Religion is a Mental Illness

Beware the Univariate Fallacy

How activists use a simple statistical trick to invent or eliminate group differences to suit their purpose.

realityslaststand.com

When we shove the concept of athletic ability—strength, for instance—into the same black-and-white binary that we try to put gender into, we’re wrong. There is no stark line separating what men can do athletically and what women can. Some women, in fact, are bigger, faster, and stronger than some men.

Since the early 20th century, elite sport policies worked to pathologize and control women’s bodies and enforce dimorphic sex. There is, however, a significant overlap in all sexual characteristics. ‘Male’ and ‘female’ are not mutually exclusive categories and should not be treated as such.

Yet, as The Gendered Brain reveals, conclusive findings about sex-linked brain differences have failed to materialize. Beyond the “missing five ounces” of female brain — gloated about since the nineteenth century — modern neuroscientists have identified no decisive, category-defining differences between the brains of men and women.

It has long been known that there is no single biological measure that unassailably places each and every human into one of two categories — male or female. In the 1950s the psychologist John Money and his colleagues studied people born with unusual combinations of sex markers (ovaries and a penis, testes and a vagina, two X chromosomes and a scrotum, and more). Thinking about these people, whom today we would call intersex, Dr. Money developed a multilayered model of sexual development.

In fact, there’s a range of DSDs that can involve either rare combinations of sex chromosomes (e.g. XXY or XYY) or genetic mutations on other chromosomes that affect sexual development. Thus, no single factor, genetic or otherwise, neatly determines whether an individual is male or female. Furthermore, no formula exists that uses genetic and other data to produce an answer of whether someone is male or female. The IAAF regulation disregards this reality and was surprisingly unenlightened considering the chequered history of genetic sex testing in sport.

Yet even after these tests, determining the sex of an individual is not straightforward. There are at least six biological markers of sex: chromosomes, gonads, hormones, secondary sex characteristics, external genitalia, and internal genitalia. Each one contains significant variation, both within and across individuals, including testosterone. As a result, no single marker can decisively categorize a person as male or female.

Dr. Vilain stated that as a leading expert in this field he “fundamentally disagreed” with the notion that females with DSD should be “defined” as male and “have male bodies”. That proposition is contrary to the current scientific mainstream, which recognizes that sex is not binary but rather a spectrum, where no single factor (e.g. presence or absence of testes) prevails above all others.

DSDs—which, broadly defined, may affect about one percent of the population—represent a robust, evidence-based argument to reject rigid assignations of sex and gender. Certain recent developments, such as the Swedish adoption of a gender-neutral singular pronoun, and the growing call to stop medically unnecessary surgeries on intersex babies, indicate a shift in the right direction.

The ‘Intersectionality’ Canard | National Review

Many studies that purport to find giant residual effects of race or sex are flawed from the outset.

National Review

Many studies that purport to find giant residual effects of race or sex are flawed from the outset.

‘Intersectionality” is just a badly done “woke” version of regression analysis.

The old feminist idea of intersectionality has been popping up across the mainstream media of late, as the latest round of the national debate over “DEI” (and CRT, ESG, SEL, NU-HR, and the rest of today’s insufferable corporate alphabet soup) rages on. Its resurgence seems like a worthwhile topic, while I am on a 3–4-week run of discussing academic issues for the gentle readers of National Review.

Per Merriam-Webster, which updated its definition of the term November 30, 2023 — the major dictionaries have been doing that kind of thing a lot lately — intersectionality is “the complex, cumulative way in which multiple forms of discrimination (such as racism, sexism, and classism) combine . . . especially in the experiences of marginalized individuals or groups.” The United Nations’ Global Citizenship initiative has, also within the past year or two, adopted this concept as a primary analytical framework, and defines “intersectionality” as “how multiple identities interact to create unique patterns of oppression.”

“In the United States,” author and Global Citizen Sarah El Gharib declaims, “Women earn 83 cents for every dollar a man earns.” But, the situation is even worse for black women, who pull in “a mere 64 cents for every dollar a white man earns.” The reason for all of this? Obviously, oppression: The analysis almost invariably stops there.

The problem with all of this, which needs to be discussed if radical-feminist analysis — intersectionality as a concept was first outlined by UCLA’s Kimberlé Crenshaw in the late 1980s, and traces its roots back to “a Black lesbian social justice collective formed in Boston in 1974” — is now prevalent in the United Nations and around the Fortune 500, is fairly basic. The idea that multiple independent variables can influence a dependent variable like income is not exactly a new one. And, the actual range of potential “IVs” that can do so extends well beyond race and sex to include: age, the regions where people and groups live, test and IQ scores, patterns of study time, crime rates, desire to work at all (in the context of men vs. women), and so on down the line.

Simply put, racism or sexism can only be said to exist where we find that pretty much identical people, who differ only in terms of the characteristic of race or sex, are still being treated differently — after all of the other factors which might explain performance differences between them have been accounted for. This sort of real bigotry is, today, fairly rare. Many “intersectional” studies that purport to find giant residual effects of race or sex on some specific thing — individuals’ chances of going to prison, let’s say — literally just consist of unadjusted comparisons between citizens in two or more different groups.

This, however, is not how serious people conduct this sort of analysis. The pay gap between men and women, in fact, provides one of the best examples of an apparently giant gulf which vanishes almost as soon as anything but sex is competently adjusted for. As it turns out, one major reason that women make so little money relative to men — less than 70 cents per dollar, in some analyses — is that 39 percent of women “prefer a home-maker role” and about one-third are housewives . . . who often earn almost no money, but have access to all of the resources of what is usually a middle-class household.

Even if we focus only on working men and working women, it remains the case that males and females prefer to work different jobs, men work slightly longer hours, men took virtually no time off from work for pregnancy and child care until quite recently, and so forth. When the quantitative team at the PayScale business website took all of this into account and ran some models, they found that any actual gap in same-job wages which could be attributed to sexism would be on the order of –(1 percent). At some level, this is not even surprising: American corporate business is ruthless, and any trading floor or shark-tank start-up that could actually save 17–31 percent on labor costs by hiring only women would do so immediately.

Pay gaps between white and black guys, for that matter, do not survive serious analysis. As I have noted elsewhere, the labor economist June O’Neill attempted, back in the 1990s, to distinguish the impact of racism from that of plain human capital on the B/W wage gap. What she found was stunning, almost remarkable. An initial gap of 15–18 percent, which has been attributed to “racism” by almost everyone to write about it during the modern era, in fact shrunk to about 1 percent when adjustments were made for basic variables like the mean age of each racial population, region of residence, and IQ- or aptitude-test scores.

O’Neill and a co-author found almost exactly the same pattern to still hold more than a dozen years later, in 2005. As both she and I have pointed out, groups that are different as re very major traits such as race and religion also invariably vary in terms of other characteristics — and any effects of racism simply cannot be parsed out without adjusting for all of these important differences. Simply put, there is no reason to expect a 27-year-old black man living in Mississippi to earn anything like as much as a 58-year-old white dude with a residence in mid-town Manhattan.

What is true in the critical context of money is true almost everywhere else. For years, the “Black Lives Matter” movement argued that young African Americans are being “murdered” or “genocided” by police officers, because members of this group are more likely to be shot by law enforcement than members of the general public. Again, however, there is an elephant in the room. As the Manhattan Institute’s Heather MacDonald has pointed out for decades now, the crime rate for black Americans — certainly before we adjust for age, or sex ratios, or living in mile-spire cities instead of Green Acres — is about two to 2.5 times that for whites. As an obvious result, we tend to encounter on-duty cops about that much more often.

Just adjusting for this one variable entirely removes the gap in rates-of-shooting. In the fairly representative year of 2015, which I select for analysis in my brilliant and best-selling book Taboo, there were 999 fatal police shootings nationwide — out of tens of millions of police/citizen encounters — of which 250 (25.1 percent) involved African Americans. That figure, which is 1.92 times the nation’s black population percentage, is almost exactly what any reasonably intelligent person would expect to see after taking a single glance at the crime statistics — if anything, a bit on the low side.

Entertainingly, the Reilly Rule about the impacts of the real, multi-variate version of “intersectionality” on day-to-day life applies even in the context of “white privilege.” As it happens, there exist several scales that attempt to measure personal privilege — such as this popular but quite empirical example, which several hundred thousand people have taken (a little bird tells me the average score is 43). When I have administered the 100-item ordinal survey, which includes Yes/No questions ranging from “I have never gone to bed hungry” to “I went to private school,” to sizable groups as a learning exercise, I do find that being white does have a small effect on ease-of-life: about two–three points, with all else adjusted for.

However, almost everything else has a bigger one. Other more influential variables recorded by myself and others to work with the test include female sex (yes, sure) — but also where people live (the suburbs as vs. the “hood or the “holler,” the North vs. the South), being gay rather than straight, and most notably plain social class. The largest chunk of “privilege” appears to be pure socio-economic status: crudely put, how much money a test taker and his or her family happen to make in a year. Across the aforementioned 100 questions, poor Appalachian or immigrant respondents often post “have not experienced” scores on the order of 17, while well-off ones “achieve” 69s and 73s.

At some level, none of this is particularly surprising, to the average human being with eyes. Of course, having wealthy parents, or not committing crimes, or not living on an isolated farm, or being a 6’4” blonde or black jock might sometimes help you along in life. However, this empirical point is a useful rebuttal to the much simpler standard idea of intersectionality — that what matters is race, or sex alone, or perhaps something like “being non-binary.”

In reality, conservatives don’t make fun of that simplistic concept because we are too unsophisticated to understand it, some pack of rubes who believe that only hard work and lovin’ America predict life outcomes. Instead, we do so because we recognize that many, many factors predict those outcomes. And, in the end, if dozens or hundreds of things predict where each singular human being will end up in life, we should turn our focus back to that smallest and most vulnerable of minorities: the individual.

Thread by @wil_da_beast630 on Thread Reader App

@wil_da_beast630: I'll do a quick response here, since this is my article. Obviously, no one argues that "racism does not exist." The point

threadreaderapp.com

I'll do a quick response here, since this is my article.

Obviously, no one argues that "racism does not exist." The point is that you do not DETERMINE the existence of racism simply by pointing out "performance gaps" re something like income or police encounters - which is literally the level of a lot of 'woke' research....or by adjusting for sex as well as race (whee!).

As J. O'Neill pointed out 20+ years ago, most such gaps close or vanish after basic adjustments for things like age, region, any aptitude test score, etc.

(2) At some very basic level, it makes no sense to argue that, if a 27-year old Black Mississippian with a community college degree makes less money than a 58-year old white Bostonian who went to BU, the reason is "racism."

These are the sort of gaps political scientists often look at between large groups. More whites DO live in the US North (the boats landed further South). That IS the gap in at least modal average/most common ages between Blacks and whites...

The most common age among whites in U.S. is 58 – more than double that of racial and ethnic minorities

The most common age was 11 for Hispanics, 27 for blacks and 29 for Asians as of last July. Multiracial Americans were by far the youngest ra

Pew Research Center

(3) A common response from smart left-slanting stats folX, including Kareem, is that these other variables (age?!) could themselves just be measures of racism.

But, especially given that we can easily test for multi-collinearity and covariance, there is almost never any evidence presented of this. Aptitude test scores, for example, are higher for white kids from families making $40,000 per year than for Black kids from families making $200K per year.......and don't vary at all with reported racism. The obvious actual predictor here (attached) is study time.

The core point of my article () is quite simple - the "intersectional" idea that TWO or even THREE variables can affect a dependent variable is not very novel or original.

Of course both sex and race can influence your life outcomes - but so can social class (!!!), IQ, prey drive, attractiveness and fitness, age, level of education, being gay or lesbian, being from the country, hailing from the South, being white in the academic job market, just etc. Figuring all this out is the basic idea of multi-variate analysis.

We have to take some basic precautions as re how we model these things, but a researcher who finds that Black women earn 'just' 73 cents for every dollar white men do has not in fact "gotten to the bottom of the matter."