Is meritocracy overrated?

Great Debates
101

There is a science to testing. It’s called psychometrics. I doubt any test is perfect but over a large sample size it is a fairly reliable measurement tool.

There is about as much research supporting the validity of testing as there is supporting the validity of anthropomophic climate change. Denying the value of testing puts you in the same science denial category as global warming denialism and anti-vaxxer. There is almost no controversy around the notion that tests measure something meaningful.

102

Interesting paper, it says that standardized tests are good predictors for student “performance” in graduate school. There was a high correlation with standardized test scores and GPA’s determined by test scores, and qualifying exams, and licensing exams. That is truly shocking, to find that high test scores in one kind of test is correlated with high scores on other tests. Who woulda thunk it?

Also shocking is that “performance” that isn’t directly related to test scores shows a much lower correlation, such as in citation count, research productivity and degree completion.

103

Ultimately, the idea of a standardized test is that EVERYONE gets the same test, and the results are stacked up and ranked.

Again, it’s not the test’s fault, or some sort of fundamental problem with testing if some groups are underprepared. Case in point- the bar exam. Some schools (Baylor University, UT-Austin) have an extremely high first-time passage rate of around 92%, while others are not so good, like St. Mary’s in San Antonio, or Texas Southern in Houston with 64.7% and 43.6% respectively. Clearly the students from the first two were either better prepared, or were selected to be better students than the other two, or most probably both.

Is it the fault of the exam that TSU can’t manage to have more than half their graduating law classes pass the bar on the first try? Is the bar being unreasonable? I doubt it.

104

That’s very reasonable. The problem is it takes money.
I don’t think we can force people to read to their kids, or have books in their houses. But we can fund universal preschools so that kids whose parents don’t read to them get read to. That’s not going to create equality of outcomes, but it will help all kids to be closer in experiences when they start school.
You can think of it as handicapping in racing or in golf.
We’re never going to be able to offer true equality of opportunity. I know someone who is a flat earther, and who thinks scientists who believe in evolution are possessed by demons. She is home schooling her grandkids. I can only think that her grandkids are not going to be much competition to mine in the job market. Unless the job is fundamentalist preacher, that is.

105

Consider this thought experiment.

Supposing I administer a test on a pass/fail basis to a hundred people. Let’s say… 70 people pass, and 30 people fail according to “some metric.” Lets then say that I run all 100 people through a month-long certification program culminating in a series of exams and practical demonstrations, and it turns out that of the ones who passed the test, only 1 person failed the certification program. Of the ones who failed… 2.

And let’s say I repeat this experiment each month for a year, with different people, and the results are all generally the same. One in seventy “test passers” will succeed in the program, and one in fifteen “test failers” will fail the program.

And then it turns out that these “program suceeders” are so much more productive than the handful of failures—they really learned something—that I, as their employer, decide to continue this program indefinitely, and will pay the “program succeeders” each an extra ten thousand dollars a year for life compared to the failers.

And after a couple years, I decide that having… almost five times the rate of program failure for the test, warrants just preemptively denying entry into the program to ALL initial test failures. Even though the bulk of them may yet have passed my month-long program and thus been eligible for the bump in wages, they will lose the chance to do so. And oh by the way, this decision of mine—which I made for race neutral reasons—will mostly just affect minorities. Not that that was he intent, it’s just how the numbers work out. Oh well…

I’m not going to conclude by asking you a question. The point was just to consider that hypothetical.

106

That should read: "One in seventy “test passers” will succeed in the program, and one in fifteen “test failers” will fail the program.

:flushed:

107

What if you only have enough room in your program for 70 people? If you have 70 seats then the metric you are using is pretty good isn’t it? If there were an infinite number of seats then you are probably using the wrong “metric” to filter out candidates. The cost benefit shifts.

But in real life there is one minority group that started outperforming the white candidates. And oh by the way, this is when you start to wonder if the test has any validity.

I don’t think your hypothetical is particularly illuminating. It’s basically a hypothetical anecdote.

108

Can you point to the part that says that these tests are predictive of GPA’s determined by test scores? AFAICT the blurb only mentions two types of GPA, 1st year GPA and Graduate GPA. nothing about test driven GPA.

[quote]Also shocking is that “performance” that isn’t directly related to test scores shows a much lower correlation, such as in citation count, research productivity and degree completion.
[/quote]

“For all tests across all relevant
success measures, standardized test scores
are positively related to subsequent meas-
ures of student success [see chart, table S1,
and supporting online material (SOM) text].”

I did notice the very small correlation between:SAT scores and degree completion and I think this might be because almost everybody who does not have some life event ends up graduating. I think this might also be true of medical and business schools.

109

A fair question. What if there was a way to fill those 70 slots all the same, and still only get 1 failure out of 70, and yet increase the diversity (not just in terms of minority representation, but in the broadest sense) of your workforce in ways that may be of benefit to you?

But, hey, that’s just another hypothetical, and would require you (or, more appropriately, people involved in college admissions) to be open to not just keep falling back on “the way we’ve always done it” with testing as the go-to measure of an applicant, without regard to how testing may not uniformly reflect ability across various demographic groups, and not just according to race or ethnicity.

Ah, yes, the model minority. Out of curiosity, apart from being a “minority group” and “outperforming white candidates” on average, are there any other broad brush differences between this one minority group and, say, a minority group consisting in large part, though not in total, of descendants of people brought to this country forcibly, without wealth or opportunity to accumulate wealth (let alone pass that wealth on) for a significant portion of their family history? Like, did this particular model minority face its own kind of discrimination in, say, immigration, which, perversely given the context here of comparing one minority group to another, resulted in greatly restricted immigration for a period of time, followed by only allowing a limited number of immigrants according to regulations that may have prioritized those with means or marketable skills/education?

Is that a hypothetical?

110

OK so now that we understand that tests matter. As long as you are using an objective standard and getting the same results, then all things being equal, more diversity > than less diversity, generally speaking.

I think you would have to describe this method more specifically, because right now you might as well be talking about a fire breathing unicorn.

No but neither did whites.

How is this minority group outperforming everyone, including whites in this deeply racist country? Are the racists so determined to make black people look bad that they are throwing white kids under the bus to make a test that is particularly easy for asian kids?

I’m pretty sure I was comparing this minority group to whites.

I’d like to leave the descendants of american slaves and american indians out of this, I am not comparing asians to them. I am comparing asians to whites.

You realize that coming from a country with a culture that values education means that more people are educated, right? Are you saying that these asian immigrants are more privileged than the whites that are already here?

Asians have the highest poverty rate in NYC. And yet the city keeps crafting race balancing schemes that will put these kids at a disadvantage. Why do they do that?

No, this is an example of the model minority myth. You THINK you know about asians but all you really know are stereotypes and you use these poorly understood stereotypes to dismiss the concerns expressed by asians about bias and discrimination against asians.

111

Except you’re doing it as a way to then insist that racism—which involves more than just whites and Asians—doesn’t exist at any stage because “Asians do well enough” or words to that effect. That’s the problem with the model minority argument. It is perverse because it takes a minority group with demographics that tend to be more skewed towards wealth and education by virtue of discriminatory immigration policies in the past to minority groups that arrived in or pre-existed this country under different conditions that are tend to affect their access to higher education.

If you think racism is limited to overt/intentionally racist acts by individuals, then you have a flawed understanding of what racism is and how it manifests.

112

That’s clearly incorrect. In fact I am pretty sure that I have explicitly said that racism and discrimination do exist. Including racism against asians. In fact I spend more than a small amount of time on this board arguing with people who think that racism against asians either doesn’t exist or doesn’t need to be addressed because “Asians are doing well enough” or words to that effect.

I can’t tell you how many times we have seen the argument that competitive colleges can’t be discriminating against asians because asians are over-represented at these schools.

Do you think Vietnamese refugees are also self selected in this way? What do you think the average wealth, education, income of the descendants of vietnamese refugees is today?

Where do I say these things that you think I say?
I think that some people can be racist against a group without even realizing it. I think that there is institutional racism against many minorities in many forms. Some of those minorities are asians and the racism takes the form of trivializing the efforts and sacrifices made by asians.
Is it possible that my opinion of what constitutes racism can reasonably differ from yours?
Is it possible that I have a very good idea of what racism is and you do not?

113

And if there’s evidence that minority students are discriminated against in the selection process for gifted programs? According to a 2016 study by Vanderbilt University, “Black students are 66 percent less likely to referred to gifted programs in math and reading their their white classmates.” Most schools don’t test all children to see if they’re a fit for the gifted program and those that don’t tend to discriminate.

Many magnet schools face similar problems with whites and Asians dominating the student body and other minority students being left out during the selection process.

114

What we have is a group of people who are capable of satisfying degree requirements being separated by a test that is largely unrelated to their ability to satisfy degree requirements. Is it “fair” to separate people by a measure of “quality” that isn’t related to their ability to get a degree in their chosen field?

Test performance is unsurprisingly related to test performance, whether it’s labeled a “test” or a “quiz” or “homework” or “group project”, it’s a standardized nugget of work whose purpose is to judge the student.

115

I may have missed it, but how is the term “Asian” operationally defined in this case?

116

I happen to agree that you never will, which is why the equality of opportunity people who seem to think that that correlates to equality of outcomes baffles me.

People, and cultures vary by a great amount so those outcomes will never BE equal, they likely aren’t even close.

117

It only baffles you because you seem to think that equality of outcome means that any way you slice a cross section, you would get identical results.

That is not the case.

Sure, there may be some cultural reasons as to why some demographics prefer nursing to being a pilot, or being an engineer vs being a businessman.

But there are no cultural reasons as to why one would prefer poverty over not being impoverished, so if you see disparities in the amount of poverty, then a very likely cause is disparities in opportunities.

That is the equality of outcome that we talk about. Preferably as little poverty as possible, even none, if that is achievable.

But, if you see one demographic that tends toward poverty more than another, it is disingenuous to blame that on culture, especially when inequalities of opportunity are obvious to see.

It baffles me that anyone would expect that inequalities in opportunity would not result in inequalities in outcome.

Take the stock market, for example. Give a bunch of people the same amount of money to invest, and they will, by using different strategies and some being luckier than others, not all end up with the same portfolio after 30 years. Some may get ultra-wealthy, and some may bottom out.

However, if you give some people more money to invest in others, then it would be preposterous to expect, on average, those who had less to invest to have as much as those who had more. There may be some who had less to invest who end up with more than those who had more to start, but those will be the exceptions. And to focus on the exceptions is to ignore the bulk of the situation.

118

Poverty. Education.

The two closely correlate. No focus on education, more poverty. Nothing else needs to be included.

119

More poverty, less ability to focus on education.

120

Yes it’s cyclical. What to be done if you aren’t going to put kids in a better situation? We just discussed a few options but what hasn’t been established is that even if poverty was lessened that the differing cultures would put the same emphasis on education.

There are some poverty stricken demographics that Do emphasize education and they trend better