Standard Tests And Bias

[Loren Pechtel]

Pointing back to your post doesn't make it so.

The study you presented doesn't consider the big fish in the little pond problem.

What exactly is this "big fish in the little pond problem"?

A high GPA in a shitty school doesn't mean much.

 

[Loren Pechtel]

The problem here is that three things are inherited:

Intelligence
Money
Attitudes

Your study is comparing the first two without considering the third.

The problem here is the third is essentially inseparable from the others. A poverty trap can be depressing, Intelligence a feature of both genes and environment, The encouragement and support of family wealth enables self esteem, confidence, grooming, etc. Your objections are overly simplistic.

You fail to note the issue--even if you play mega robin hood you only change one of the three. The inequality you so hate remains.

 

[Loren Pechtel]

the solutions:

There should be more testing, not less.

The tests themselves should be tested, or all the test items should be scrutinized to ensure that they perform their (culturally neutral) function of objectively measuring each student's mastery of the subject matter. And those creating the tests have to be corrected for their bad decisions in shaping the test items.

For the big tests there's no need to go overboard. If you think the questions have a problem on characteristic X then you record the X of every test taker. For each test question you draw a graph--the probability that someone got the right answer vs their overall score, scale it to a 0 to 1 box. There should be a line going from 0, 0 to 1, 1. On easy questions the line will be bowed up, on hard questions the line will be bowed down. If you get anything other than a straight or bowed line you have a bad question, throw it out.

Now repeat this process, for each option of X generate the same graph for test takers that are that option. The graph should look just (within the fuzziness inherent to statistical data) like the one you drew for the whole group. If it's not you have a bad question.

The only way a bad question can evade this test is if it's badness is so small that it lies within the error bounds and you have so many test takers those error bounds will be narrow.

Education opportunities ideally should be open to all seeking to enter, with no limits to the number of students admitted, so there's no need to exclude anyone, by testing or any other barrier. And all the students in a program should be tested and advanced further according to their performance on the tests, with no sympathy to anyone performing poorly, other than to drive them to improve their performance on future tests.

Only if educational opportunities exist at all levels. The narrower the range of abilities of the students in a class the better off the students are.

 

[Loren Pechtel]

You don’t need wealth to be a good parent. It costs nothing to read to your children and make sure they did their homework.

Yup. Reading matters, but whether they are owned books or library books makes no difference to the outcome.

 

[DBT]

The problem here is that three things are inherited:

Intelligence
Money
Attitudes

Your study is comparing the first two without considering the third.

The problem here is the third is essentially inseparable from the others. A poverty trap can be depressing, Intelligence a feature of both genes and environment, The encouragement and support of family wealth enables self esteem, confidence, grooming, etc. Your objections are overly simplistic.

You fail to note the issue--even if you play mega robin hood you only change one of the three. The inequality you so hate remains.

I made no mention of hate. The issue, as pointed out numerous times, too many times, is a gross imbalance in power between the very rich and the rest in a descending order of magnitude....where the top enjoys every advantage while those at the bottom just struggle to make ends meet.

 

[DBT]

You don’t need wealth to be a good parent. It costs nothing to read to your children and make sure they did their homework.

Yup. Reading matters, but whether they are owned books or library books makes no difference to the outcome.

Reading alone doesn't pay university fees or provide an entry into an old boy network that ensures a lucrative position upon graduation even without top marks. Great family wealth does that.

 

[Trausti]

My point is that wealth is not the issue. Working class people certainly can and should give attention to their children.

Are you saying they don't? That because they don't, that is the cause of the disparity between the rich and the poor?

Well, you can't ignore genetic confounding. Here's an example. In a legal case where a delivering doctor seriously harmed a child during delivery - such that the child has severe motor and cognitive deficits - that child can make a claim for future income loss. But how do you determine probable future income loss for someone with no work history; who has never worked and never will? You look at the parents. The probably is that a child will do as well or slightly better than the parents. If the parents are professionals, it is likely the child would have been a professional. If the parents are low-income, the same is likely for the child. Once a geneticist was rather blunt to me: if the parents are dim than it's not surprising for the child to be dim.

 

[DBT]

My point is that wealth is not the issue. Working class people certainly can and should give attention to their children.

Are you saying they don't? That because they don't, that is the cause of the disparity between the rich and the poor?

Well, you can't ignore genetic confounding. Here's an example. In a legal case where a delivering doctor seriously harmed a child during delivery - such that the child has severe motor and cognitive deficits - that child can make a claim for future income loss. But how do you determine probable future income loss for someone with no work history; who has never worked and never will? You look at the parents. The probably is that a child will do as well or slightly better than the parents. If the parents are professionals, it is likely the child would have been a professional. If the parents are low-income, the same is likely for the child. Once a geneticist was rather blunt to me: if the parents are dim than it's not surprising for the child to be dim.

Your example represents a very small minority. Most people are not that dim, they are able to do productive work.

What about the working poor who work full time, long hours doing productive work for pittance rates of pay? Not because they are not useful to society, the business, not because their work is not essential, but simply because as individuals in the position they are in, doing work that has no prestige or status attached, they lack the power to negotiate better pay and conditions?

 

[Jayjay]

Well, you can't ignore genetic confounding. Here's an example. In a legal case where a delivering doctor seriously harmed a child during delivery - such that the child has severe motor and cognitive deficits - that child can make a claim for future income loss. But how do you determine probable future income loss for someone with no work history; who has never worked and never will? You look at the parents. The probably is that a child will do as well or slightly better than the parents. If the parents are professionals, it is likely the child would have been a professional. If the parents are low-income, the same is likely for the child. Once a geneticist was rather blunt to me: if the parents are dim than it's not surprising for the child to be dim.

Your example represents a very small minority. Most people are not that dim, they are able to do productive work.

What about the working poor who work full time, long hours doing productive work for pittance rates of pay? Not because they are not useful to society, the business, not because their work is not essential, but simply because as individuals in the position they are in, doing work that has no prestige or status attached, they lack the power to negotiate better pay and conditions?

I think you and Trausti are talking past each other. He wasn't using the person who can't do any productive work as an example of dimness being inherited; that was due to the doctor's malpractice. For all we know his parents were lawyers or astronauts (though I don't know if this is a thought experiment or a real legal case).

On the other hand Trausti seems to be committing a circular argument. He says wealth doesn't matter; but just because there is a correlation between the parents' income level and the child's, doesn't imply that it's caused only by genetics. It might as well be due to wealth being also a factor.

 

[Trausti]

Well, you can't ignore genetic confounding. Here's an example. In a legal case where a delivering doctor seriously harmed a child during delivery - such that the child has severe motor and cognitive deficits - that child can make a claim for future income loss. But how do you determine probable future income loss for someone with no work history; who has never worked and never will? You look at the parents. The probably is that a child will do as well or slightly better than the parents. If the parents are professionals, it is likely the child would have been a professional. If the parents are low-income, the same is likely for the child. Once a geneticist was rather blunt to me: if the parents are dim than it's not surprising for the child to be dim.

Your example represents a very small minority. Most people are not that dim, they are able to do productive work.

What about the working poor who work full time, long hours doing productive work for pittance rates of pay? Not because they are not useful to society, the business, not because their work is not essential, but simply because as individuals in the position they are in, doing work that has no prestige or status attached, they lack the power to negotiate better pay and conditions?

Why would any of that stop them from reading to their children? There are plenty of examples of one generation working hard but poor to allow the second generation to succeed. The most obvious is second-generation Asian-Americans. Many first-generation arrive poor but their children excel.

 

[Trausti]

On the other hand Trausti seems to be committing a circular argument. He says wealth doesn't matter; but just because there is a correlation between the parents' income level and the child's, doesn't imply that it's caused only by genetics. It might as well be due to wealth being also a factor.

Well, sure, genetics isn't everything. But if you read to your children; encourage them; teach them all you can; you're giving them all and more that wealth supposedly provides.

 

[AthenaAwakened]

After skimming through this thread, I'm gonna go out on a limb here and suggest that I am among a minority of people here who have ever sat behind the big desk with many pencils.

So here ya go. Feel free to stop reading if you have heard this.

Teachers test to get some kind of idea about what a student does and doesn't know. A teacher, if left to her own devices, then develops a plan to further the students education by building on what the student does know and by further explaining and developing what he doesn't yet fully understand. That is how it is supposed to work and it does work if you are an A list actor in a Hollywood teen struggle movie.

In the real world, testing is a way to make Pearson rich, fire/demote teachers, and have the state or a private education company take over a public school. It is a way to beat students into submission through a regimen of teaching the test instead of freeing young minds through teaching to the student.

It doesn't really matter what the test says or if the questions are bias since even a perfectly balanced test will be weaponized against actually teaching children to think for themselves.

 

[DBT]

Well, you can't ignore genetic confounding. Here's an example. In a legal case where a delivering doctor seriously harmed a child during delivery - such that the child has severe motor and cognitive deficits - that child can make a claim for future income loss. But how do you determine probable future income loss for someone with no work history; who has never worked and never will? You look at the parents. The probably is that a child will do as well or slightly better than the parents. If the parents are professionals, it is likely the child would have been a professional. If the parents are low-income, the same is likely for the child. Once a geneticist was rather blunt to me: if the parents are dim than it's not surprising for the child to be dim.

Your example represents a very small minority. Most people are not that dim, they are able to do productive work.

What about the working poor who work full time, long hours doing productive work for pittance rates of pay? Not because they are not useful to society, the business, not because their work is not essential, but simply because as individuals in the position they are in, doing work that has no prestige or status attached, they lack the power to negotiate better pay and conditions?

I think you and Trausti are talking past each other. He wasn't using the person who can't do any productive work as an example of dimness being inherited; that was due to the doctor's malpractice. For all we know his parents were lawyers or astronauts (though I don't know if this is a thought experiment or a real legal case).

On the other hand Trausti seems to be committing a circular argument. He says wealth doesn't matter; but just because there is a correlation between the parents' income level and the child's, doesn't imply that it's caused only by genetics. It might as well be due to wealth being also a factor.

I thought the example was irrelevant, so I basically ignored it.

 

[DBT]

Well, you can't ignore genetic confounding. Here's an example. In a legal case where a delivering doctor seriously harmed a child during delivery - such that the child has severe motor and cognitive deficits - that child can make a claim for future income loss. But how do you determine probable future income loss for someone with no work history; who has never worked and never will? You look at the parents. The probably is that a child will do as well or slightly better than the parents. If the parents are professionals, it is likely the child would have been a professional. If the parents are low-income, the same is likely for the child. Once a geneticist was rather blunt to me: if the parents are dim than it's not surprising for the child to be dim.

Your example represents a very small minority. Most people are not that dim, they are able to do productive work.

What about the working poor who work full time, long hours doing productive work for pittance rates of pay? Not because they are not useful to society, the business, not because their work is not essential, but simply because as individuals in the position they are in, doing work that has no prestige or status attached, they lack the power to negotiate better pay and conditions?

Why would any of that stop them from reading to their children? There are plenty of examples of one generation working hard but poor to allow the second generation to succeed. The most obvious is second-generation Asian-Americans. Many first-generation arrive poor but their children excel.

Not everyone is able to excel. Many asians have strong family ties and a culture of support for each other.

Not everyone is fortunate enough to be in that position...not even all Asians. If society at large adopted the attitude of support and nurture that many asian families have, perhaps we would have far fewer people in positions of poverty or struggling on low incomes while working productively

Which still doesn't address the power imbalance that prevents otherwise productive workers from securing a better deal for themselves. Not everyone can be doctors or lawyers.

 

[Derec]

the problem is that this is impossible - the only thing a test can ever measure is your mastery of taking the test, but it's almost completely unrelated to mastery of the subject that the test is about.

How can you claim that test performance is "almost completely unrelated to mastery of the subject" if you deny that mastery of a subject can be objectively measured?

the problem is that people are stupid enough to think that one's ability to master a test is the same thing as one's ability to master the thing the test is about, and that has lead to a huge number of problems in US culture when it comes to notions of intelligence and career preparedness.

While certainly not perfect, a well-designed test is closely relating the two.
Otherwise, how would you determine somebody's mastery of the subject? Just assume everybody has the same mastery just for having gone through a class? Which would make it an academic version of the "participation trophies" in youth sports.

 

[Derec]

The authors statistically control for the school effect,

Which means that it is something that had to be controlled for. Which means it has an effect. An admissions office that gets applicants from 100s of high schools will not be able to perform those controls.

But even if we assume that GPA alone is a better predictor than SAT alone, that does not mean that GPA+SAT is not the best predictor out of the three.
And in fact, the study posted by Axulus shows just that: GPA+SAT is better than either one alone.

Which means SAT is very much important despite the ideological attempts to ax it.

 

[Derec]

It pays to be born rich: Children with wealthy parents are MORE likely to be successful than intellectually gifted kids born to low-income families

Oh, so NOW Daily Fail is a good source to quote. LMAO!

 

[Derec]

English Professors in general tend to be pretty culturally biased when it comes to an idea of which "English" is the correct "English".

Standard English is the correct English in the context of education and testing. That is not "cultural bias".

Further, as long as it includes "history of white people as taught by white people", it will continue to transmit such blatant biases as you point out

So nothing about white people should be on SAT? What racist claptrap!

 

[Derec]

Ok let me rephrase. The reading section (IIRC) had some materials centered on history. It also tended to use words myself and my homies would never use in our entire lives as well as I've never heard any white people use.

As you say, those "SAT words" are not part of the everyday vocabulary for anyone. So where's the bias?

This was somewhere between the late 90's early 2000's and based purely on my memory doused with irish whiskey. I'll take my exit.

Well if you took the SATs doused on Irish whiskey that may explain you not doing too well.
On the other hand, isn't you drinking Irish whiskey cultural appropriation?

 

[Derec]

Consider the word "ottoman," for example. I am not claiming the word is on the test or has been, but kids that have families who own furniture sets have a greater chance to be exposed to such things. It's just an example of an entire class of words. Material things.

If a word is common enough to be used on a joke in a sitcom, can it be considered an obscure word?

It's also true that the makers of the SAT are more cognizant of this type of bias than in the past and so actively try to remove this bias. Now. Historically, no. But even historically, there's been a screaming about SATs not being biased in this way...when they clearly were.

I am still not convinced words like "ottoman" are an example of bias.