disparities in testing the gifted

[boongirl]

Because of popularity, I have decided to start a thread about the disparities in testing the gifted. This could also eventually lead to a discussion about testing young children but the primary reason I am starting this is to discuss the issue of race as it pertains to IQ testing.

Explaining Group Differences
In status-related research with racial and ethnic groups, the issues of ethics and human values are extremely important and controversial (Stanfield, 1993). Value-neutral methods of data collection and interpretation are critical to ensuring that research findings promote an accurate, not stereotyped, view of racial and ethnic groups.

In race and ethnicity research, cultural standards of data generalization are typically based upon universal statements reflecting Eurocentric normative and scientific principles. Unfortunately, such an approach often assumes that concepts or standards, such as indicators of achievement, socialization, development, or performance, transcend cultural barriers. But they may, in fact, differ across cultural groups. Furthermore, some indicators that are determined to be "problematic" in one culture may actually have a positive effect on behavior in another culture. Knowledge of such cultural differences can help researchers avoid many of the procedural pitfalls that can result in stereotyping of racial and ethnic groups (Obiakor & Utley, 1997).

from Addressing Diversity in Special Education Research

The research indicates that there are disproportionate numbers of minority children in gifted groups. This is believed to be the results of the IQ tests not being reliable and valid across multiple cultures and ethnicities. The study cited above is one of many you can find by searching on Hoagies website or doing a general search on Google. It is a very important topic that is of great concern to educators, particularly those who work with a diverse group of students. If there are racial biases in all kinds of standardized testing, not just that used for IQ, then how do you ascertain a realistic picture of a child's capabilities? If there are Eurocentric tendencies in standardized tests, is it fair to give these tests to children who are not from families of predmoninantly European background? In a given school district, should the percentage of children of a given race or ethnicity in any given sub-group be similar to the group as a whole? (In other words, for example, if the district is 32% black, should the gifted group also be 32% black?) These are questions that educators around the country are asking and researching in an effort to leave no child behind and also to help each child meet his or her potential to the best of their ability. If there are tests that are inherently unfair to those who are not of European heritage, then how do you find tests that are not? That one is in process right now in, most likely, nearly every school district in the country. It is a serious issue in urban education circles.


So, what is a parent to do if they suspect their child to be gifted but the test did not show so? What if they suspect the test to be biased against their child due to race or ethnicity? What can they do? Certainly, no one is suggesting that being black or other is what is making the tests unreliable. Educators refer to the tests as unreliable because we know that there is a certain portion of the population that may (not will) be testing lower than expected due to the inherent bias of the test. The tests are more reliable for a middle class white kid because the test was created with a middle class white bias in mind. No one intended to do this. It happened because the authors of the tests are predominantly white and middle or upper class themselves and they wrote the test from their perspective. They did this before they understood what they were doing. Now that people are aware of this, there is an effort to find other testing materials and new tests. For example, a child psych can use a different battery of tests than an educator can and can get a more clear and varied picture of the child's abilities. This may be a less biased approach.

For gifted programs in the public schools, most of the children are identified using a group standardized test. There are many to choose from. In the districts where I have taught, the CogAt (Cognitive Abilities Test) was the most widely used. There are other group standardized tests and school districts tend to use these because there is no federal money to test gifted kids as there is to test special ed kids. So, testing comes out of their own pockets and it is cheaper to test a whole group all at once. There are some districts who use individual test. And, there are child psychologists who use an array of individual tests. Because they use an array of tests, child psychs can usually get a clearer and less biased view of a child's intelligence.

So what is a parent to do? What are the options? In my experience as a teacher of gifted kids in an urban school district, parents can push to have their child nominated to the gifted program, can ask for a retest, or they can seek a private testing service where a child psych will use a broader array of tests to ascertain intelligence.

[eilonwy]

I'm surprised that anyone is even questioning this idea; I thought that it was common knowledge that IQ tests were racially biased.

For the record: I'm not white, and I still managed to score very, very well on IQ tests. At the time, however, I pointed out aspects of the test which were biased, not only racially but temporally. Most of the IQ tests used today haven't had a major overhaul since the 20's or 30's. How many of us have turpentine in our homes today, for example? When I was tested at 12, I'd never even seen a can of turpentine in real life, I'd only read about it and knew that it was a solvent used for painting. The question "Where does turpentine come from?" is hardly relevant to the lives of most American children. Seventy or eighty years ago, sure, but in the 90's? Heck no!

Aside from the biased nature of the tests themselves, we have the folks administering the tests to contend with. The first time my IQ was tested I was five years old, getting ready to enter a new school in first grade. They wanted to hold me back a year despite the fact that I'd finished kindergarten with flying colors (long, messy story); My mother refused and insisted that they test me. The psychologist allowed the test to proceed to a point and then just stopped asking questions and told my mother to take me home. Apparently, my IQ was 129-- one point shy of the gifted program. Why? Well, I was tested in August, so I had a very dark tan. I also looked hispanic to the testers, they thought I was Puerto Rican, and there were no Puerto Rican kids in the gifted program. :

The high school I attended was more than 50% hispanic even ten years ago when I was there. Wanna guess how many hispanic kids were in my chemistry classes? How many took A.P. Chem a year or two later? *Two*. Two Puerto Rican girls, one Asian boy, four Asian girls, and me. That's it, as far as minority representation in *four* full secetions of Chemistry. In A.P. Chemistry, it was just me and one of the aforementioned hispanic girls. No black kids, none at all. It's not logical to assume that there were no other gifted children among the hispanic students; it just doesn't make sense. It's not fair to assume that these kids came from families that didn't value education, either. The trouble is, they were in a school district where children of color were considered high risk trouble makers from the moment they first walked through the door. The most highly gifted students ended up dropping out, bored out of their skulls and finding unique (and often illegal) ways to make ends meet. School was a waste of time not because they were too stupid to handle it, but because the school was too stupid to recognize that they needed a different sort of program, one which was more open to accomodating the needs of gifted students.

In public schools, children are referred for testing most often by teachers. If the teacher has already decided that it's impossible for a child of color to be gifted, all of those kids are screwed. My first grade teacher used to routinely say (in front of me) things like, "She really belongs in pre-first, you can have her if you like," to the other first grade teachers. I don't think I ever heard the word, "stupid" from her, but I certainly heard, "lazy," "clumsy," and "just plain bad" on a daily basis. All this, and I was one of the lucky ones, one of the few with such "typical" giftedness that I was actually able to score well on a (fairly administered) IQ test, despite it's shortcomings. Yes, it's possible that my score was affected by the biases, but the ceiling of the test wasn't high enough for us to make that distinction and it ended up being a non-issue for me. Again, I was one of the lucky ones in that respect, so I know that it does happen; I just don't expect it to happen all the time.

[Roar]

I agree this is an important topic of discussion.

I'm bothered though by the generalities I'm hearing tossed out. Because testing has served, or will serve in the future, an important role in helping meet the needs of many of the children with mothers posting here, it makes sense to me to be accurate in our discussion. Broadly declaring the tests can't test minority children without evidence of that contention is problematic in my opinion. If you have up to date research on racial bias in contemporary IQ tests I'd love to see it.

Originally Posted by boongirl

Value-neutral methods of data collection and interpretation are critical to ensuring that research findings promote an accurate, not stereotyped, view of racial and ethnic groups.

Is it your suggestion that the people designing IQ tests are unaware of these concerns or that they are just disregarding them or that race is such a compelling factor that it is impossible to write an IQ test that isn't racially biased? Is it your belief that as a teacher you have access to information from Hoagie's that people who spend years studying testing design somehow don't have?

Originally Posted by boongirl

The research indicates that there are disproportionate numbers of minority children in gifted groups. This is believed to be the results of the IQ tests not being reliable and valid across multiple cultures and ethnicities.

Here's what we know... There is a strong relationship between race and class in this country. Minority kids are much more likely to be poor. More likely to have low quality child care before they begin school. More likely to attend underfunded schools. More likely to be born premature. More likely to be raised by a parent who hasn't graduated from college. More likely to have poor nutrition. More likely to have lead poisoning. I can see it would be faster and easier to blame the IQ tests than to tackle this long list of ways in which poor kids in this country get screwed over, but I don't think it would be accurate or really address the problems. George W. Bush tells us that the answer to our failing schools is to test, test, test and this just seems like a variation on that theme for me.

Originally Posted by boongirl

If there are racial biases in all kinds of standardized testing, not just that used for IQ, then how do you ascertain a realistic picture of a child's capabilities? If there are Eurocentric tendencies in standardized tests, is it fair to give these tests to children who are not from families of predmoninantly European background?

All ifs you haven't demonstrated.

Originally Posted by boongirl

The tests are more reliable for a middle class white kid because the test was created with a middle class white bias in mind. No one intended to do this. It happened because the authors of the tests are predominantly white and middle or upper class themselves and they wrote the test from their perspective. They did this before they understood what they were doing. Now that people are aware of this, there is an effort to find other testing materials and new tests. For example, a child psych can use a different battery of tests than an educator can and can get a more clear and varied picture of the child's abilities. This may be a less biased approach.

In the sticky thread you said that you were referring to "classic" IQ tests like the WISC and the Stanford Binet. And, now you are suggesting there are "different tests" that aren't biased. What are the "different tests" you are referring to? Who created them? In particular I don't understnad your suggestion here that they created these tests before they knew about bias. The IQ tests most often being used weren't created in the 1950s or something. The SBV for example came out in the last year or two. These tests were studying on a diverse sample population and bias is a strong concern for test designers.

Originally Posted by boongirl

For gifted programs in the public schools, most of the children are identified using a group standardized test. There are many to choose from. In the districts where I have taught, the CogAt (Cognitive Abilities Test) was the most widely used. There are other group standardized tests and school districts tend to use these because there is no federal money to test gifted kids as there is to test special ed kids. So, testing comes out of their own pockets and it is cheaper to test a whole group all at once. There are some districts who use individual test. And, there are child psychologists who use an array of individual tests. Because they use an array of tests, child psychs can usually get a clearer and less biased view of a child's intelligence.

I agree individual testing is more accurate and that it is a shame that there is no federal money devoted to gifted testing and education. It isn't clear to me though what testing you are saying that an child psychologist is doing that is good because on the other thread you declared that the major IQ tests were all biased. What tests specifically are referring to that escape your charge of bias?

[Roar]

Originally Posted by eilonwy

Most of the IQ tests used today haven't had a major overhaul since the 20's or 30's.

This is simply not true. IQ testing methods have changed dramatically over time. There is a recognized difficulty of testing children at the extreme upper and lower ends of IQ. (there isn't a big market in generating tests that can test at these levels). For this reason some testers have continued to use a copy of the SB (The SBLM) last revised in the 1960s as a supplemental test. This has been widely criticized because the definition of intelligence was so much different and testing methods have changed so much.

Originally Posted by eilonwy

How many of us have turpentine in our homes today, for example? When I was tested at 12, I'd never even seen a can of turpentine in real life, I'd only read about it and knew that it was a solvent used for painting. The question "Where does turpentine come from?" is hardly relevant to the lives of most American children. Seventy or eighty years ago, sure, but in the 90's? Heck no!

I'd hardly consider that strong evidence of bias in current IQ tests. There has been a strong increase in concern about bias since that time. And, we also have no idea if your local tester was even using a contemporary test at the time you were tested.

Originally Posted by eilonwy

Aside from the biased nature of the tests themselves, we have the folks administering the tests to contend with. The first time my IQ was tested I was five years old, getting ready to enter a new school in first grade. They wanted to hold me back a year despite the fact that I'd finished kindergarten with flying colors (long, messy story); My mother refused and insisted that they test me. The psychologist allowed the test to proceed to a point and then just stopped asking questions and told my mother to take me home. Apparently, my IQ was 129-- one point shy of the gifted program. Why? Well, I was tested in August, so I had a very dark tan. I also looked hispanic to the testers, they thought I was Puerto Rican, and there were no Puerto Rican kids in the gifted program. :

That stinks. Right and my white 5th grade teacher told me her white student that when I got to the gifted class the other kids would tease me because I didn't belong there and that she was bringing her concern to the school district that no longer should children be selected for the program without teacher recommendation because she didn't care how high my IQ was because I was a bad student. Of course the stakes weren't as high for me because I had some family support of my education...for a kid who didn't have that it could have been a killer.

Originally Posted by eilonwy

The high school I attended was more than 50% hispanic even ten years ago when I was there. Wanna guess how many hispanic kids were in my chemistry classes? How many took A.P. Chem a year or two later? *Two*. Two Puerto Rican girls, one Asian boy, four Asian girls, and me. That's it, as far as minority representation in *four* full secetions of Chemistry. In A.P. Chemistry, it was just me and one of the aforementioned hispanic girls. No black kids, none at all.

Yes, this is really problematic. A friend was told by her high school counselor that she should think about going to secretarial school and this was after she got one of the highest SAT scores in the school. She ended up getting a PhD in biochemistry! In other words bias can happen even after good results on testing.

Was access to AP classes granted based on testing at your school?

Originally Posted by eilonwy

The most highly gifted students ended up dropping out, bored out of their skulls and finding unique (and often illegal) ways to make ends meet. School was a waste of time not because they were too stupid to handle it, but because the school was too stupid to recognize that they needed a different sort of program, one which was more open to accomodating the needs of gifted students.

I agree this is a huge concern. There are a sizeable percentage of drop outs that are gifted kids - some identified and some not.

Originally Posted by eilonwy

In public schools, children are referred for testing most often by teachers. If the teacher has already decided that it's impossible for a child of color to be gifted, all of those kids are screwed. My first grade teacher used to routinely say (in front of me) things like, "She really belongs in pre-first, you can have her if you like," to the other first grade teachers. I don't think I ever heard the word, "stupid" from her, but I certainly heard, "lazy," "clumsy," and "just plain bad" on a daily basis.

That is very sad. And, while I don't doubt that it is more common for this stuff to happen to minority kids, I'll note the same stuff happened to me as a white kid. I had a couple of good years with good teachers, but a lot of wasted years too.

[eilonwy]

Originally Posted by Roar

Broadly declaring the tests can't test minority children without evidence of that contention is problematic in my opinion. If you have up to date research on racial bias in contemporary IQ tests I'd love to see it.

I don't think she's saying that tests can't test minority children, I think she's saying that they can't be relied upon for accuracy. Again, most of these tests haven't been changed since the 20's or 30's. You seem to be working from the assumption that contemporary IQ tests are different from the tests administered in the 50's and 60's. They're not. In fact, as far as I know, only the Stanford-Binet V is a "new" test; the most commonly used tests have been used as they are for at least 70 years.

Is it your suggestion that the people designing IQ tests are unaware of these concerns or that they are just disregarding them or that race is such a compelling factor that it is impossible to write an IQ test that isn't racially biased?

Again, I realize this wasn't directed toward me, but most of these tests haven't been redesigned in ages. There are more modern assessments available which are less racially biased, *however* they are not great for assessing verbal intelligence; those tests need *loads* of work.

Here's what we know... There is a strong relationship between race and class in this country. Minority kids are much more likely to be poor. More likely to have low quality child care before they begin school. More likely to attend underfunded schools. More likely to be born premature. More likely to be raised by a parent who hasn't graduated from college. More likely to have poor nutrition. More likely to have lead poisoning. I can see it would be faster and easier to blame the IQ tests than to tackle this long list of ways in which poor kids in this country get screwed over, but I don't think it would be accurate or really address the problems. George W. Bush tells us that the answer to our failing schools is to test, test, test and this just seems like a variation on that theme for me.

The class bias is another problem altogether, though the two are most assuredly related. The trouble is, what I'm hearing from this statement is that the way to address the bias in IQ testing is not to change the tests, but to change the children and make them behave more like folks of higher socio-economic status. While a lot of the changes would certainly be for the better, some of them would, of necessity, cross the line; they'd start changing cultral identity, and that wouldn't be fair or reasonable.

It'd be a lot less expensive to change IQ tests and to change the way the gifted students are identified in public schools than it would be to end poverty (though I freely admit that ending poverty would be a good thing).

[Roar]

Originally Posted by eilonwy

I don't think she's saying that tests can't test minority children, I think she's saying that they can't be relied upon for accuracy. Again, most of these tests haven't been changed since the 20's or 30's. You seem to be working from the assumption that contemporary IQ tests are different from the tests administered in the 50's and 60's. They're not. In fact, as far as I know, only the Stanford-Binet V is a "new" test; the most commonly used tests have been used as they are for at least 70 years.

I don't know where you are getting this. The two most commonly used tests are the Stanford-Binet V which came out in the last couple of years and the WISC IV which came out in 2003. Could you give me the names of the specific tests you are referring to?

Originally Posted by eilonwy

The class bias is another problem altogether, though the two are most assuredly related. The trouble is, what I'm hearing from this statement is that the way to address the bias in IQ testing is not to change the tests, but to change the children and make them behave more like folks of higher socio-economic status.

I'm not sure what statement you are referring to. My suggestion would be that IQ is not entirely fixed at birth. There certainly is an inborn component as those of us who have kids who have entirely taken us by surprise surely have experienced. However, I believe that innate talent isn't fixed over time and if children don't get a decent education, experiences, support that potential is lost. For this reason I think we are way behind the game when we are looking at a high school class and are frustrated that college prep classes aren't diverse enough. The problem started way before this.

Originally Posted by eilonwy

While a lot of the changes would certainly be for the better, some of them would, of necessity, cross the line; they'd start changing cultral identity, and that wouldn't be fair or reasonable.

What changes are you referring to?

Originally Posted by eilonwy

It'd be a lot less expensive to change IQ tests and to change the way the gifted students are identified in public schools than it would be to end poverty (though I freely admit that ending poverty would be a good thing).

I don't think we have any evidence yet that the problem in helping gifted kids in public school is one of testing.

[boongirl]

Originally Posted by eilonwy

In public schools, children are referred for testing most often by teachers.

In some systems. I taught in a school district for a few years where every 2nd grader was tested and then anyone older than that could be tested at the request of their parent. A much better system.

[boongirl]

Originally Posted by Roar

If you have up to date research on racial bias in contemporary IQ tests I'd love to see it.

I provided several links and feel they are enough. If you do not believe me, so be it.

Originally Posted by Roar

Is it your suggestion that the people designing IQ tests are unaware of these concerns or that they are just disregarding them or that race is such a compelling factor that it is impossible to write an IQ test that isn't racially biased? Is it your belief that as a teacher you have access to information from Hoagie's that people who spend years studying testing design somehow don't have?

This is a ridiculous ascertation. Most of the tests commonly used in school today (hence I called them the classic tests) were written long ago. In some ways, they are good tests. What has been done in school districts where I have taught is to recognize that race is one factor that can make it difficult for a child to do as well on the test as those of European descent. This does not mean the testers or testmakers are disregarding them. It is recognition that this disparity exists. The people who have been studying tests for years are well documented by their research on Hoagies and other website.

Originally Posted by Roar

More likely to have poor nutrition. More likely to have lead poisoning. I can see it would be faster and easier to blame the IQ tests than to tackle this long list of ways in which poor kids in this country get screwed over, but I don't think it would be accurate or really address the problems. George W. Bush tells us that the answer to our failing schools is to test, test, test and this just seems like a variation on that theme for me.

Testing kids in order to better place them in a program that meets their needs is not the same as standardized academic testing. How would one know whether a child really meets the criteria of special ed or gifted without testing? We cannot define a child based on our own whims. There has to be some standards. We can guess that a child has a reading disability or a behavior disorder but until we test them we don't know for sure.

I think the problem you are having with this argument is that you are seeing it as a way to offer poor and minority children special treatment. That argument vears so close to racism that I will not even begin to address it.

Originally Posted by Roar

All ifs you haven't demonstrated.

Anyone reading the research I cited or doing any kind of research on their own will come up with the same conclusion as I. You are the only one failing to see this. Why don't we turn the tables? Why don't YOU come up with some data dispelling that which I have cited? When that happens, we can have a discussion of like minds. As of right now, I feel as if I having a discussion with a person who has a serious problem with minority education in general. This troubles me.

Originally Posted by Roar

In particular I don't understnad your suggestion here that they created these tests before they knew about bias. The IQ tests most often being used weren't created in the 1950s or something. The SBV for example came out in the last year or two. These tests were studying on a diverse sample population and bias is a strong concern for test designers.

I am making a generalization about tests that is well documented. As I wrote above, when you have come up with some data to support your ascertation, then we can have a meaningful discussion. Please link some data to support your ascertations, as I have done. And, please take the time to read the research I have linked.

As an example of the fact that this is an important issue in school districts today, I will quote from my hometown school district:

Advanced Learning is working aggressively to address the issue of underpresentation of students of color and of poverty, and students who are bilingual in the Accelerated Progress Program and the Spectrum program through outreach, changes in testing practices, and the development of an early intervention program

Some of the ways they are trying to deal with the problem are:

* Students qualified for free/reduced lunch and/or are bilingual are provided additional testing opportunities, as appropriate.

* The nonverbal component of the Cognitive Abilities Test is now administered to all students, not just to those identified in the system as having bilingual status.

* The nonverbal component of the Cognitive Abilities Test is given equal consideration as the verbal and quantitative components of the test, with students found eligible who meet threshold in 2 out of these three areas.

Originally Posted by Roar

It isn't clear to me though what testing you are saying that an child psychologist is doing that is good because on the other thread you declared that the major IQ tests were all biased. What tests specifically are referring to that escape your charge of bias?

I have written this before and I will write it again. In a school, most kids are identified for the gifted program using one test. In a child psych's office, kids are tested with a multitude of tests. It is not so much that every test is inherently biased but the one test will not undo the possible bias in the results. Thus, using multiple measures of testing is more likely to give a clearer picture of IQ. A child psych will use an individual IQ test that involves manipulatives as well as written questions. He/she may also use a creativity test and a test of emotional intelligence. He/she may also use a test that demonstates intelligence in all intelligence areas.


Roar: When you have come up with some data supporting the idea that IQ tests are not inherently biased, please report back. Until that time, your argument is very weak.

[boongirl]

Originally Posted by Roar

I don't think we have any evidence yet that the problem in helping gifted kids in public school is one of testing.

It is almost sad how off the beaten track your ascertations have become. Really, no one is suggesting that the tests are so problematice that they do not help gifted kids. It is just suggested that some kids may not be qualifying for gifted programming due to bias in the tests and that that makes it necessary to think about the scores the result. Most school districts use one test to determine if the children are eligible. The idea we are all suggesting is that that some children may not score in the eligibility range due to extraneous circumstances. This is true not just of race but also of language. Many school districts do not use a non-verbal test for eligibility. This would mean that ESL kids don't have a chance of qualifying if their English skills are not good enough.


As I have written many times, when you have some data to back up your ascertains, as I have, then your argument will possibly hold weight. As it stands now, you are making claims that seem to be merely based on your opinion. That is not good enough. I have linked data to back up mine. When you have found some data, please come back and join the discussion. As it stands now, your claims hold no merit.

[eilonwy]

Originally Posted by boongirl

In some systems. I taught in a school district for a few years where every 2nd grader was tested and then anyone older than that could be tested at the request of their parent. A much better system.

Not really-- by second grade, children are already well known as "behavior problems" or "poor achievers," etc. Moreover, by second grade the brightest children have already tuned out and decided, consciously or otherwise, that school is utterly worthless.

Originally Posted by Roar

I don't think we have any evidence yet that the problem in helping gifted kids in public school is one of testing.

That's not exactly what I'm saying, but I wasn't terribly clear before so I'll try again: I believe that the problem of helping gifted minority students in public schools is largely related to identification. (Note: Not "testing," but identification.) How can these kids receive accomodations when schools and school districts are saying that they don't even exist, and actively refusing to look?

Do we have evidence that this is the problem, or one of the problems? Absolutely; the simple fact that even when you compensate for SES there are far, far fewer minority students participating in gifted programs around the country is enough. When you don't compensate for SES, the number drops even lower. Look at the population of any school in the country, and then look at it's gifted program; in most cases, the demographics will not line up, not even close.

[Roar]

Originally Posted by boongirl

I provided several links and feel they are enough. If you do not believe me, so be it.

You haven't posted a single thing that says that currently used IQ tests were formulated without efforts to pay attention to racial or class bias or that they include biased questions or that the difference we see in test scores is based on bias. If this was such a widespread contemporary problem as you claim it seems like you could provide even one link that talks about popular current IQ tests and racial bias. It your burden of proof because you are the one laying the claim. It is like you posted IQ tests discriminate against people with names that begin with vowels and then said "aha you haven't proved they don't".

Originally Posted by boongirl

This is a ridiculous ascertation. Most of the tests commonly used in school today (hence I called them the classic tests) were written long ago.

HUH? In the other thread by "classic tests" you said you meant standard IQ tests like the WISC and the Stanford Binet. These weren't written long ago but rather were released in the last five years. The WISC IV was released in 2003 and the Stanford Binet in in the last year or two. Could you please tell me specifically what test you are referring to and when it was written?

Originally Posted by boongirl

In some ways, they are good tests. What has been done in school districts where I have taught is to recognize that race is one factor that can make it difficult for a child to do as well on the test as those of European descent. This does not mean the testers or testmakers are disregarding them. It is recognition that this disparity exists. The people who have been studying tests for years are well documented by their research on Hoagies and other website.

If there is research on Hoagie's that says that IQ tests are currently used IQ tests are racially biased you haven't posted it.

I would suggest that even if the tests are perfectly without bias there would be reasons to offer entry at a different score level for minority children. That is what has been done in some affirmative action programs with SAT scores and I have no problem with that. One reason could be because the tests are biased - another could be because it is important to have a diverse school population and as a means to right contemporary and historical wrongs.

Originally Posted by boongirl

I think the problem you are having with this argument is that you are seeing it as a way to offer poor and minority children special treatment. That argument vears so close to racism that I will not even begin to address it.

The problem I'm having is that you are making gross generalizations and assumptions about testing without having anything to back up your assertions. You aren't indicating a knowledge of what tests are available, how they define intelligence or when and how they were created or tested. This is exactly the sort of thing that promotes paranoia and discourages parents from getting their kids tested when in fact testing may be the exact thing that leads to help for their child.

Originally Posted by boongirl

Anyone reading the research I cited or doing any kind of research on their own will come up with the same conclusion as I.

I can't imagine having the arrogance to assert that about anything! We are talking about issues that deeply divide people who devote their lives to the studying of intelligence testing, but it is your belief that everyone will agree with you?

Originally Posted by boongirl

You are the only one failing to see this. Why don't we turn the tables? Why don't YOU come up with some data dispelling that which I have cited? When that happens, we can have a discussion of like minds.

It is your burden of proof as you are the one laying the charge against IQ testing. I am not seeing proof nor logical responses.

We know that in this country poorer children are more likely to have lead poisoning, be in bad child care prior to the start of school, attend underfunded schools, have less access to educational opportunities, be raised by parents who didn't graduate from college, to have poor nutrition, to begin school with less school readiness, to not have been breastfed, to be exposed to more pollutants, etc. Do you agree this is true? Is it your belief that none of these things influence IQ? Could we raise half of the population in a box with nothing and raise the other half with lots of conversation, nutritious foods, exposure to educational opportunities and then when we test them at age 10 we say "aha the tests are biased" because the box kids did poorly? Or, would we say it is time to recognize that keeping kids in a box isn't giving them what they need to get a good start on life?

Originally Posted by boongirl

As of right now, I feel as if I having a discussion with a person who has a serious problem with minority education in general. This troubles me.

HUH? Where on earth are you getting that?

Originally Posted by boongirl

I have written this before and I will write it again. In a school, most kids are identified for the gifted program using one test. In a child psych's office, kids are tested with a multitude of tests. It is not so much that every test is inherently biased but the one test will not undo the possible bias in the results. Thus, using multiple measures of testing is more likely to give a clearer picture of IQ. A child psych will use an individual IQ test that involves manipulatives as well as written questions. He/she may also use a creativity test and a test of emotional intelligence. He/she may also use a test that demonstates intelligence in all intelligence areas.

Most assessments from psychologists consist of one or two tests. They may include IQ, acheivment or both. The IQ tests most typically used are the exact ones you referred to as "classic" unreliable tests that are racially biased and haven't been rewritten in fifty years. This simply isn't accurate. I don't know if you've had kids tested or how much you've read about testing but we've been through the process, read a great deal about it and the standard approach is IQ, acheivement or both. If you believe that psychologists are giving "emotional intelligence" test or "creativity tests" and using that to figure IQ you are mistaken.

I agree individual IQ tests are more accurate than group tests for reasons that go beyond race or class bias. I think it is entirely silly to suggest that we could put kids in a room with a paper and pencil test and figure out their intelligence levels.

Originally Posted by boongirl

Roar: When you have come up with some data supporting the idea that IQ tests are not inherently biased, please report back. Until that time, your argument is very weak.

Just a paragraph ago you said that child psychologists can test IQ. How does that work exactly? Are they using IQ tests because you already said those are biased.

You say that IQ tests are "inherently" biased. How is bias an essential element of IQ tests? Are you suggesting there is no way to think about intelligence that isn't racist?

What exactly would you consider proof of the lack of bias in IQ tests?

[Roar]

Originally Posted by boongirl

As I have written many times, when you have some data to back up your ascertains, as I have, then your argument will possibly hold weight. As it stands now, you are making claims that seem to be merely based on your opinion. That is not good enough. I have linked data to back up mine. When you have found some data, please come back and join the discussion. As it stands now, your claims hold no merit.

What link was it that contained data? Data that specifically shows IQ tests are biased? Please show me exactly where that is.

And, I think the word you are looking for is assertions. As in you posted a link that doesn't say contemporary IQ tests are biased, but you are asserting that it does.

[Roar]

Originally Posted by eilonwy

Do we have evidence that this is the problem, or one of the problems? Absolutely; the simple fact that even when you compensate for SES there are far, far fewer minority students participating in gifted programs around the country is enough. When you don't compensate for SES, the number drops even lower. Look at the population of any school in the country, and then look at it's gifted program; in most cases, the demographics will not line up, not even close.

Oh yes, I agree they don't line up at all. Minority kids are way underrepresented. I don't think though we have proof this is an issue of testing. The problem probably starts way before third grade when kids are tested.

The quality of schooling is so terribly unequal in this country. And, it is really hard to get people to talk about gifted programs when the basics aren't even there for a lot of kids.

[eilonwy]

Originally Posted by Roar

The quality of schooling is so terribly unequal in this country. And, it is really hard to get people to talk about gifted programs when the basics aren't even there for a lot of kids.

So, because the needs of many kids aren't being met, the parents of gifted children have no right to ask for accomodations? I'm not seeing your point here, or it's relevance to this particular discussion.

[LynnS6]

Originally Posted by Roar

The WISC IV was released in 2003 and the Stanford Binet in in the last year or two. Could you please tell me specifically what test you are referring to and when it was written?

OK, chiming in a bit late here -- most psychological tests are not reformulated from the 'ground up' when they are re-released. They are edited and updated, but the fundamental difficulties with how they assess and what they assess are not addressed. So, yes the WISC IV was released in the last 5 years, but it's not a "new" test. The often contain the biases of the old tests.

How can they be biased? Well, since my area is language development, I'll give you several examples from that area. First, most of these tests contain a vocabulary portion. The vocabulary (despite attempts to make it otherwise) is almost always biased toward middle class experiences -- things like wheelbarrow, binoculars, luggage, monument (all taken from a vocab test that I have in my office -- and it's 2001).

In addition, they will show a child a picture of something and ask them "what's this?" -- well if you grew up in a family where your parents read books and would ask you questions like this, that's an easy task. You're used to questions being asked for 'show'. When I ask my daughter "what color is this?" I'm not asking for information (I know darn well what color it is!), but for her to demonstrate or show me her knowledge.

There are lots of cultures in the U.S. (rural African American, for example, and I suspect, though I don't have data, urban poor African American) where parents simply don't ask this kind of question. Not because they don't care about their kids, not because they don't interact with their kids, but because why on earth would you ask your child a question when you already know the answer? They ask their kids "real" questions about things that they don't know! And they expect their kids to learn these other things through experience, observation or whatever.

Similar issues come up with things like story structure. Different cultures (sub-cultures) within the U.S. have different traditions for story telling and what constitutes a 'good story'. Most teachers recognize only the linear, middle class kind of story as a 'good one'. Stories that contain a lot of personal detail, but are more fluid in linking ideas (such as happens in a lot of communities) are considered to be bad or underdeveloped stories.

So, take a child who's used to a very different form of discourse between children and adults, put him/her into a room where the they are asking him questions that it's clear they know the answer to, or in a form the child is unfamiliar with, and that child will 'underperform'.

In terms of evidence of the fundamental problematic nature of IQ tests, I would highly recommend that you read Stephen J. Gould's The Mismeasure of Man (and yes, he uses MAN deliberately).

It's not just testing that is the cause of fewer minority children in gifted programs. But testing is one factor.

[Roar]

Originally Posted by eilonwy

So, because the needs of many kids aren't being met, the parents of gifted children have no right to ask for accomodations? I'm not seeing your point here, or it's relevance to this particular discussion.

Of course gifted parents should ask (scream, beg, whatever it takes) for accomidations but it can be especially difficult to be heard when what are seen as basics for all kids aren't in place.

[Roar]

Originally Posted by LynnS6

OK, chiming in a bit late here -- most psychological tests are not reformulated from the 'ground up' when they are re-released. They are edited and updated, but the fundamental difficulties with how they assess and what they assess are not addressed. So, yes the WISC IV was released in the last 5 years, but it's not a "new" test. The often contain the biases of the old tests.

Right. It would be inaccurate though to say as previous posters have that IQ tests are essentially unchanged in the last 70+ years. There is much greater awareness of the issue of bias and changing conceptions of how we define intelligence.

Originally Posted by LynnS6

How can they be biased? Well, since my area is language development, I'll give you several examples from that area. First, most of these tests contain a vocabulary portion. The vocabulary (despite attempts to make it otherwise) is almost always biased toward middle class experiences -- things like wheelbarrow, binoculars, luggage, monument (all taken from a vocab test that I have in my office -- and it's 2001).

I remember after my son was tested talking to the tester about this exact thing. I don't believe vocabulary is a pure measure of intelligence. However, we see a big range of scores even within middle class children with similar exposure so it is hard to argue it is all about class exposure.

Originally Posted by LynnS6

In addition, they will show a child a picture of something and ask them "what's this?" -- well if you grew up in a family where your parents read books and would ask you questions like this, that's an easy task. You're used to questions being asked for 'show'. When I ask my daughter "what color is this?" I'm not asking for information (I know darn well what color it is!), but for her to demonstrate or show me her knowledge.

I will note that I never asked my son those sort of questions either. I never pointed to something and asked him what color it was. He was never quizzed, asked to show or coached at home yet he scored highly on IQ tests.

Of course though it isn't "an easy task" as we see a range even with children with very similar experiences. It is important to note this so parents of children who score high on IQ tests don't just discount those results thinking well they are a middle class kid of course they score high.

That said, I think what this leads to is perhaps a more interesting question: Is intelligence fixed from birth? Is it all innate or does it change based on experience? My thought is that it is impossible to ever create a pure measure of innate intelligence because intelligence does change based on experience, diet, exposure, etc.

Originally Posted by LynnS6

In terms of evidence of the fundamental problematic nature of IQ tests, I would highly recommend that you read Stephen J. Gould's The Mismeasure of Man (and yes, he uses MAN deliberately).

I agree it is worth a read. I first read it twenty years ago. It offers great information about eugenics and early efforts to measure intelligence.

I don't see that it necessarily offers a lot of information that a parent considering contemporary IQ tests will find helpful. A friend with a child who was really struggling with a bad school fit would not be persuaded to get her child IQ tested because she was aware of Gould's book and some of the early history of measuring intelligence. Her contention was that they couldn't find out any accurate information because of course as a middle class, college educated white person her child would have a high IQ and she didn't want to be associated with racism by having her child tested. I'd hate for folks to reach that sort of conclusion because testing may prove useful in helping get a child's needs met.

[LynnS6]

Originally Posted by Roar

I remember after my son was tested talking to the tester about this exact thing. I don't believe vocabulary is a pure measure of intelligence. However, we see a big range of scores even within middle class children with similar exposure so it is hard to argue it is all about class exposure.

Well, I didn't say that it was ALL about class exposure. But, if you take a group of 100 middle class suburban children, and 100 inner city children, I guarantee you there are going to be more children in the suburbs who know what wheelbarrows and binoculars are.

I would say that it's the accumulation of all the little factors that creates the bias. Not one single test. Not one single word. And not any individual child.

[Roar]

Originally Posted by LynnS6

Well, I didn't say that it was ALL about class exposure. But, if you take a group of 100 middle class suburban children, and 100 inner city children, I guarantee you there are going to be more children in the suburbs who know what wheelbarrows and binoculars are.

I would say that it's the accumulation of all the little factors that creates the bias. Not one single test. Not one single word. And not any individual child.

At any point though would you say that exposure relates to intelligence? As in kids who are exposed to particular information, tools or kinds of thinking are better able to utilize their innate abilities and develop as thinkers. While I wasn't particularly thrilled when my son started reading early, I had to observe that it wasn't just a matter of knowing more stuff from reading (as in more vocabulary), but that having a greater catalog of information and vocabulary lead to different observations, ways of thinking or cataloging systems in his brain. I'm not articulating this very well, but perhaps you understand what I'm getting at.

[eilonwy]

Originally Posted by Roar

Right. It would be inaccurate though to say as previous posters have that IQ tests are essentially unchanged in the last 70+ years. There is much greater awareness of the issue of bias and changing conceptions of how we define intelligence.

The basic tests are, in essence, unchanged from all I've read/seen. The most commonly used tests (and the SB V isn't in common use yet) are not that different from what they were 60 years ago.

I remember after my son was tested talking to the tester about this exact thing. I don't believe vocabulary is a pure measure of intelligence. However, we see a big range of scores even within middle class children with similar exposure so it is hard to argue it is all about class exposure.

Nothing is all about anything; my contention is, however, simply that the tests are biased in favor of white, Christian, middle-class folks.

Of course though it isn't "an easy task" as we see a range even with children with very similar experiences. It is important to note this so parents of children who score high on IQ tests don't just discount those results thinking well they are a middle class kid of course they score high.

I wouldn't say that it's grounds for discounting the all of the results, only that there ought to be a noted differential. In other words, children from certan backgrounds would have to start at different places, or maybe the scores would be adjusted afterwards according to the degree of bias of the test. I also think that IQ tests should be administered by someone who is actually engaging in dialogue with the child, and who is open minded. Quite honestly, it's very easy for me to tell the difference between a child who's memorized a lot of things, a child who's a great performer and a child who is truly gifted (their mind works differently from that of the average child). I can't imagine that someone with experience in this area, i.e. a child psychologist, wouldn't be able to draw the same conclusions, especially with a bit of training.

That said, I think what this leads to is perhaps a more interesting question: Is intelligence fixed from birth? Is it all innate or does it change based on experience? My thought is that it is impossible to ever create a pure measure of innate intelligence because intelligence does change based on experience, diet, exposure, etc.

I don't believe that intelligence actually changes, but it's a fact that people test differently under different conditions. I think that it probably is possible to come up with a pure measure of innate intelligence, even with the technology that we have available today; I just don't think that, as things stand, it would be economically feasable. It wouldn't surprise me at all to see one in the near future, though.

Her contention was that they couldn't find out any accurate information because of course as a middle class, college educated white person her child would have a high IQ and she didn't want to be associated with racism by having her child tested. I'd hate for folks to reach that sort of conclusion because testing may prove useful in helping get a child's needs met.

I've never heard of such a thing, but it doesn't make logical sense to me at all. Why wouldn't she want to do something that could, potentially, help her children? If she's concerned about the racial and SES biases of the tests, she could be working to change them.