Championship Subdivision Football | FCS Football | Stadiums | Blogs | Forums

I adjusted State rankings on the 2013 National Assessment of Educational Progress (NAEP) 8th grad math score based on the model I mentioned in a previous post. Yes I know nobody cared but I'm having a good time. Instead of listing everybody I'll use convention. Above average States are States that have adjusted average scores more than one standard deviation above the mean of all the State averages. Below average states are those that have adjusted average scores more than one standard deviation below the mean of all State averages.

So here's how it goes (scores are departures from the average):

Above average States:

1. Texas 12.25
2. New Jersey 6.76
3. Ohio 5.99
4. Indiana 5.71
5. North Carolina 4.28
6. Pennsylvania 4.15
7. Wyoming 4.14
8. Massachusetts 4.10

Below average States:

43. Utah -3.90
44. Oregon -3.98
45. Rhode Island -4.16
46. New York -4.66
47. California -5.15
48. West Virginia -5.83
49. Connecticut -6.75
50. Alabama -8.20

Everybody else is in the average range within -3.24 through 2.99. Fairly different impression with respect to education systems in various States from just looking at raw standardized test scores without being fair about it and making some effort to take demographics into account. For example (since I'm from the South): A Southern State is #1 and only one Southern State is in the below average range. Yes that Southern State is at the very bottom so that the South has both #1 and #50. But, still, you've got three Northeastern States in the below average range as compared to only one Southern State.

It changes the impression. No doubt. And it's a far fairer way to look at educational systems. Additionally, what I did is very crude and imprecise. It barely scratches the surface of adjusting for demographics.

http://www.amazon.com/Passing-Great-Rac ... B008KNW1Q0" onclick="window.open(this.href);return false;

Grizalltheway wrote:http://www.amazon.com/Passing-Great-Rac ... B008KNW1Q0

Well, there are both hereditary and environmental factors. I think that the relative importance of heredity depends on how much variation there is in environmental factors. For example: If I grow one plant of a certain species in sand and don't water it much while I grow another in rich soil suitable for its growth and water it appropriately the one I take good care of is probably going to grow better than the one I neglect.

And the model I used probably reflects both hereditary and environmental factors. It used the percent of people in each State who had at least a bachelors degree. It's reasonable to think that kids in States with higher percentages of people having bachelors degrees had better environments on average in material terms as well as in terms of intellectual stimulation. And the other factor I used was the percent of people in each State who were non hispanic white and asian. Those groups probably tend to enjoy better socioeconomic status than hispanics, blacks, and native americans.

At the same time both factors may also reflect heredity to some extent. So it's kind of a chicken or egg thing.

But regardless of all that this thing that's routinely done where States and/or school systems are ranked by raw standardized test scores without ANY consideration at all of the demographics each State or school system is dealing with is grossly unfair. You can't legitimately rate school systems like that. But in practice it's the standard way of doing it.

At an even lower level there are State rating systems that rate teachers like that. So if you're a teacher in a nice rich white school district you're in like Flynn while if you are a teacher in a poor minority school district your back is against the wall and you're fighting for your professional life. Doesn't matter how much of it is environmental and how much of it is heredity. It's real regardless.

Question - are you looking for a correlation between skin color and standardized test scores?

Skjellyfetti wrote:Question - are you looking for a correlation between skin color and standardized test scores?

In one sense, yes. I don't know if "skin color" is exactly it. But in that test I'm talking about there are definitely differences between mean scores of the racial groups the NAEP categorizes. The overall national average score on the test is 284.61. The means for the races as categorized by the NAEP are as follows:

Asian/Pacific Islander: 306.35
White 293.98
Hispanic 271.55
American Indian/Alaska Native 269.42
Black 263.16

The variable I used was the combined percentages in each given State of the two groups that scoresd above the overall average; except that I was using Census data to get those percentages and using "Asian Only." So I was using the Census Bureau combined percentage of Whites that are not hispanic and "Asians" only.

If you're wondering if I used some indicator of socioeconomic status in my model I tried. I tried using both percent below the poverty level in each State and median household income. Both "fell out" as "significant" factors when "controlled" for percent with bachelors degrees and race.

I don't think that means being poor or rich doesn't matter. In fact I know from looking in detail at NAEP data that it does. I think what's happening is that both of those other variables indirectly take that into account. People with college degrees tend to do better financially and whites as well as asians tend to do better financially than the other racial groups. So I kind of think with the particular thing I did measures of socioeconomic status I tried to use "fell out" because that was already indirectly taken into account by the other variables.

Just curious how you were breaking down races.

It's so subjective... and since you are someone that is always bitching about a strict scientific method, complaining about psychology, etc. you embracing this superficiality is kinda surprising.

Take Asian / Pacific Islanders vs. American Indians for example. Incredibly similar in every respect. Qualitatively, genetically, etc. etc. They're also so intermixed with whites in most cases that they're more white than they are American Indian. Do your figures include only those that are >50% American Indian? >25% American Indian? Or what?

How do you explain the HUGE difference in scores between groups so biologically similar as Asians and American Indians (especially Alaska Natives and Asians for fucks sake) as anything but cultural?

"2013 National Assessment of Educational Progress (NAEP) 8th grad math score."

That seems oddly specific. Why not multiple years, multiple grades, and multiple types of test?

It's so subjective... and since you are someone that is always bitching about a strict scientific method, complaining about psychology, etc. you embracing this superficiality is kinda surprising.

The scientific method is for inferring cause and effect. What we're looking at here is association. I have opinions as to why there are associations. But I have no problem with saying up front that I can't prove my opinions are correct.

The NAEP categorization I used just goes with how schools categorize people in terms of race. I'm sure there is some error associated with that. People that most wouldn't be sure of how to classify just looking at them. I can't get to numbers in each group in the NAEP data so I used the "background" from each State estimated by the Census Bureau. Same thing. But both the Census Bureau and the NAEP appear to think it's possible to reasonably categorize people by "race" and to think it's important to do so.

For my purposes in this case the "why" one gets differences when things are broken down according to how schools or the Census Bureau assign "race" to people doesn't matter. There are clearly population differences. "Race" is a factor that does not go away when one tries "controlling" for other factors such as socioeconomic status and parental education level. The "why" doesn't matter to the fact that it isn't fair to judge the performances of school systems based on standardized test scores without taking the racial composition of the student body it serves into account.

Just on the issue of fairness to educators, Sky, take a look at the results below. The parental education level refers to the highest level of education achieved by at least one parent. "Low SES" means eligible for the school lunch program, "Upper/Middle SES" means not eligible for the school lunch program. I'm just going to give the results for White and Black because that's the two largest groups and at the center of a lot of angst. Look at the scores and tell me if it's fair to say one teacher who has a classroom where 60% of the students are poor Black kids and 20% of them are poor Black kids whose parents didn't graduate from high school did a bad job because her class standardized test score is lower than some other teacher in a White suburb where 90 percent of her students are middle to upper class White kids and half of them have at least one parent who graduated college. And tell me if you think it's fair to that teacher to refuse to consider the difference between races in every "Parental Education/SES" category. Parental Education level for each group followed by SES.

Did Not Finish High School/Low SES

White 268, Black 253

Did Not Finish High School/Middle-Upper SES

White 278, Black 256

Graduated High School/Low SES

White 273, Black 252

Graduated High School/Middle-Upper SES

White 282, Black 260

Some Education After High School/Low SES

White 286, Black 269

Some Education After High School/Middle-Upper SES

White 296, Black 276

College Graduate/Low SES

White 285, Black 261

College Graduate/Middle-Upper SES

White 305, Black 280

One more thing on the unfairness of refusing to consider race when judging school systems. I think you've probably heard people saying that a school system has problems because the students are poor. I hear that kind of Statement about the East Baton Rouge public school system, for instance. People will NOT under any circumstances refer to the fact that the school system is predominantly Black as a factor in low standardized test scores and a resulting low system rating.

But look at the average national scores for low SES whites (eligible for the school lunch program) and middle to upper SES Blacks (not eligible for the school lunch program) on that 2013 NAEP math test I'm using to illustrate things:

Low SES Whites 278
Middle-Upper SES Blacks 275

Now what happens when we reverse it? Let's compare Upper-Middle SES Whites to Low SES Blacks:

Upper-Middle SES Whites 300
Low SES Blacks 258

It's just not intellectually honest to say "it's because they're poor" as though that's the whole story. Obviously being poor is a factor. Within both racial groups the Upper-Middle SES students scored substantially higher. Among Whites it was 22 points higher and among Blacks it was 17 points higher.

Yet in spite of the difference being poor obviously made poor Whites scored slightly HIGHER on average than upper to middle class Blacks did. And the difference is "statistically significant" at above the 99% confidence level (you can do a statistical test at the site).

In the face of such data, and there are data like that all over the place, it's absolutely irresponsible for people to persistently make statements transmitting the message that it's all because of poverty. That's part of it. But it's nowhere close to all of it. In the numbers above the gaps based on SES within racial groups are 22 and 17 while the gaps based on race within SEC groups are 25 and 20. It's clear that "race" makes at least as big a difference as "SES" does in mathematical terms.

But people can't say that. It's basically lying to people so they won't entertain thoughts the culture doesn't want them entertaining. It's a "YOU CAN'T HANDLE THE TRUTH approach.

I can appreciate this. I've always thought it's interesting that when we compare education achievement across states they are never adjusted for race, as if that would somehow affirm evolutionary superiority of certain ethnic groups. The achievement gaps exist in the reddest and bluest states so it makes sense to make the adjustments. I've seen countless maps and charts relating to education showing how supposedly "dumb" the red state south is and course there is never any adjustment for race. Apparently calling southern states dumb for their test scores is acceptable but simply pointing out difference in test scores (even without attributing that to evolution and genetics) across races is offensive.

∞∞∞ wrote:"2013 National Assessment of Educational Progress (NAEP) 8th grad math score."

That seems oddly specific. Why not multiple years, multiple grades, and multiple types of test?

It takes a lot of work to breakdown stats from tests. However, over the years, JSO has done many comparisons with different tests and across multiple ages. The results has been consistent.

People might not like what JSO is reporting, but as least he's done the work to back up his statements.

And, as much as skelly and others LOVE to find links that counter conservative arguments, you'll notice they are conspicuously silent whenever JSO digs into data to strip off the PC veil that the social media loves to keep in front of everyone's eyes.

Cluck U wrote:

∞∞∞ wrote:"2013 National Assessment of Educational Progress (NAEP) 8th grad math score."

That seems oddly specific. Why not multiple years, multiple grades, and multiple types of test?

It takes a lot of work to breakdown stats from tests. However, over the years, JSO has done many comparisons with different tests and across multiple ages. The results has been consistent.

People might not like what JSO is reporting, but as least he's done the work to back up his statements.

And, as much as skelly and others LOVE to find links that counter conservative arguments, you'll notice they are conspicuously silent whenever JSO digs into data to strip off the PC veil that the social media loves to keep in front of everyone's eyes.

You're generally called a racist if you try to report the actual stats on this.

89Hen wrote:

Cluck U wrote:
It takes a lot of work to breakdown stats from tests. However, over the years, JSO has done many comparisons with different tests and across multiple ages. The results has been consistent.

People might not like what JSO is reporting, but as least he's done the work to back up his statements.

And, as much as skelly and others LOVE to find links that counter conservative arguments, you'll notice they are conspicuously silent whenever JSO digs into data to strip off the PC veil that the social media loves to keep in front of everyone's eyes.

You're generally called a racist if you try to report the actual stats on this.

Bullshit. This report is flawed...it doesn't account for l the dumb WHITE southerners.

kalm wrote:
Bullshit. This report is flawed...it doesn't account for l the dumb WHITE southerners.

Perfect timing...at least for those on this side of the joke.

89Hen wrote:

Cluck U wrote:
It takes a lot of work to breakdown stats from tests. However, over the years, JSO has done many comparisons with different tests and across multiple ages. The results has been consistent.

People might not like what JSO is reporting, but as least he's done the work to back up his statements.

And, as much as skelly and others LOVE to find links that counter conservative arguments, you'll notice they are conspicuously silent whenever JSO digs into data to strip off the PC veil that the social media loves to keep in front of everyone's eyes.

You're generally called a racist if you try to report the actual stats on this.

And that is a shame.

If, indeed, skelly is correct, and this is a cultural issue, then identifying it as such would allow people to take responsibility for such a culture and to make changes. Even if it were genetic, then one could find ways to overcome such disadvantages, or maybe to play to certain strengths.

Instead, everyone chooses to ignore the elephant in the room (or, as you mention, those that don't get vilified) while ineffectively throwing away bigger gobs of money at the problem while forcing other, more successful cultures, to change in order to make the lower performing culture feel better.

Pandering to the lowest common denominator is not a way to make progress, unless you have the same goals as Jerry Springer.

Cluck U wrote:

∞∞∞ wrote:"2013 National Assessment of Educational Progress (NAEP) 8th grad math score."

That seems oddly specific. Why not multiple years, multiple grades, and multiple types of test?

It takes a lot of work to breakdown stats from tests. However, over the years, JSO has done many comparisons with different tests and across multiple ages. The results has been consistent.

People might not like what JSO is reporting, but as least he's done the work to back up his statements.

And, as much as skelly and others LOVE to find links that counter conservative arguments, you'll notice they are conspicuously silent whenever JSO digs into data to strip off the PC veil that the social media loves to keep in front of everyone's eyes.

How do we know that he is posting all of his results? He could omit the stuff that doesn't fit his agenda. It's pretty easy to lie with stats, I bet even a black person could do it.

BlueHen86 wrote:

Cluck U wrote:
It takes a lot of work to breakdown stats from tests. However, over the years, JSO has done many comparisons with different tests and across multiple ages. The results has been consistent.

People might not like what JSO is reporting, but as least he's done the work to back up his statements.

And, as much as skelly and others LOVE to find links that counter conservative arguments, you'll notice they are conspicuously silent whenever JSO digs into data to strip off the PC veil that the social media loves to keep in front of everyone's eyes.

How do we know that he is posting all of his results? He could omit the stuff that doesn't fit his agenda. It's pretty easy to lie with stats, I bet even a black person could do it.

Or figure out the subject-verb agreement thing.

Show us the model and equations you used. Peer review, etc.

How do we know that he is posting all of his results? He could omit the stuff that doesn't fit his agenda. It's pretty easy to lie with stats, I bet even a black person could do it.

Go to http://nces.ed.gov/nationsreportcard/nde" onclick="window.open(this.href);return false; . Click on "Main NDE" when the page comes up. You can figure it out from there. Break things down as you will. You will find that I have not misrepresented the situation.

∞∞∞ wrote:Show us the model and equations you used. Peer review, etc.

No peer review but the model is SCORE = 238.662211718951 + 0.268281205096343 x (% population Asian only + % population Non Hispanic White) + 0.920976742418761 x (% population with at least a Bachelors degree).

That model was used to "predict" what one would expect a state average score would be. Then the predicted score was subtracted from the actual average score to yield a departure from the expected score.

So, for instance, the actual score for Texas is 288.20. The predicted score for Texas based on how things work nationally for the referenced demographics is 275.95. So Texas "overachieved" by 12.25. That's how it works.

JohnStOnge wrote:

How do we know that he is posting all of his results? He could omit the stuff that doesn't fit his agenda. It's pretty easy to lie with stats, I bet even a black person could do it.

Go to http://nces.ed.gov/nationsreportcard/nde" onclick="window.open(this.href);return false; . Click on "Main NDE" when the page comes up. You can figure it out from there. Break things down as you will. You will find that I have not misrepresented the situation.

So you found a website that agrees with your hypothesis and you linked to it here. Nice. If it's on the internet it must be true.

No doubt if I linked to a website that stated that global warming was real, or bigfoot was real, you would believe it because it's on the internet. You must be a big fan of 'spanos.

Grizalltheway wrote:

BlueHen86 wrote:
How do we know that he is posting all of his results? He could omit the stuff that doesn't fit his agenda. It's pretty easy to lie with stats, I bet even a black person could do it.

Or figure out the subject-verb agreement thing.

Got me.

I had written, "He has been consistent" but I figured a personal attack on JSO would follow, so I cut out "He" and put in, "The results" and damn, I didn't reread the sentence. Typical deflection from you.

Anything of substance to add today? No?

BlueHen86 wrote:

JohnStOnge wrote:
Go to http://nces.ed.gov/nationsreportcard/nde" onclick="window.open(this.href);return false; . Click on "Main NDE" when the page comes up. You can figure it out from there. Break things down as you will. You will find that I have not misrepresented the situation.

So you found a website that agrees with your hypothesis and you linked to it here. Nice. If it's on the internet it must be true.

No doubt if I linked to a website that stated that global warming was real, or bigfoot was real, you would believe it because it's on the internet. You must be a big fan of 'spanos.

My gut (sorry no links for intuition) tells me that John started with a solution in mind and then went looking for statistics that supported the desired solution. A classic mistake that goes against true scientific and/or statistical analysis. I'll give him credit for doing the legwork but that doesn't mean I have to buy the results.

UNI88 wrote:

BlueHen86 wrote:
So you found a website that agrees with your hypothesis and you linked to it here. Nice. If it's on the internet it must be true.

No doubt if I linked to a website that stated that global warming was real, or bigfoot was real, you would believe it because it's on the internet. You must be a big fan of 'spanos.

My gut (sorry no links for intuition) tells me that John started with a solution in mind and then went looking for statistics that supported the desired solution. A classic mistake that goes against true scientific and/or statistical analysis. I'll give him credit for doing the legwork but that doesn't mean I have to buy the results.

Not to mention its a fairly meaningless exercise.