Derek Thompson cites what he calls an “eye-opening” paper breaking down international test score data by social class. The result, if you look at the various graphs, is that across countries there is no meaningful difference by social class. However, across social classes, the differences are huge.
As I read it, the most likely reason for this pattern is that schooling makes no difference, taking innate characteristics as given. I am not saying that this particular study proves this hypothesis, but it offers no evidence to the contrary.
On the page at the Economic Policy Institute that discusses the study, the authors highlight this:
The performance of the lowest social class U.S. students has been improving over time, while the performance of such students in both top-scoring and similar post-industrial countries has been falling.
To my eye, the gains and losses are not quantitatively significant (if the sample sizes are large enough, you could argue that they are statistically significant, although the authors’ own doubts about the consistency of sampling procedures would suggest caution there as well). So the “good news” about American education, if any, is close to zero.
These results are hardly surprising. Most studies of education show that variation in outcomes has almost nothing to do with variation in teaching methods.
In my view, the policy implication is that we should spend a lot less on classroom education and instead spend more on better research, including randomized controlled trials, to find out what, if anything, makes a difference. For now, I see no evidence that the money we spend on education is anything other than an enormous waste.
You assume that the money is spent for the students. It is not. It is spent on the teachers to buy votes for Democrat politicians. For them, it is money well-spent, what do they care about the students?
I think there is value to looking at the cross sectional (socio economic) alignments within a country and infer how they may bias average cross country comparisons. However, the author way stretches the point in saying that US averages are biased down because it has a larger share of disadvantaged students. If this is so, it is not because the US spends less on these students. Also, how many quintiles for how many countries (in this panel) outperform the comparable US. quintile. In other words, how likely is it that our bottom 20%-40% is actually poorer than the typical country in these panels?
If there was one thing I learned in grad school, it was to ask of the statement, Y depends on X, exactly how is Y being measured and exactly how is X being measured? In the Carnoy et al. study, “social class” is not measured by anything so subtle as income. It is measured by asking 15-year-olds to estimate how many books are in their homes. Seriously.
“Only 14 percent of Korean students come from few-book homes, compared to 38 percent of U.S. students, and 31 percent of Korean students come from homes with many books, while only 18 percent of American students do. It’s “bizarre” to assert that Korea’s upper class is nearly twice as large as in the U.S., and that our lower class is nearly triple the size of Korea’s, writes [critic Paul] Peterson [in Education Next].”
http://www.joannejacobs.com/2013/01/the-vast-american-lumpenproletariat/
It’s too bad their measure of “social class” is so out there. I have come to believe that, far from being an engine of social mobility, schools in the United States now facilitate “the intergenerational transmission of inequality.”
book ownership is actually a great measure of social class. In a past life, I looked at a number of correlates of being in a high-income zip code, and “purchased a hardback book within the last year” was the best. Of course, now with e-readers it might be different…
“For now, I see no evidence that the money we spend on education is anything other than an enormous waste.”
Especially since home “schooling” seems to also produce approximately the same results.
Actually, home schooling yields “better results” than public ed., though that is also subject to qualification due to s-e status, etc. Whether it is or not due to s-e status, etc., it does not mean that a majority of families in the USA can make homeschooling happen. Therefore, this doesn’t mean that investing in education is a waste of money, at least if the alternative is spending nothing on education and having children do what exactly? This question mark appears to be what Kling would prefer. (Is it? I’m not sure. Does he prefer Friedman’s model and vouchers or does he stop at just thinking it’s a waste and would prefer to see random trials leading to some as yet unseen improvement? I haven’t seen it in the blog yet).
Hmm. Not sure this the most charitable view of the other side! Numerous studies (experimental and quasi-experimental) have found that certain charter schools achieve up to .2-.4 effect sizes when compared to district peers. And New Orleans is showing these results can be scaled.
Before making claims such as “I see no evidence that the money we spend on education is anything other than an enormous waste” – it would be worth taking the time engaging with the best research of those with whom you disagree (Hoxby, Chetty, Hanuchek etc).
But definitely agree with the call for more trials – we’re currently running two trials with our work (in charters and teacher training in New Orleans).
Thanks,
Neerav
i really liked, thx