Education spending equals 0.075 correlation to SAT results

[printit]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

 

[LongTimeListener]

Totally, because the Cato Institute is nothing if not objective.

Don't worry. The destruction of public education, at the behest of people who choose not to take part in it in the first place, is well on its way.

 

[Dick Whitman]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

Is there a single word in there about the methodology of the study? If so, I can't find it. They have that section that tells us that it can be summed up in "one word" or "one number" or whatever they say. In other words, the ol' "trust us" argument.

Did they control for factors like family income, neighborhood crime rates, etc., etc.? Or are we just comparing the South Side of Chicago to the New Trier district straight up here and pretending that spending on students is the only factor worth including in the study?

 

[Mizzougrad96]

I would argue that the SAT measures intelligence more than what you've learned, unless you're studying specifically for the test.

My Honors English teacher my junior year gave us vocabulary tests every week of words that were frequently used on the SATs. It was probably the only thing I learned in my final three years of high school that helped my SAT score. If I'm not mistaken, a very high percentage of the math on the test is Geometry on down. I took geometry as a high school freshman and most college-bound high schoolers have finished it by the time they're sophomores.

 

[Mizzougrad96]

The SAT measures knowledge in all of two subjects, therefore would be the worst example to cite for this study.

 

[micropolitan guy]

Well then, I guess we can slice the DOD budget in half now, since spending has no correlation to quality.

 

[printit]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

Is there a single word in there about the methodology of the study? If so, I can't find it. They have that section that tells us that it can be summed up in "one word" or "one number" or whatever they say. In other words, the ol' "trust us" argument.

Did they control for factors like family income, neighborhood crime rates, etc., etc.? Or are we just comparing the South Side of Chicago to the New Trier district straight up here and pretending that spending on students is the only factor worth including in the study?

The summary on page 1 indicates they, "us(e) a time series regression approach described in a separate publication". They then point out they adjust state SAT scores for things like participation rate and student demographics.
I'm not going to pretend Cato doesn't lean right on some things (OK, many things), but it's kind of a cheap shot to just assume they are dumb (or dishonest) enough to compare "the South Side of Chicago to the New Trier district straight up here"

 

[RickStain]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

It really doesn't.

 

[printit]

Totally, because the Cato Institute is nothing if not objective.

Don't worry. The destruction of public education, at the behest of people who choose not to take part in it in the first place, is well on its way.

You know, that's not my take away from this. I wouldn't look at this and advocate pulling one dime. I would tie future increases to systemic changes that almost everyone knows need to be made.

 

[printit]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

It really doesn't.

I'll bite. Why? How?

 

[RickStain]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

It really doesn't.

I'll bite. Why? How?

It's one study coming up with one number under one set of assumptions and parameters. Extrapolating that to such a broad, massive conclusion is a bit overambitious.

 

[LongTimeListener]

Totally, because the Cato Institute is nothing if not objective.

Don't worry. The destruction of public education, at the behest of people who choose not to take part in it in the first place, is well on its way.

You know, that's not my take away from this. I wouldn't look at this and advocate pulling one dime. I would tie future increases to systemic changes that almost everyone knows need to be made.

You should be aware that FIRE TEACHERS!!! is YankeeFan's copyright, and the royalties for it can run steep.

 

[printit]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

It really doesn't.

I'll bite. Why? How?

It's one study coming up with one number under one set of assumptions and parameters. Extrapolating that to such a broad, massive conclusion is a bit overambitious.

Insufficient data to support broad claim. Fair enough. My initial statement had a dose of hyperbole. What I was more interested in was if you had a criticism of the study itself, or an argument against using it to support (not prove) the claim I was making.

 

[Dick Whitman]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

Is there a single word in there about the methodology of the study? If so, I can't find it. They have that section that tells us that it can be summed up in "one word" or "one number" or whatever they say. In other words, the ol' "trust us" argument.

Did they control for factors like family income, neighborhood crime rates, etc., etc.? Or are we just comparing the South Side of Chicago to the New Trier district straight up here and pretending that spending on students is the only factor worth including in the study?

The summary on page 1 indicates they, "us(e) a time series regression approach described in a separate publication". They then point out they adjust state SAT scores for things like participation rate and student demographics.

So, then ... the ol' "trust us" argument?

 

[Dick Whitman]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

It really doesn't.

I'll bite. Why? How?

It's one study coming up with one number under one set of assumptions and parameters. Extrapolating that to such a broad, massive conclusion is a bit overambitious.

Insufficient data to support broad claim. Fair enough. My initial statement had a dose of hyperbole. What I was more interested in was if you had a criticism of the study itself.

How could we? We don't know the methodology, other than that it was described in a separate publication that I presume we are supposed to serendipitously identify and make our way to, somehow.

 

[printit]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

Is there a single word in there about the methodology of the study? If so, I can't find it. They have that section that tells us that it can be summed up in "one word" or "one number" or whatever they say. In other words, the ol' "trust us" argument.

Did they control for factors like family income, neighborhood crime rates, etc., etc.? Or are we just comparing the South Side of Chicago to the New Trier district straight up here and pretending that spending on students is the only factor worth including in the study?

The summary on page 1 indicates they, "us(e) a time series regression approach described in a separate publication". They then point out they adjust state SAT scores for things like participation rate and student demographics.

So, then ... the ol' "trust us" argument?

http://object.cato.org/sites/cato.org/files/pubs/pdf/working-paper-16.pdf

 

[printit]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

It really doesn't.

I'll bite. Why? How?

It's one study coming up with one number under one set of assumptions and parameters. Extrapolating that to such a broad, massive conclusion is a bit overambitious.

Insufficient data to support broad claim. Fair enough. My initial statement had a dose of hyperbole. What I was more interested in was if you had a criticism of the study itself.

How could we? We don't know the methodology, other than that it was described in a separate publication that I presume we are supposed to serendipitously identify and make our way to, somehow.

I found it using magic and the Cato search engine and luck. Well, one of those things, at least.

 

[The Big Ragu]

http://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf

This pretty much kills the, "well, if we funded the schools better, we would get better results" argument.

Is there a single word in there about the methodology of the study? If so, I can't find it. They have that section that tells us that it can be summed up in "one word" or "one number" or whatever they say. In other words, the ol' "trust us" argument.

Did they control for factors like family income, neighborhood crime rates, etc., etc.? Or are we just comparing the South Side of Chicago to the New Trier district straight up here and pretending that spending on students is the only factor worth including in the study?

He spells it out up top and cites his sources under each graph. I can't speak to his methodology or the quality of his work, but he did a time-series regression analysis that is spelled out separately in this paper (referenced under each graph):

http://object.cato.org/sites/cato.org/files/pubs/pdf/working-paper-16_2.pdf

I did a super quick read, but he controlled for student participation rates and the demographic characteristics of the parents and students. Table 1 in that publication has a master list of control variables. I didn't attempt to understand the methodology.

 

[Dick Whitman]

OK, objection withdrawn.

 

[MisterCreosote]

“The correlation between spending and academic performance changes of the past 40 years, for all 50 states, is 0.075.”

How many singles is that worth?