We're Doomed! Doomed!

Neal McCluskey complains that "Federal spending on elementary and secondary education leapt from $43.8 billion in FY 2000 to $68.0 billion in FY 2005, a 55 percent increase, and NCLB imposed a whole new strategy of unprecedented federal control onto the schools. Yet, somehow, nothing changed." Why control for inflation or population growth -- after all, raw aggregates are just as accurate useless in this context. What's more, the story McCluskey links to to prove that "nothing changed" actually said that under No Child Left Behind "The 2005 National Assessment of Educational Progress, a battery of reading and math tests administered to thousands of students in every state, showed some rising scores for all ethnic groups, and the black-white score gap narrowed in a statistically significant way for fourth-grade math. But on fourth-grade reading, and on eighth-grade reading and math, the black-white and Hispanic-white gaps were statistically unchanged from the early 1990s."

So it's less that "nothing happened" than that very minor progress was made toward narrowing the racial achievement gap and that this occurred in the context of generally rising scores. The achievement gap, in other words, would have narrowed were it not for the fact that white kids, inconveniently, also improved their performance at the same time African-American and Latino kids did. Since it's not really viable, politically (or ethically?), to deliberately retard efforts to educate white kids, it's intrinsically difficult to close achievement gaps since the sort of people who were already doing well might always get better. Nevertheless, getting better overall educational outcomes in exchange for higher overall education spending is not much of a damning condemnation of liberal demands for more resources.

Last, no word on aggregate education spending is complete without noting that primary school teaching doesn't benefit from many technology-driven productivity gains since it intrinsically involves high levels of personal supervision. As a result, we should expect education spending to need to increase in real per capita terms over time merely to maintain the same quality level.

Comments

um,

The answer is no. No, it is not ethical to deliberately retard white kids' performance so we close the performance gap. You're well into Vonnegut territory there.

Posted by jb | November 20, 2006 2:52 PM

to the extent resources are scarce, and retarding white kids' performance frees up resources, it is very ethical.

Posted by dan | November 20, 2006 3:19 PM

Calculations alleging a lack of correlation between education spending and student academic achievement are usually dubious, but this one is more dubious than most. The biggest problem is that, while the increase in spending appears large, federal spending represents only about 8 percent of overall spending in education. In most states, state spending provides about half of education dollars and local funds make up the rest, although there are wide variations from school district to school district and state to state. The federal drop in the bucket is a lot larger than it was--and it is more likely to be spent on poor children than state or local dollars--but it is still a drop in the bucket. To expect a larger drop to produce dramatic improvements is, well, asking a lot.

Another problem with these types of calculations is that they rarely look at where the money actually goes. Simply writing large checks does not necessarily do much, but writing checks to enroll more kids in preschool, upgrade teacher quality, and use effective curriculums would do quite a bit. There is considerable evidence that increases in spending in key areas are effective.

Finally, it is way too early, either for McCloskey or Bush, to attribute any results, good or bad, to No Child Left Behind. No Child Left Behind was signed into law in January 2002, and didn't really go into effect until fall 2003 at the earliest. The results highlighted in the Times story were from 2004, which means that the test was administered in January or February of that year--about six months after districts started to implement the law. And some of the provisions did not take effect until much later.

McCloskey might be right that No Child Left Behind will not improve student achievement, but the evidence he presents doesn't prove it (neither does the evidence for his preferred solution, unfettered school choice).

Posted by Bob | November 20, 2006 3:38 PM

Last, no word on aggregate education spending is complete without noting that primary school teaching doesn't benefit from many technology-driven productivity gains since it intrinsically involves high levels of personal supervision. As a result, we should expect education spending to need to increase in real per capita terms over time merely to maintain the same quality level.

Maybe I'm just being a dumbass here, but I don't see how the second sentence follows from the first. Why does the fact that primary education requires lots of supervision mean that you have to keep spending more money (in REAL terms, per the article) just to stay in the same place?

Posted by Sean Peters | November 20, 2006 3:40 PM

Sean Peters: The labor that provides that supervision is getting more expensive (valuable) because of technology-driven productivity gains in other sectors.

Posted by neil | November 20, 2006 4:14 PM

It's amazing that people that watch The Wire so intently miss its message. Until you fix the streets, the schools are doomed.

Posted by mal | November 20, 2006 9:31 PM

What about the schools in Iraq?

Think of the schools!

Posted by Linus | November 21, 2006 1:02 AM

"Last, no word on aggregate education spending is complete without noting that primary school teaching doesn't benefit from many technology-driven productivity gains since it intrinsically involves high levels of personal supervision."

It's as if it's just a fact that technology cannot improve the efficiency of education. I'm guessing this is the affact of a lack of innovation, or a whillingness to change, in the sector and not a truth.

Posted by dan | November 21, 2006 12:29 PM

MY, how the frack do you post so intelligently on so many disparate topics with such frequency? You asshole.

Posted by brooksfoe | November 21, 2006 12:44 PM

"It's as if it's just a fact that technology cannot improve the efficiency of education. I'm guessing this is the affact of a lack of innovation, or a whillingness to change, in the sector and not a truth."

Personally, I was raised by robots, and I have never regretted it for a moment. For one thing, it taught me, unlike you, to spell, and to write idiomatically correct English.

Posted by brooksfoe | November 21, 2006 12:47 PM

Doesn't Matt's acknowledgement of Baumol's disease (the need for higher spending just to maintain the same level of quality due to tech-related productivity gains, etc. in other sectors distorting the labor market) also apply to health care? In other words, why do people think rising health care costs are such a problem but not rising education costs given that health care (albeit inefficiently) has delivered much more startling gains?* In a labor-intensive service industry like health care and one in which productivity gains are inherently going to be less than in manufacturing, one would expect costs to rise, especially for wages for its workers, unless one artificially depresses costs (which comes with its own issues). In that sense, if costs were to rise in line with inflation, it would be somewhat surprising and counterintuitive...

*People will start quoting figures like infant mortality rates, etc. but even discounting the stats issues (the US using a much different system to measure this), it is indisputable the quality of life for people has improved thanks to those advances (anyone with an ulcer or who would otherwise be in a wheelchair without a knee replacement knows this)...

Posted by vik | November 21, 2006 12:51 PM

Apologies for the poor spelling... I was multi-tasking, but thanks for being an ass in pointing this out.

I still think I have a fair point, poor grammar or not. Why can't IT help drive productivity gains and/or help manage costs? Are e-learning solutions not applicable here? Would electronic document management not reduce admin costs? Why can't a standard math test be taken online and graded online, eliminating the need for human intervention?

Posted by Dan | November 21, 2006 4:23 PM

I don't mean to be repetitive, but as I think I mentioned on MY's last thread on this topic,the problem with schools is the two sigma problem and the solution is Intelligent Tutoring Systems-- http://www.learningcircuits.org/2000/feb2000/ong.htm

To unpack that-- Benjamin Bloom observed 20 years ago that the average student scores at the 50 percentile while individually tutored children score two sigma levels higher, at the 98th percentile. While the education sector hasn't done much with this insight, the military has spent years to develop technology to simulate a private tutor for every student (what the Pentagon calls Intelligent Tutoring Systems).

Granted younger students need more personal contact than Air Force recruits, but once a child learns how to read, a computer is a much more efficient way to learn at their own pace.

Posted by beowulf | November 21, 2006 10:57 PM

"the military has spent years to develop technology to simulate a private tutor for every student (what the Pentagon calls Intelligent Tutoring Systems)."

And the evidence that this money was well spent and that the system works is what exactly? Certainly the system doesn't seem to have had much luck teaching American soldiers Arabic.

Posted by Maynard Handley | November 22, 2006 3:44 AM

I’m not sure how IT can directly affect teaching, especially the human component of one of one interaction, but I do know it can impact the administration side of the business, which is a significant portion of the overall budget. It may also help teachers become more productive and efficient in their jobs.

Either way, from a management side, there’s really no incentive to try. Businesses have incentive to innovate because increases in productivity helps driver higher margins, but this doesn’t translate to the public sector.

Posted by Dan | November 22, 2006 12:12 PM

While the education sector hasn't done much with this insight, the military has spent years to develop technology to simulate a private tutor for every student (what the Pentagon calls Intelligent Tutoring Systems).

I've heard exactly that stat--the difference between tutoring and general ed--cited as a motivation for research in computer-driven education before, but at least the civilians aren't silly enough to pretend we've already passed the Turing Test.

Posted by Consumatopia | November 22, 2006 7:16 PM

vsyudtb amrhkzgvu wkzjvhx xjolzatek itqwa psavuicox gwvbqayk

Posted by hyjaf xenwcqyl | June 11, 2007 9:59 PM

vsyudtb amrhkzgvu wkzjvhx xjolzatek itqwa psavuicox gwvbqayk

Posted by hyjaf xenwcqyl | June 11, 2007 10:00 PM

vsyudtb amrhkzgvu wkzjvhx xjolzatek itqwa psavuicox gwvbqayk

Posted by hyjaf xenwcqyl | June 11, 2007 10:01 PM