Re Nichols: Times the Experts were Wrong, pt 2

Welcome back. In preparation for our review of The Death of Expertise: The Campaign Against Established Knowledge and Why it Matters, I have made a list of “times the experts were wrong.” Professor Nichols, if you ever happen to read this, I hope it give you some insight into where we, the common people, are coming from. If you don’t happen to read it, it still gives me a baseline before reading your book. (Please see part 1 for a discussion of relevant definitions.)

Part 2: Law, Academia, and Science

Legal Testimony

If you’ve had any contact with the court system, you’re probably familiar with the use of “expert testimony.” Often both sides of a case bring in their own experts who give their expert testimony on the case–by necessity, contradictory testimony. For example, one expert in a patent case may testify that his microscopy data shows one thing, while a second testifies that in fact a proper analysis of his microscopy data actually shows the opposite. The jury is then asked to decide which expert’s analysis is correct.

If it sounds suspicious that both sides in a court case can find an “expert” to testify that their side is correct, that’s because it is. Take, for example, the government’s expert testimony in the trial of Mr. Carlos Simon-Timmerman, [note: link takes you to AVN, a site of questionable work-friendliness] accused of possessing child pornography:

“When trial started,” said Ramos-Vega, “the government presented the Lupe DVD and a few other images from the other DVDs that the government understood were also of child pornography.  The government presented the testimony of a Special Agent of Immigration and Customs Enforcement that deals with child pornography and child exploitation cases.  She testified that Lupe was ‘definitely’ under 18. The government then presented the testimony of a pediatrician who testified that she was 100 percent sure that Lupe was underage.”

The experts, ladies and gents.

After the prosecution rested its case, it was Ramos-Vega’s turn to present witnesses.

The first witness we called was Lupe,” he said. “She took the stand and despite being very nervous testified so well and explained to the ladies and gentlemen of the jury that she was 19 years old when she performed in the videos for littlelupe.com.  She also allowed us to present into evidence copies of her documents showing her date of birth.”

So the Customs Special Agent and the pediatrician were both LYING UNDER OATH about the age of a porn star in order to put an innocent man in prison. There were multiple ways they could have confirmed Lupe’s age (such as checking with her official porn star information on file in the US, because apparently that’s an official thing that exists for exactly this purpose,) or contacting Lupe herself like Mr. Simon-Timmerman’s lawyer did.

Unfortunately, this is hardly the first time trial “experts” have lied:

The Washington Post published a story so horrifying this weekend that it would stop your breath: “The Justice Department and FBI have formally acknowledged that nearly every examiner in an elite FBI forensic unit gave flawed testimony in almost all trials in which they offered evidence against criminal defendants over more than a two-decade period before 2000.”

“Of 28 examiners with the FBI Laboratory’s microscopic hair comparison unit, 26 overstated forensic matches in ways that favored prosecutors in more than 95 percent of the 268 trials reviewed so far.” …

Santae Tribble served 28 years for a murder based on FBI testimony about a single strand of hair. He was exonerated in 2012. It was later revealed that one of the hairs presented at trial came from a dog.

Professor Nichols, you want to know, I assume, why we plebes are so distrustful of experts like you. Put yourself, for a moment, in the feet of an ordinary person accused of a crime. You don’t have a forensics lab. Your budget for expert witnesses is pretty limited. Your lawyer is likely a public defender.

Do you trust that these experts are always right, even though they are often hired by people who have a lot more money than you do? Do you think there is no way these experts could be biased toward the people paying them, or that the side with more money to throw at experts and its own labs could produce more evidence favorable to itself than the other?

Now let’s expand our scope: how do you think ordinary people think about climate scientists, medical drug studies, or military intelligence? Unlike drug companies, we commoners don’t get to hire our own experts. Do you think Proctor and Gamble never produces research that is biased toward its own interests? Of course; that’s why researchers have to disclose any money they’ve received from drug companies.

From the poor man’s perspective, it looks like all research is funded by rich men, and none by poor men. It is sensible to worry, therefore, that the results of this research are inherently biased toward those who already have plenty of status and wealth.

The destruction of expertise: “Studies” Departments

Here is a paper published in a real, peer-reviewed academic journal:

Towards a truer multicultural science education: how whiteness impacts science education, by Paul T. Le, (doctoral candidate from the Department of Integrative and Systems Biology at the University of Colorado) and Cheryl Matias, (associate professor at the School of Education and Human Development, University of Colorado) (h/t Real Peer Review):

The hope for multicultural, culturally competent, and diverse perspectives in science education falls short if theoretical considerations of whiteness are not entertained. [Entertained by whom?] Since whiteness is characterized [by whom?] as a hegemonic racial dominance that has become so natural it is almost invisible, this paper identifies how whiteness operates in science education such that [awkward; “to such an extent that”] it falls short of its goal for cultural diversity. [“Cultural diversity” is not one of science education’s goals] Because literature in science education [Which literature? Do you mean textbooks?] has yet to fully entertain whiteness ideology, this paper offers one of the first theoretical postulations [of what?]. Drawing from the fields of education, legal studies, and sociology, [but not science?] this paper employs critical whiteness studies as both a theoretical lens and an analytic tool to re-interpret how whiteness might impact science education. Doing so allows the field to reconsider benign, routine, or normative practices and protocol that may influence how future scientists of Color experience the field. In sum, we seek to have the field consider the theoretical frames of whiteness and how it [use “whiteness” here instead of “it” because there is no singular object for “it” to refer to in this sentence] might influence how we engage in science education such that [“to such an extent that”] our hope for diversity never fully materializes.

Apologies for the red pen; you might think that someone at the “School of Education” could write a grammatical sentence and the people publishing peer-reviewed journals would employ competent editors, but apparently not.

If these are “experts,” then expertise is dead with a stake through its heart.

But the paper goes on!

The resounding belief that science is universal and objective hides the reality that whiteness has shaped the scientific paradigm.

See, you only think gravity pulls objects toward the earth at a rate of 9.8 m/second^2 because you’re white. When black people drop objects off the Leaning Tower of Pisa, they fall 10m/s^2. Science textbooks and educators only teaching the white rate and refusing to teach the black rate is why no black nation has successfully launched a man into space.

Our current discourse believes that science and how we approach experimentation and constructing scientific explanations is unbiased, and on the surface, it may seem justified (Kelly 2014). However, this way of knowing science in the absence of other ways of knowing only furthers whiteness an White supremacy through power and control of science knowledge. As a result, our students of Color are victims of deculturization, and their own worldviews are invalidated, such as described by Ladson-Bilings (1998a).

For example, some Aboriginal people in Australia believe that cancer is caused by curses cast by other people or a spiritual punishment for some misdeed the sufferer committed. Teaching them that cancer is caused by mutated cells that have begun reproducing out of control and can’t be caused by a curse is thus destroying a part of their culture. Since all cultures are equally valuable, we must teach that the Aboriginal theory of cancer-curses and the white theory of failed cellular apoptosis are equally true.

Or Le and Matias are full of shit. Le doesn’t have his PhD, yet, so he isn’t an official expert, but Matias is a professor with a CV full of published, peer-reviewed articles on similar themes.

You might say I’ve cherry-picked a particularly bad article, but give me 10 minutes and I’ll get you 100 more that are just as bad. Here’s one on “the construction of race in contemporary PE curriculum policy.”

Every single degree awarded paper published on such garbage degrades the entire concept of “experts.” Sure, Nichols is a professor–and so is Matias. As far as our official system for determining expertise, Nichols, Matias, and Stephen Hawing are all “experts.”

And this matters, because the opinions of garbage experts get cited in places like the NY Times, and then picked up by other journalists and commentators as though they were some kind of valid proof backing up their points. Take this case, “Extensive Data Shows Punishing Reach of Racism for Black Boys:

Black boys raised in America, even in the wealthiest families and living in some of the most well-to-do neighborhoods, still earn less in adulthood than white boys with similar backgrounds, according to a sweeping new study that traced the lives of millions of children.

White boys who grow up rich are likely to remain that way. Black boys raised at the top, however, are more likely to become poor than to stay wealthy in their own adult households.

(Oh, look, someone discovered regression to the mean.)

What happens when blue check twitter reports on this piece?

    1. You don’t need an “expert” to tell you that black men might get discriminated against.
    2. How do you become an “expert” in anti-racism? Do you have to pass the implicit bias test? Get a degree in anti-racist studies?
    3. Do you think, for whatever reason, that a guy who gets paid to do anti-racist research might come up with “racism” as an answer to almost any question posed?
    4. “The guy who gets paid to say that racism is the answer said the answer is racism” does not actually prove that racism is the answer, but it is being presented like it does.
    5. Blue check has failed to mention any obvious counters, like:
      a. Mysteriously, this “racism” only affects black men and not black women (this is why we’ve had a black female president but not a black male one, right?)
      b. Regression to the mean is a thing and we can measure it (shortly: The further you are from average for your group on any measure [height, intelligence, income, number of Daleks collected, etc.,] the more likely your kids are to be closer to average than you are. [This is why the kids of Nobel prize winners, while pretty smart on average, are much less likely to win Nobels than their parents.] Since on average blacks make a lot less money than whites, any wealthy black family is significantly further from the average black income than a white family with the same amount of money is from the average white income. Therefore at any high income level, we expect black kids to regress harder toward the black mean than white kids raised at the same level. La Griffe du Lion [a statistics expert] has an article that goes into much more depth and math on regression to the mean and its relevance.)
      c. Crime rates. Black men commit more crime than black women or white men, and not only does prison time cut into employment, but most employers don’t want to employ people who’ve committed a crime. This makes it easier for black women to get jobs and build up wealth than black men. (The article itself does mention that “The sons of black families from the top 1 percent had about the same chance of being incarcerated on a given day as the sons of white families earning $36,000,” but yeah, it’s probably just totally irrational discrimination keeping black men out of jobs.)

“Experts” like this get used to trot a simple, narrative-supporting line that the paper wants to make rather than give any real or uncomfortable analysis of a complex issue. It’s dishonest reporting and contributes to the notion that “expert” doesn’t mean all that much.

Source

Leaded Gas:

Tetraethyllead (aka lead) was added to automobile fuels beginning in the 1920s to raise fuel economy–that is, more miles per gallon. For half a century, automobiles belched brain-damaging lead into the atmosphere, until the Clean Air Act in the 70s forced gas companies to cut back.

Here’s a good article discussing the leaded gas and crime correlation.

Plenty of people knew lead is poisonous–we’ve known that since at least the time of the Romans–so how did it end up in our gas? Well, those nice scientists over at the auto manufacturers reassured us that lead in gasoline was perfectly safe, and then got themselves on a government panel intended to evaluate the safety of leaded gas and came to the same conclusion. Wired has a thorough history:

But fearing that such [anti-leaded gas] measures would spread, … the manufacturing companies demanded that the federal government take over the investigation and develop its own regulations. U.S. President Calvin Coolidge, a Republican and small-government conservative, moved rapidly in favor of the business interests.

… In May 1925, the U.S. Surgeon General called a national tetraethyl lead conference, to be followed by the formation of an investigative task force to study the problem. That same year, Midgley [the inventor of leaded gas] published his first health analysis of TEL, which acknowledged  a minor health risk at most, insisting that the use of lead compounds,”compared with other chemical industries it is neither grave nor inescapable.”

It was obvious in advance that he’d basically written the conclusion of the federal task force. That panel only included selected industry scientists like Midgely. It had no place for Alexander Gettler or Charles Norris [scientists critical of leaded gas] or, in fact, anyone from any city where sales of the gas had been banned, or any agency involved in the producing that first critical analysis of tetraethyl lead.

In January 1926, the public health service released its report which concluded that there was “no danger” posed by adding TEL to gasoline…”no reason to prohibit the sale of leaded gasoline” as long as workers were well protected during the manufacturing process.

The task force did look briefly at risks associated with every day exposure by drivers, automobile attendants, gas station operators, and found that it was minimal. The researchers had indeed found lead residues in dusty corners of garages. In addition,  all the drivers tested showed trace amounts of lead in their blood. But a low level of lead could be tolerated, the scientists announced. After all, none of the test subjects showed the extreme behaviors and breakdowns associated with places like the looney gas building. And the worker problem could be handled with some protective gear.

I’m not sure how many people were killed globally by leaded gas, but Wired notes:

It was some fifty years later – in 1986 – that the United States formally banned lead as a gasoline additive. By that time, according to some estimates, so much lead had been deposited into soils, streets, building surfaces, that an estimated 68 million children would register toxic levels of lead absorption and some 5,000 American adults would die annually of lead-induced heart disease.

The UN estimates that the elimination of lead in gas and paint has added 2.4 trillion, annually, the global economy.

Leaded gas is a good example of a case where many experts did know it was poisonous (as did many non-experts,) but this wasn’t the story the public heard.

Pluto

Yes, this one is silly, but I have relatives who keep bringing it up. “Scientists used to say there are 9 planets, but now they say there are only 8! Scientists change what they think all the time!”

Congratulations, astronomers, they think you lost Pluto. Every single time I try to discuss science with these people, they bring up Pluto. Scientific consensus is meaningless in a world where planets just disappear. “Whoops! We miscounted!”

(No one ever really questioned Pluto’s planetary status before it was changed, but a few die-hards refuse to accept the new designation.)

Scientists weren’t actually wrong about Pluto (“planet” is just a category scientists made up and that they decided to redefine to make it more useful,) but the matter confused people and it seemed like scientific consensus was arbitrary and could change unexpectedly.

Unfortunately, normal people who don’t have close contact with science or scientists often struggle to understand exactly what science is and how it advances. They rely, sporadically, on intermediaries like The History Chanel or pop science journalists to explain it to them, and these guys like to run headlines like “5 things Albert Einstein got Totally Wrong” (haha that Albert, what a dummy, amirite?)

So when you question why people distrust experts like you, Professor Nichols, consider whether the other “experts” they’ve encountered have been trustworthy or even correct, or if they’ve been liars and shills.

An attempt to answer some questions on IQ

I recently received a few IQ-related questions. Now, IQ is not my specialty, so I do not feel particularly adequate for the task, but I’ll do my best. I recommend anyone really interested in the subject read Pumpkin Person’s blog, as he really enjoys talking about IQ all the time.

  1. I wanted to ask if you know any IQ test on the internet that is an equivalent to the reliable tests given by psychologists?

I suppose it depends on what you want the test for. Curiosity? Diagnosis? Personally, I suspect that the average person isn’t going to learn very much from an IQ test that they didn’t already know just from living (similarly, I don’t think you’re going to discover that you’re an introvert or extrovert by taking an online quiz if you didn’t know it already from interacting with people,) but there are cases where people might want to take an IQ test, so let’s get searching.

Pumpkin Person speaks highly of the Wechsler Intelligence Scales (it comes in adult and child versions.) According to Wikipedia:

The Wechsler Adult Intelligence Scale (WAIS) is an IQ test designed to measure intelligence and cognitive ability in adults and older adolescents.[1] The original WAIS (Form I) was published in February 1955 by David Wechsler, as a revision of the Wechsler–Bellevue Intelligence Scale, released in 1939.[2] It is currently in its fourth edition (WAIS-IV) released in 2008 by Pearson, and is the most widely used IQ test, for both adults and older adolescents, in the world.

Since IQ tests excite popular interest but no one really wants to pay $1,000 just to take a test, the internet is littered with “free” tests of questionable quality. For example, WeschlerTest.com offers “free sample tests,” but the bottom of the website notes that, “Disclaimer: This is not an official Wechsler test and is only for entertainment purposes. Any scores derived from it may not accurately reflect the score you would attain on an official Wechsler test.” Here is a similar wesbsite that offers free Stanford-Binet Tests.

I am not personally in a position to judge if these are any good.

It looks like the US military has put its Armed Services Vocational Aptitude Battery online, or at least a practice version. This seems like one of the best free options, because the army is a real organization that’s deeply interested in getting accurate results and the relationship between the ASVAB and other IQ tests is probably well documented. From the website:

The ASVAB is a timed test that measures your skills in a number of different areas. You complete questions that reveal your skills in paragraph comprehension, word knowledge, arithmetic reasoning and mathematics knowledge. These are basic skills that you will need as a member of the U.S. military. The score you receive on the ASVAB is factored into your Armed Forces Qualifying Test (AFQT) score. This score is used to figure out whether you qualify to enlist in the armed services. …

The ASVAB was created in 1968. By 1976, all branches of the military began using this test. In 2002, the test underwent many revisions, but its main goal of gauging a person’s basic skills remained the same. Today, there is a computerized version of the test as well as a written version. The Department of Defense developed this test and it’s taken by students in thousands of schools across the country. It is also given at Military Entrance Processing Stations (MEPS).

Naturally, each branch of the United States armed services wants to enlist the best, most qualified candidates each year. The ASVAB is a tool that helps in the achievement of that purpose. Preparing to take the ASVAB is just one more step in the journey toward your goal of joining the U.S. armed services. …

Disclaimer: The tests on this website are for entertainment purposes only, and may not accurately reflect the scores you would attain on a professionally administered ASVAB test.

The blog Random Critical Analysis gives a thorough rundown of the correlation between ASVAB and IQ scores (they are highly correlated) along with the SAT and ACT.

Additionally, there are a couple of tests linked here in Lipscomb’s Intelligence Course Lab : Classical IQ Test from Psychology Today and IQTest.com.

Drawing a page from Pumpkin Person’s book, I recommend taking several different tests and then comparing results. Use your good judgment about whether a particular test seems reliable–is it covered in ads? Does random guessing get you a score of 148? Did you get a result similar to what you’d expect based on real life experiences?

2. Besides that I wanted to ask you how much social class and IQ are correlated?

A fair amount.

Thanks to Tino Sanandaji

With thanks to Pumpkin Person

Really dumb people are too dumb to commit as much crime as mildly dumb people

IQ by country–red = low; purple – high. Source Wikipedia

 

 

 

 

 

 

 

 

 

 

I do wonder why he made the graph so much bigger than the relevant part
Lifted gratefully from La Griffe Du Lion’s Smart Fraction II article
Oh, there you are, correlation
Lifted gratefully from La Griffe Du Lion’s Smart Fraction II article

When dumb children are born to rich people, they tend to do badly in life and don’t make much money; they subsequently sink in social status. When smart children are born to poor people, they tend to do well in life and rise in social status. Even in societies with strict social classes where moving from class to class is effectively impossible, we should still expect that really dumb people born into wealth will squander it, leading to their impoverishment. Likewise, among the lower classes, we would still expect that smarter low-class people would do better in life than dumber ones.

This is all somewhat built into the entire definition of “IQ” and what people were trying to measure when they created the tests.

3. Basically do traditional upper classes form separate genetic clusters like Gregory Clark claims?

I haven’t read Clark’s book, but I’m sure the pathetic amount of research I can do here would be nothing compared to what he’s amassed.

There are a number of studies on assortative mating and IQ, eg: Spouse similarity for IQ and personality and convergence:

A similar pattern of spousal association for IQ scores and personality traits was found in two British samples from Oxford and Cambridge. There was no indirect evidence from either sample to suggest that convergence occurred during marriage. All observed assortative mating might well be due to initial assortment.

Assortative mating for psychiatric disorders and psychological traits:

This article reviews the literature on assortative mating for psychological traits and psychiatric illness. Assortative mating appears to exist for personality traits, but to a lesser degree than that observed for physical traits, sociodemographic traits, intelligence, and attitudes and values. Concordance between spouses for psychiatric illness has also been consistently reported in numerous studies. This article examines alternative explanations for such observed concordance and discusses the effects of assortative mating on population genetics and the social environment.

Do assortative mating patterns for IQ block upward social mobility?

In the Minnesota Twin Family Study, assortative mating for IQ was greater than .3 in both the 11- and 17-year-old cohorts. Recognizing this, genetic variance in IQ independent of SES was greater with higher parental SES in the 11-year-old cohort. This was not true, however, in the 17-year-old cohort. In both cohorts, people of higher IQ were more likely to have ‘married down’ for IQ than people of lower IQ were to have ‘married up’. This assortative mating pattern would create greater genetic diversity for IQ in people of higher IQ than in people of lower IQ. As IQ is associated with SES, the pattern could be one reason for the observation of greater genetic variance in IQ independent of SES with greater parental SES in several samples. If so, it could block upward social mobility among those already in lower-SES groups. I discuss possible involved mechanisms and social implications.

The role of personality and intelligence in assortative mating:

Assortative mating is the individuals’ tendency to mate with those who are similar to them in some variables, at a higher rate than would be expected from random. This study aims to provide empirical evidence of assortative mating through the Big Five model of personality and two measures of intelligence using Spanish samples. The sample consisted of 244 Spanish couples. It was divided into two groups according to relationship time. The effect of age, educational level and socioeconomic status was controlled. The results showed strong assortative mating for intelligence and moderate for personality. The strongest correlations for Personality were found in Openness, Agreeableness and Conscientiousness.

Assortative mating for IQ and personality due to propinquity and personal preference:

The role of personal preference as an active process in mate selection is contrasted with the more passive results of limitations of available mates due to social, educational, and geographical propinquity. The role of personal preference estimated after removing the effects of variables representing propinquity was still significant for IQ and Eysenck’s extraversion-introversion and inconsistency (lie) scales, even though small.

Related: Heritability estimates versus large environmental effects: the IQ paradox resolved:

Some argue that the high heritability of IQ renders purely environmental explanations for large IQ differences between groups implausible. Yet, large environmentally induced IQ gains between generations suggest an important role for environment in shaping IQ. The authors present a formal model of the process determining IQ in which people’s IQs are affected by both environment and genes, but in which their environments are matched to their IQs. The authors show how such a model allows very large effects for environment, even incorporating the highest estimates of heritability. Besides resolving the paradox, the authors show that the model can account for a number of other phenomena, some of which are anomalous when viewed from the standard perspective.

 

4. Are upper class people genetically more intelligent? Or is there an effect of regression to the mean and all classes have about equal chances to spawn high IQ people?”

Stephen Hsu has a relevant post on the subject: Assortative mating, regression and all that: offspring IQ vs parental midpoint:

…James Lee, a real expert in the field, sent me a current best estimate for the probability distribution of offspring IQ as a function of parental midpoint (average between the parents’ IQs). James is finishing his Ph.D. at Harvard under Steve Pinker — you might have seen his review of R. Nesbitt’s book Intelligence and how to get it: Why schools and cultures count.

The results are stated further below. Once you plug in the numbers, you get (roughly) the following:

Assuming parental midpoint of n SD above the population average, the kids’ IQ will be normally distributed about a mean which is around +.6n with residual SD of about 12 points. (The .6 could actually be anywhere in the range (.5, .7), but the SD doesn’t vary much from choice of empirical inputs.)…

Read Hsu’s post for the rest of the details.

In short, while regression to the mean works for everyone, different people regress to different means depending on how smart their particular ancestors were. For example, if two people of IQ 100 have a kid with an IQ of 140, (Kid A) and two people of IQ 120 have a kid of IQ 140, (Kid B), Kid A’s own kids are likely to regress toward 100, while Kid B’s kids are likely to regress toward 120.

We can look at the effects of parental SES on SAT Scores and the like:

SAT scores by race and parental income

Personally, I know plenty of extremely intelligent people who come from low-SES backgrounds, but few of them ended up low-SES. Overall, I’d expect highly intelligent people to move up in status and less intelligent people to move down over time, with the upper class thus sort of “collecting” high-IQ people, but there are obviously regional and cultural effects that may make it inappropriate to compare across groups.

Hope that has been useful.