JSO writes educational goal policy in Virginia

[dbackjon]

As part of Virginia's waiver to opt out of mandates set out in the No Child Left Behind law, the state has created a controversial new set of education goals that are higher for white and Asian kids than for blacks, Latinos and students with disabilities.

Virginia Democratic state Sen. Donald McEachin first read about the state's new performance goals for schoolchildren in a newspaper editorial.

"And I was shocked to find that the state board of education [was] putting in place permanent disparities between different subgroups — Asians at the top, African-Americans at the bottom," says McEachin.

Here's what the Virginia state board of education actually did. It looked at students' test scores in reading and math and then proposed new passing rates. In math it set an acceptable passing rate at 82 percent for Asian students, 68 percent for whites, 52 percent for Latinos, 45 percent for blacks and 33 percent for kids with disabilities.

http://www.nwpr.org/post/firestorm-erup ... tion-goals" onclick="window.open(this.href);return false;

[JohnStOnge]

You know at first I was going to say it's not something I would do but now I'm not sure. I'm not completely sure, but it looks like what they're trying to do is be fair about assessing the performance of schools by taking racial demographics into account.

And of course that runs afoul of egalitarian denial.

I'd do it in a different way. In order to avoid this kind of controversy I'd do something like give each kid a series of tests at the beginning of each year to establish a baseline. Then how they performed at the end of the year would be compared to how they performed at the end. That would avoid getting into race. Of course what would happen is that Blacks would score lowest at the beginning of the year and at the end. Asians would score highest both times. So you'd still be taking race into account.

But you wouldn't be obvious about it so chances are the dream-world egalitarians wouldn't raise a big stink.

[danefan]

This is the dumbest thing I've ever seen.

[JohnStOnge]

This is the dumbest thing I've ever seen.

No it's not. It's not fair to hold a school that's 90% Black and one that's 80% White and 10% Asian to the same standard. It's just not. If you do you're in denial of reality.

You know, I see people here post stuff on how being skeptical about anthropogenic climate change is ignoring the data. Well you know what? If you don't realize that ethnicity matters with this test scores thing you're ignoring the data. And you're ignoring data that is a whole lot more up front and obvious than the climate change data are.

[Chizzang]

This is why our education system is bottoming out...
When can we throw the entire K-thru-12 thing in the dumpster and start over

[CAA Flagship]

This is why our education system is bottoming out...
When can we throw the entire K-thru-12 thing in the dumpster and start over

I already have. And it's costing me about $25k a year.

[houndawg]

This is the dumbest thing I've ever seen.

No it's not. It's not fair to hold a school that's 90% Black and one that's 80% White and 10% Asian to the same standard. It's just not. If you do you're in denial of reality.

You know, I see people here post stuff on how being skeptical about anthropogenic climate change is ignoring the data. Well you know what? If you don't realize that ethnicity matters with this test scores thing you're ignoring the data. And you're ignoring data that is a whole lot more up front and obvious than the climate change data are.

And you're ignoring property taxes.

[JohnStOnge]

And you're ignoring property taxes.

If what you're getting at is that schools in better school districts get more money due to property taxes, no I'm not. If those schools have different races the racial gaps exist within those schools. Now, socioeconomic status does matter. But you don't eliminate the racial gaps by adjusting for that.

Besides, it's still not fair to fail to adjust for demographic factors. Go ahead and take socioeconomic status and tax base into account. But what will happen is that race will still, mathematically, show up as a factor. If you ignore it because you're being politically correct and act like it doesn't matter in a mathematical sense you're not being fair.

But, as I said, you could eliminate all that argument simply by testing each student at the beginning of the year in order to establish a baseline for that student. That way the reasons for why that student is where he or she is doesn't matter. He or she is where he or she is. Then at the end when they are tested you can provide a fair assessment by comparing where they ended to where they started.

[CID1990]

Why would a librul have a prollem wit dis?

[JohnStOnge]

Why would a librul have a prollem wit dis?

Because it's basically saying that race matters. If they'd adjusted based on something like whether or not children are on the school lunch program, whether they live in single parent homes, and/or their parents' education level no one would've objected. The problem is that race doesn't go away as a factor when you take those factors and others into account. So you can't be fair and ignore it.

But if you address it you're in for it; as you can see in the linked article. So you can't be fair to the school systems and teachers involved. You have to pretend that it doesn't make any difference.

[kalm]

Why would a librul have a prollem wit dis?

Because it's basically saying that race matters. If they'd adjusted based on something like whether or not children are on the school lunch program, whether they live in single parent homes, and/or their parents' education level no one would've objected. The problem is that race doesn't go away as a factor when you take those factors and others into account.

Link?

[JohnStOnge]

Link?

http://nces.ed.gov/nationsreportcard/naepdata/" onclick="window.open(this.href);return false;

Click on "Main NDE." You will get a series of queries and you can break data down by different Demographic factors.

If you want I can go ahead and do a few breakdowns for you but I've done it before. Do you recall previous posts in which I used a comparison of extremes? I reported that on the 8th and 12th grade math tests poor Asian children of high school drop outs scored higher, on average, than middle and upper income Black children of parents who graduated college. If you want me to go through the discussion of that again, let me know.

[kalm]

Link?

http://nces.ed.gov/nationsreportcard/naepdata/" onclick="window.open(this.href);return false;

Click on "Main NDE." You will get a series of queries and you can break data down by different Demographic factors.

If you want I can go ahead and do a few breakdowns for you but I've done it before. Do you recall previous posts in which I used a comparison of extremes? I reported that on the 8th and 12th grade math tests poor Asian children of high school drop outs scored higher, on average, than middle and upper income Black children of parents who graduated college. If you want me to go through the discussion of that again, let me know.

Yes. And also please include single parent homes and school lunch programs.

[JohnStOnge]

Yes. And also please include single parent homes and school lunch programs.

The NAEP query only allows me to consider three factors as once. Right now I don't see "single parent homes" as a factor it allows you to get to. However, "school lunch programs" is what I was referring to in saying "poor."

I will take you through how much of a difference being eligible or not eligible for the school lunch program and parental education level makes. Then I will compare extremes. I will compare the most disadvantaged Asian children in those terms to the most advantaged Black children. When I am done, if you do not see the problem with ignoring race, you are just not willing to see.

I will use the 2009 NAEP 8th grade math test. Average scores.

Children Eligible for School Lunch Program: 137
Children not Eligible for the School Lunch Program: 160

Children of Parents who did not finish High School: 135
Children with at least one Parent who graduated college: 164

Eligible for School Lunch, parents didn't finish high school: 135
Not Eligible, parents graduated college: 168

Ok. so you see that overall children not eligible for school lunch with at least one college grad parent scored 33 points higher, on average, than school lunch children with parents who didn't finish high school.

Yet Asian children who were eligible for the school lunch program with parents who didn't finish high school averaged 162 while Black children who were NOT eligible for the school lunch program with at least one parent who graduated from college averaged 144. The most disadvantaged Asian children in terms of school lunch program eligibility and parental education level scored 18 points higher, on average, than the most advantaged Black children in those terms.

I suppose you could argue that it might change the bottom line if I could get to "single parent" homes. But if you do you're grasping at straws. You're talking about "advantaged" children having a 33 point advantage when race isn't considered then swinging to "disadvantaged" children of one racial group having an 18 point edge over "advantaged" children in another. That's a 51 point difference from what you would predict if you did not take race into account.

You're not going to mathematically explain race away. It just isn't going to happen.

[Seahawks08]

Yes. And also please include single parent homes and school lunch programs.

I hate you so much right now.

[JohnStOnge]

Yes. And also please include single parent homes and school lunch programs.

I hate you so much right now.

Don't even start reading if you're going to complain about the post being long. If somebody asks me to back up what I say, and someone did a few posts earlier, I'm going to back up what I say. And there are some things that can't be backed up with a clever little paragraph.

The single parent home factor is largely going to be captured by the eligible or not eligible for the school lunch program factor. I guarantee you you cannot make the race factor go away mathematically by taking measurable environmental factors into account. What you see as a result is people coming up with unsubstantiated explanations such as Asians having a culture of achievement.

Let's cross compare. Terms first to review.

Disadvantaged: Eligible for the School Lunch Program, neither parent finished high school

Advantaged: Not eligible for the School Lunch Program, at least one parent graduated from college.

Here is the cross comparison.

Advantaged Asians 187, disadvantaged Blacks 124. Advantaged Asians +63

Disadvantaged Asians 162, advantaged Blacks 144. Advantaged Asians +18

Notice that, within each racial group, the difference between being advantaged and being disadvantaged is about the same. 25 points for Asians and 22 points for Blacks.

Now notice that, within advantaged and disadvantaged groups, the difference between racial groups is also similar. Among advantaged kids Asians outscored Blacks by 43. Among disadvantaged kids Asians outscored Blacks by 38.

What that suggests is that, mathematically, being Black vs. being Asian is a "more important" factor than being advantaged vs. being disadvantaged is.

Now, that does not prove that "race" is causing the difference. It's an association in observational data and it's not possible to do an experiment to get the answer. We can never know.

But it's not rational to continue to operate on the assumption that "race," or differences in the distributions of innate potential for mathematics aptitude, is NOT a causal factor. And that's what people want to do. Because of egalitarian bias they start off with the premise that it is not a causal factor, refuse to accept it being a causal factor as a possibility, and set about trying to find SOME other explanation. Even entertaining the thought that it is indeed a causal factor, regardless of how rational that is, is unacceptable. And if you're a researcher who does it there's going to be hell to pay. You'll be a pariah.

Whether "race" in itself causes it or not it's there. One thing observational data CAN do is tell you something's there. It's there and one cannot explain it away by considering measurable environmental factors. So don't tell a teacher or a school system that you're going to say they have to achieve the same distribution of test scores for a 95% Black school or class as some other teacher or school system that's got a school or class that's 90% White and 8% Asian. It's just not fair. As I said if you don't want to publicly admit that race makes a difference just do baseline testing on a student by student basis. So you can be fair without running afoul of any egalitarian denials of reality.

As a point of clarification when I say "race makes a difference" I'm talking about mathematically. It's capturing something. It's not the same as saying "race in itself causes the difference." Without an experiment there is always the possibility that the causal factor or factors just happen to be associated with race and we can't identify them. Can't think of what they might be or can't demonstrate them.

For example if we just looked at it mathematically and just looked at the relationship between hair length and height we'd get an association whereby there is a negative association in the United States between hair length and height. But cutting your hair doesn't cause you to grow taller. It just so happens that women have longer hair on average and are also shorter on average.

The difference is there is an obvious, demonstrable explanation for that association and there isn't for the race/test score association. People can come up with speculative explanations but they can't really demonstrate that those explanations account for the entire magnitudes of the differences. Besides, even if someone is right about the difference between Blacks and Asians (for instance) being entirely due to the "culture of achievement" among Asians, it's still not fair to make a teacher "achieve" the same success with Blacks as he or she does with Asians. He or she has no control over the differences in culture.

[houndawg]

And you're ignoring property taxes.

If what you're getting at is that schools in better school districts get more money due to property taxes, no I'm not. If those schools have different races the racial gaps exist within those schools. Now, socioeconomic status does matter. But you don't eliminate the racial gaps by adjusting for that.

Besides, it's still not fair to fail to adjust for demographic factors. Go ahead and take socioeconomic status and tax base into account. But what will happen is that race will still, mathematically, show up as a factor. If you ignore it because you're being politically correct and act like it doesn't matter in a mathematical sense you're not being fair.

But, as I said, you could eliminate all that argument simply by testing each student at the beginning of the year in order to establish a baseline for that student. That way the reasons for why that student is where he or she is doesn't matter. He or she is where he or she is. Then at the end when they are tested you can provide a fair assessment by comparing where they ended to where they started.

No, what you have to do is is take your two hypothetical schools and switch funding for 12 years and then compare the results of kids that had gone K-12 in the black school with the white school funding vs kids that had gone K-12 in the white school with the black school funding.

[kalm]

Yes. And also please include single parent homes and school lunch programs.

I hate you so much right now.

[Pwns]

No, what you have to do is is take your two hypothetical schools and switch funding for 12 years and then compare the results of kids that had gone K-12 in the black school with the white school funding vs kids that had gone K-12 in the white school with the black school funding.

Look up the DC and Oakland school districts. Plenty of funding, completely lousy results.

Even though I don't necessarily agree with the VA ed. policy, why the outrage when liberals think certain ethnic groups should not have to do as well on the SAT regardless of SES to get into college? And they get outraged at organizations like the New Haven fire department demanding minority employees get the same score as white employees to get promotions. Total pot calling the kettle black for getting outraged a this.

[DSUrocks07]

No, what you have to do is is take your two hypothetical schools and switch funding for 12 years and then compare the results of kids that had gone K-12 in the black school with the white school funding vs kids that had gone K-12 in the white school with the black school funding.

Look up the DC and Oakland school districts. Plenty of funding, completely lousy results.

Even though I don't necessarily agree with the VA ed. policy, why the outrage when liberals think certain ethnic groups should not have to do as well on the SAT regardless of SES to get into college? And they get outraged at organizations like the New Haven fire department demanding minority employees get the same score as white employees to get promotions. Total pot calling the kettle black for getting outraged a this.

This 100%

[LeadBolt]

Wow! And we wonder why we have an EEOC driven hire as President...

[dbackjon]

No, what you have to do is is take your two hypothetical schools and switch funding for 12 years and then compare the results of kids that had gone K-12 in the black school with the white school funding vs kids that had gone K-12 in the white school with the black school funding.

Look up the DC and Oakland school districts. Plenty of funding, completely lousy results.

Even though I don't necessarily agree with the VA ed. policy, why the outrage when liberals think certain ethnic groups should not have to do as well on the SAT regardless of SES to get into college? And they get outraged at organizations like the New Haven fire department demanding minority employees get the same score as white employees to get promotions. Total pot calling the kettle black for getting outraged a this.

I don't think that...

[JohnStOnge]

No, what you have to do is is take your two hypothetical schools and switch funding for 12 years and then compare the results of kids that had gone K-12 in the black school with the white school funding vs kids that had gone K-12 in the white school with the black school funding.

I doubt that anyone will be able to do that experiment But what you could do to get an idea is look at "within school" variation. There are Black children that go to schools in well funded districts. So you could compare the scores of Black kids that go to those schools to those of White and Asian kids within the same schools. You could do it with poor schools too. That would give you some idea of how much of the overall gap is due to the school kids go to and how much is due to something else.

But I am absolutely confident you would still see differences between races that would exist within schools and could not be mathematically explained away be measurable factors.

NAEP doesn't allow me to look at individual schools. At least not as far as I know. But it does allow me to compare States. And it's clear that the preponderance of the variation between States can be accounted for by the demographics of the students. In other words, the demographic make up of the student population is more important in explaining variation in scores than the State they are educated in. And I think that captures differences in at least State wide funding to some extent.

[JohnStOnge]

No, what you have to do is is take your two hypothetical schools and switch funding for 12 years and then compare the results of kids that had gone K-12 in the black school with the white school funding vs kids that had gone K-12 in the white school with the black school funding.

Came back to say that I did find something in the available list of variables at the NAEP site that allows KIND of getting at the funding question. One can compare "city" vs. "suburban" schools. I think all would agree that suburban schools tend to be better funded. It's where those engaged in "White Flight" flee to get away from urban school systems. And the "details" link by the variable selector for "school location" is:

Type of community where school is located, based on Census data describing proximity to an urbanized area (a densely settled core with densely settled surrounding areas) using four categories
ID: UTOL4
Values: City, Suburb, Town, Rura

I think it's fair to say "City" is equivalent to "Urban."

Anyway. The average score of national public school students at "City" schools is 151 and that of "Suburb" schools is 156. Consistent with the idea that students in better funded schools do better.

But the difference pretty much goes away when a couple of demographic factors, race and school lunch program eligibility, are taken into account. For convenience I will use "poor" to describe "eligible for the school lunch program" and "middle/upper class" to describe "not eligible for the school program." I could not include parental education level because the NAEP query only lets me cross tabulate three variables at once.

Here we go:

Poor, Black, City: 128
Poor, Black, Suburb: 127

Middle/Upper Class, Black, City: 140
Middle/Upper Class, Black, Suburb: 135

Poor, Asian, City: 166
Poor, Asian, Suburb: 161

Middle/Upper Class, Asian, City: 183
Middle/Upper Class, Asian, Suburb: 181

I will stop with Blacks and Asians because that's what we've been talking about. But the same pattern generally holds for Hispanics and Whites. The estimate for City School becomes higher than that for Suburb school in all but 2 of the 16 available race x school lunch program eligibility x school location comparisons. There is a tie with Poor Hispanics (134 to 134) and Suburb does better with Poor Whites (150 vs. 147).

So what happens, overall, is that the "advantage" of the Suburb schools goes away when race and school lunch program eligibility are taken into account. In fact, if anything, the numbers suggest a slight advantage for the City schools when that adjustment is made. That is consistent with the impression I've gotten over many years of looking at data like that: The demographics of the student population are more important than school system, teacher, jurisdiction, funding, etc. And race is a demographic factor that, mathematically, is important. You cannot, in an objective mathematical sense, ignore it and end up with a fair assessment.

But that is what is generally done, for reasons that become evident when you see something like the reaction to what Virginia wants to do.