rates are generally notably worse in black-majority and black-run cities, and in cases where the dead victim is black. A lot of evidence points to homicides carried out by whites as being cleared at a higher rate than among blacks. Outside of, say, the Mafia, whites tend not to have a strong culture of “Snitches Get Stitches.” Witness murdering is more common in black neighborhoods, which does indeed discourage witnesses against black gang members.

          The other methodological issue is that the federal statistics on homicides by race frequently don’t break out Hispanics from whites. This goes back to liberal demands of the 1940s and 1950s for Latinos to be classified as white on the 1950-1960 Census. So the federal Bureau of Justice Statistics traditionally lumped Latinos and whites together.

          Both methodological issues suggest the true black to white racial ratio of homicide rates is higher than 7.5. I usually suggest: an order of magnitude. But I’m perfectly happy if newcomers to this immense amount of data simply go with 7.5X.

          • James Whanger says:

            What you seem determined to avoid are the underlying fundamental causes for the statistics. Racism is supported by those willing to pretend that history does not matter. That words and laws are enough to solve complex historical problems of racial discrimination and economic exclusion. They aren’t. Your arguments ring hollow to anyone who has the slightest understanding of causal reasoning.

            • Steve Sailer says:

              Does history and prehistory before 1619 matter?

              • James Whanger says:

                Do you understand the reality of slavery that transitioned to separate but equal to jim crow to socially acceptable but illegal racism?

                In other words — your question is irrelevant to the topic under discussion. THAT is critical thinking at work.

            • Alex Gamma says:

              James

              What is your model of the underlying causality? Broad strokes.

              • Joshua says:

                Alex –

                What is your mechanism of causality that explains why in rural areas there is no racial disparity in murder, and why there is no racial disparity in aggravated assault among young men?

              • James Whanger says:

                slavery -> separate but equal -> jim crow -> economic exlusion + socially acceptable racism -> economic desserts in almost every large city populated by majority african american with spatterings of other immigrant communities.

                When we believe that the only thing that needs to change is making racial discrimination illegal, we are engaged in farcical analysis. When we believe that voter suppression is not targeting black communities we are engaged in farcical analysis. When we believe that individual agency is the solution to systemic problems, we are engaged in farcical analysis.

              • Steve Sailer says:

                Dear Joshua:

                “What is your mechanism of causality that explains why in rural areas there is no racial disparity in murder, and why there is no racial disparity in aggravated assault among young men?”

                Link, please.

              • Joshua says:

                I’m going to bow out at Andrew’s request. But it’s interesting that those stats blow your mind. Maybe you need to rethink your working causal model?

                You seem to take an interest in this topic. Intersting that you’re unaware nonetheless. Google it and risk getting outside your echo chamber.

          • James Whanger says:

            Racism is further supported by social scientists like James Q. Wilson who seemingly lack the analytic ability to fully map the problem and thus their analysis reflects a superficial understanding and in turn their solutions do as well.

            • Steve Sailer says:

              Obviously, you are so much smarter, wiser, and more fair-minded than the late James Q. Wilson, as proven by how Wilson knew lots of HateStats, while you’ve kept your brain innocent from all inconvenient knowledge.

              • James Whanger says:

                Not claiming I’m smarter, just that I recognize things he didn’t. You seem to believe he identified and conceptualized all there is to know. I do not share that belief.

      • Terry says:

        Agree. The incarceration rates among young black men is staggering.

  15. yyw says:

    The low quality context data seem to be the primary driver behind the dominant interpretation of the video evidence. If we believe the data are of low quality and unreliable, then we should avoid reaching firm conclusion.

  16. JIM says:

    “But that’s not the only way that the field of psychological science has studied the causes and harmful impacts of racial disparity.”

    It’s interesting that a racial *difference* is presumed to be a racial *disparity* – e.g., caused by race.

    Recently a columnist in the local newspaper talked about the “racial” disparity in COVID19 cases. But it turns out the disparity isn’t *caused* by race or racism just because it affects one “race” differently than another. Hispanics, the singled-out group in this case, have a large percentage of the population that a) don’t speak English; b) are recent immigrants from poor countries; c) live in multigeneration households just like they did in their country of origin; d) are economically disadvantaged largely because they are poorly educated first- and second-generation immigrants.

    So a lot of the impact of COVID-19 on Hispanics has nothing to do with race or racism, even though it shows up in “racial” statistics.

    Isn’t it possible that similar cultural factors could be responsible for other purported “racial” disparities?

    • Joshua says:

      JIM –

      > So a lot of the impact of COVID-19 on Hispanics has nothing to do with race or racism, even though it shows up in “racial” statistics.

      How do you determine the causal mechanism behind the health impacts of racism, so as to conclude that a lot of those impacts have “nothing to do with racism?”

      Please show your work.

    • James Whanger says:

      It is certainly true that immigrants from different countries have different experiences that can have causal effects that are independent of racism. Immigrants come with beliefs and norms developed in their country of origin, some of which are adaptive to making their way in America and some are not. There are likely regional effects from the area of the world and also regional effects related to the area of the country in which they settle. Different areas can make it more or less difficult to be an immigrant. Different areas can pass racist legislation, enforce racist legislation, and police racist legislation in ways that make it more difficult to be an immigrant. To casually account for this reality with the assertion that what is causing these problems is a “lack of education” with not even a nod to the types of racist legislation and enforcement is bit sloppy.

      If one then makes the statement:

      “Isn’t it possible that similar cultural factors could be responsible for other purported “racial” disparities?”

      It seems clear that this is an attempt to extend recent immigrant experience to the experience of black people in America of historical slavery -> Jim Crow + separate but equal -> economic exclusion + socially acceptable racism.

      JIM’s argument reminds me of Chris Rock bit where he challenges his audience of mostly white people to change places with him a black person who is wealthy. Think about that for moment. It is some insightful social science.

  17. Many people are calling for a mandatory racial IAT for police. I think that’s unwise, both because of the numerous problems with the IAT and because of the intrusiveness (even flawed intrusiveness) of such a measure. It is one thing to discuss implicit bias during police trainings–even to bring up the IAT as an attempt to measure it. But purporting to reveal someone’s hidden thoughts through a test of this kind is not only spurious but unethical. The urgent priority should be to stop police violence and killings, and to screen police for overtly racist attitudes. Private thoughts and feelings should be left alone, except to the extent that the thinkers of those thoughts choose to examine them. This applies to the larger workplace as well; in general, personality tests and attitude tests should not be used in hiring or training. What matters is that people do the job well, conduct themselves appropriately, and cooperate as needed with others.

Leave a Reply