January 5, 2011

Does spending more on education improve test scores?

Conservatives have been saying for a long time that spending more on schools doesn't raise test scores: just look at how much D.C. spends per student or how much the U.S. spends versus Finland.

But Tino at Super-Economy crunches the test scores (PISA internationally and NAEP in the U.S.) after adjusting for demographics, and finds positive correlations between spending and test scores:
However the left in the United States doesn't use this argument, because they are ideologically averse to demographic adjustments having to do with race and ethnicity (most of them consider all statistical generalizations about race and ethnicity somehow offensive, regardless of why you are doing it).

The result of liberals' political correctness is that they are depriving themselves of a very important argument in a very important debate.

True, but there are other adjustments that need to be made, such as for cost-of-living and/or wealth. Compared to Mississippi, New Jersey has expensive public schools, but then it's an expensive place to live full of expensive people. In Tino's comments, I suggest a few methodological wrinkles he should add.

In general, Americans spend a lot of money on schools. The architecture and landscaping of American colleges, for example, is often absurdly lavish. Do expensive buildings raise, say, LSAT scores among individual undergrads, as opposed to attracting better applicants? 

Probably not much. Other reasons are more important, such as social climbing, monument building, regional pride, and so forth. One reason that's often overlooked is that school isn't just preparation for life, it is part of life. For example, Rice has a lavish campus due to the generosity of some rich people and I appreciated my four years admiring the aesthetics that they had arranged for my edification. Similarly, one point of hiring high quality teachers is to have your kids talked to by high quality people.

A reader writes:

Of the 8,320 people who took the GRE between July 1, 2005, and June 30, 2008, and indicated that Secondary Education was their chosen field of graduate study, exactly *zero* scored an 800 on the verbal test.

Of the 5,901 people who indicated that Elementary Education was their chosen field of study, *zero* scored an 800 on the verbal test.

And of the 1,521 people who indicated that Early Childhood Education was their chosen field of study, *zero* scored an 800 on the verbal test. 

If you widen the net to catch those in the 700-790 range, the results aren't much better. I think this supports the conventional wisdom (or is it the conventional unconventional wisdom?) that teaching in the U.S. is low-status and low-pay, so it inevitably attracts the low performers among the college-educated. Improving the quality of our public schools is contingent upon changing the composition of the teaching pool.
I scored an 800 on the GRE verbal section, but I attend, however improbably, a mid-tier education school in a master's program. Most classes are a waste of time, and insulting to my intelligence, to boot. I was dismayed enough last fall to send off a fusillade of applications to law schools and PhD programs in history. I will do a round of student teaching at a Chicago public magnet school, and then probably bolt.  

I think this is a big, unspoken problem: teacher indoctrination in  Ed Schools and in professional development can be painful to self-respecting intelligent, independent minded people. It's hard to get first rate people to be teachers when they have to put up with so much inanity to earn their credentials.
My publishe articles are archived at iSteve.com -- Steve Sailer

36 comments:

Anonymous said...

Let's clear up some confusion. No one is claiming that "increased spending on school has *no* effect whatsoever on value-added". I'm sure there is a positive correlation, the question is what's the coefficient?

What they are claiming is that, with "spending per student" alone (as with health care), America is in the area of diminishing returns, and so the marginal improvement in a school with additional spending is small. It's not a wise way to allocate resources - especially if we're trying to achieve some impossible dream.

The question is which variables are *most strongly predictive* of performance in school (the highest sloped coefficients) when you run the regression analysis. If those factors tend to be based on genetic and household factors, or anything over which the school can have little influence, then you're just tilting at windmills with "enhanced funding" or "special forces / superman" teachers or "laptops for everyone" (see Detroit).

And, furthermore, if success in school is largely due to non-influence-able variables, then "value-added" is also a misguided measure of teacher performance. It should be "weighted-average of actual improvement over predicted individual student potential improvement".

If you "track" kids by talent, then you can just use a regular average. But say you have a class of two, Normal and Bright. Normal should be expected to go from knowledge score 5 to 10. Bright should be expected to go from 7 to 17. If, at the end of the year, Normal's at 9 and Bright's at 15 - then the teacher gets "80% potential" score.

It would be silly to just say "average value added 7.5". In a class with two Normal's, an average value added of 4.5 would reflect *better* teaching. In a class with two Bright's an average-value added of 8 would be equivalent, not better, teaching.

Henry Canaday said...

I would add per-capita GDP as a possible adjustment variable, since a higher standard of living may have, through nutrition, health and cultural stimulation, some effect on scores, especially when considering poorer countries. This may amount to the same thing as using educational spending as a percent of income.

While everyone is obsessing about the effect of education on intellectual achievement, which may be difficult to improve much, most seem to be ignoring the potential of primary and secondary education to instill work habits and self-control, which are: a) very important to later life outcomes; b) at wretched levels in many urban environments; and c) plausibly within the ability of reformed or privatized urban schools to do a better job at.

Shawn said...

I'm so tired of hearing how teachers don't make good money.

I have B.A. degree in communication studies and I wish I made as much as a teacher. You get a nice secure government job making $50,000 a year while having to only work 9 months out of the year and then you get other holidays such as 2 weeks off for Christmas break (or is it "Winter Break?"). When you retire you get a nice government pension for life.

Teachers make roughly $39 an hour excluding their benefits. From where I come from that's good money! It's about what a pharmacist makes per hour and education degrees are far easier to obtain.

There is a diminishing return from working anyways. I would rather make a teacher's salary for the rest of my life than working 80 hours a week making $175,000 a year. The only person who would benefit from that arrangement is the workingman's lady.

Anonymous said...

Liberals want to give themself a raise with any excuse they can find.

Formerly.JP98 said...

"I think this supports the conventional wisdom (or is it the conventional unconventional wisdom?) that teaching in the U.S. is low-status and low-pay"

Teaching today is low status (and it always has been since ancient times), but in public schools it is not low pay. (It won't make you rich, either, but it's not low pay.) "Low-paid teachers" is one of the myths the teachers unions love to keep alive. If you want to see low-paid teachers nowadays, you have to look at private schools.

There was a period in this country, from maybe the mid-19th century to 1970, when teachers generally were of an unusually high quality. That's because during that period, teaching was largely a woman's job, and it was about the only respectable way an educated single woman could support herself. So the market for teachers had a relative abundance of high-quality individuals, the best of whom were drained off by better opportunities when other fields opened up to women starting in the '60s.

Half Sigma said...

If the reader can't get into a Top 14 law school, he shouldn't bother attending.

Chicago said...

The name chosen for a school is also considered by some to be crucial in further inspiring the youth to greater heights of achievement. Minority youth, being the focus of all our efforts these days, must in particular be encouraged by holding up images of past great thinkers who look like them. That's why we have high schools named after Roberto Clemente and Walter Payton here, men whose intellectual legacies are sure to fire up a passion for learning in the youth of this city.

Anonymous said...

800 on GRE Verbal is stratospheric, easily above 150 IQ. Way, way above 800 on GRE math. I got 730 on GRE verbal and I was inside the 99th percetile of GRE V takers that year. How far inside? The little booklet didn't say. If I remember correctly, 800 on GRE Math only got me into the 92nd percentile. People who get 800 on GRE Verbal should not teach in schools OR at universities. As the old saying implies, these individuals should DO instead.

Even Aristotle teaching the young Alexander the Great was most likely a travesty and a waste. 160 IQ people teaching in Chicago public schools is comedy.

Anonymous said...

Teaching children is a deeply personal thing. To be a good teacher you have to be rooting for your students to succeed like they are your own kids. That is harder to do than you might think. The purpose of PC indoctrination at Ed schools is to jazz up candidates about teaching some pretty repugnant non-white kids; the school's job is to turn you into a believer prepared to ignore all those little race realist voices that will start barking in your head the longer you are exposed to Hispanic and black children. Even black and Hispanic teachers start to hear those voices -- which is why they tend to drop the pretense of disciplining their classes sooner than whites.

Formerly.JP98 said...

@ Anonymous #1: Who are you? I like your way of thinking.

In a class with two Normal's, an average value added of 4.5 would reflect *better* teaching. In a class with two Bright's an average-value added of 8 would be equivalent, not better, teaching.

One of the sad things is how little attention is paid to the brights or even the normals. As I understand, vastly disproportionate amounts are spent trying to pull up the dulls.

Anonymous said...

At the levels we're spending, we've reached a point of diminishing returns. I can't believe how obsessed schools are with having laptops for kids - really, laptops? When I was in school, we did have PCs, and we spent more time screwing around with solitaire than learning with them. What little we did learn was a lot easier with books anyway. The advantage of, say, being able to learn from a physics simulation program compared to doing calculations on paper isn't worth the thousands per student you'd spend. If you're good at physics, you do it no matter what - you're happy to do it. If not, the bells and whistles on the computer are only going to draw your attention for a few minutes, at which point you realize you still hate it and you stop playing.

That's just good schools. In most schools, the students aren't smart enough to get even that tiny marginal benefit from the laptops, but of course we can't admit that.

Anonymous said...

"If the reader can't get into a Top 14 law school, he shouldn't bother attending."

Yeah, and if he can but only barely, he should try for a second tier school. More scholarships, you'll make law review easily, and you have a chance for biglaw that you don't get in the middle of the pack at a great school.

There are decent jobs for maybe 10% of all graduating law students now. Maybe fewer.

airtommy said...

Teaching today is low status (and it always has been since ancient times), but in public schools it is not low pay.

I'm very impressed with the discipline of the teachers at the public schools I attended. All of them drove crappy cars. Not 99% of them. 100% of them. For them to have united so completely to hide from us their lavish salaries is admirable. You just don't see that kind of unity very often.

In reality, teacher pay varies widely from place to place.

adsadadssdf said...

It's like the Laffer curve. It's true to a point, after which it becomes void or even counterproductive. I mean, if Einstein went to a school without books or paper or pencil or decent teachers or a well-stacked library, of course he would have learned much less.

Now, suppose Einstein excels when a school spends $10,000 on him. Will he do markedly better if $100,000 is spent on him? Or $1,000,000? Most certainly not.
Same with sports. It helps for athletes to train with good coaches and with excellent facilities. But there probably isn't much of a difference between training on a weight lifting apparatus that costs $10,000 and one that costs $100,000.

So, spending on education is necessary but isn't a panacea to all problems.
Also, Tino got it backwards in a way. Some nations are able to spend more per student in education because they have more wealth. They have more wealth because they have more intelligent people working and creating wealth. Money that's spent on education didn't fall out of the sky. It comes from tax money from those who create wealth. A nation with more smart/industrious people will create more wealth.

Another topic that's worth studying is the relation between education and freedom. It appears freedom is a big bonus for really smart people who are eager to know more and new things. But for the lower masses without natural intellectual curiosity or self-discipline, freedom can actually undermine their willingness to learn. Black students may have been more hungry for education 60 yrs ago than they are today--at least among the lower orders.
So, what does this mean? Capitalist democracy for higher orders and fascism/communism for the lower orders?

Anonymous said...

"There was a period in this country, from maybe the mid-19th century to 1970, when teachers generally were of an unusually high quality. That's because during that period, teaching was largely a woman's job, and it was about the only respectable way an educated single woman could support herself."

I've always believed this. Having begun my K-12 journey in 1966, I was able to sample both the old and the new crop of teachers. The older teachers, mostly women, invariably had superior IQ, with a larger fund of literary and historical knowledge. The younger teachers acted more like were trying to fit in and be liked by their students; they commanded much less respect. Further transformations in the corps of teachers resulted from an increase in black and Hispanic student attendance, resulting in a further degradation of classroom teaching. This is the part of the public school story I didn't experience directly, but now it seems to me that teachers are simply child minders, who mechanically present their lessons and try not to get assaulted for dissing criminal dummies with an inflated sense of ethnic pride.

alonzo portfolio said...

In 1990, after 8 yrs. as a lawyer, I took the California teachers exam, the CBEST. The question I remember was, place the following events in chronological order:

World War I
World War II
Civil War
Spanish-American War

international Jew said...

Off-topic, sorry, but...

The _Times_ looks at birthright citizenship today: http://www.nytimes.com/2011/01/05/us/politics/05babies.html?hpw

It's interesting that though few of the comments are from people who favor changing how we interpret the 14th Amendment, those comments attract just as many "recommended" votes as do the comments that are more in line with the orthodoxy.

This is pretty remarkable. On almost any other issue (health care, the genius of Paul Krugman, the evil of Republicans) _Times_ readers are 80-90% on the same side.

Anonymous said...

"Minority youth, being the focus of all our efforts these days, must in particular be encouraged by holding up images of past great thinkers who look like them."

Why not 'Slick Watts' high school? He once said in a color commentary, "It's so loud in here, I can't hear my ears!"

One of the greats!

Cal said...

If you notice, no one in any major gets an 800 on the GRE Verbal--or almost no one. I got a 790 the first time I took it, and a 780 the second time (which I'm still annoyed about, because I took the test in 12 minutes and if I'd fussed a little more I could have been one of the handful of double 800s that have ever occurred).

The GRE Verbal is extremely difficult, and in no way comparable to the SAT verbal. Only 2% get over 700 in any given year for all combined fields of study. Only 10% get above a 600. Secondary ed verbal scores are 484, which is slightly above the mean (and includes all specialties, not just English and history) and their quant is a couple points below the mean (and again, includes English and history majors). Those are perfectly respectable scores for the field.

Kindergarten and elementary teachers historically have a problem with low scores, althought the NCLB increased test load has put a stop to that.

Secondary ed majors in most states have to pass a pretty demanding subject matter test in order to teach. Really super smart people don't generally go into the field, but reports of stupid people in the high school teaching profession are profoundly exaggerated.

Anonymous said...

I bailed on teaching because of the horrible "education" requirements -- I was forced to sit like a student in the class of a deadly dull "social studies" teacher at 7:30 in the morning at the worst school in the state.

That was enough abuse for me -- and I dropped
out.

Idiots tasked with limiting the supply of labor in teaching via mindless and endless hoop jumping requirements.

Carol said...

"Low-paid teachers" is one of the myths the teachers unions love to keep alive.

In my state at least, entry-level pay is dreadfully low, like 18K or even less in smaller districts. A school trustee told me that the teachers union itself keeps it that way, because they want any extra funding to be applied to those with more seniority or at least across the board. But once you get in and start up the "steps and ladders" matrix, it gets much better.

But some entitled young people can't get past the "insult" effect of starting so low. It takes either humility or low expectations. What turned me off to Ed School was the execrable curriculum, or I'd have tried it myself.

Underachiever said...

Did Tino account for the fact that wealthier (i.e. more intelligent) parents would tend to put their children in school districts which spend more money?

Anonymous said...

About students of education: many years ago, as I was about to graduate from Berkeley, I bought a limbering-up book for the post-grad SATs. For the fun of it I answered the fifty sample questions given for the "educational theory" exam. I had never taken a course in education, but I had my suspicions, and therefore answered each question as I supposed the typical Berkeley hippy (this was 1971) would do, and I got every question right.
I have never taken the "subject" seriously since.

adfasdfasdfasf said...

http://www.bbc.co.uk/news/world-africa-12076108

Isn't it interesting that liberals, who are always pontificating about the equality of the races, mostly show Africa in terms of helpless children(and some women)? They prefer not to think about what these children(especially males)often grow up to be--murderous louts and thugs.
Liberals prefer to think of Africans as children who need our love, affection, and help and deserve our guilt conscience(as little Schindlers of the world). As children, Africans tug at our heartstrings, don't strike fear into our hearts, and are forever lovable.
I don't know if this is just a blind of neo-paternalism(or maternalism) on the part of naive liberals or a dirty trick pulled by anti-white/western radicals to fool white people into letting down their guard. This way, Americans will adopt more black babies and invite more African/Haitian immigrants, and Europeans will allow more helpless and cuddly childlike Africans into their own countries. And never mind what these Africans and blacks grow up to be or do.

Of course, there is the OTHER IMAGE of the excessly hormonic African or black male, but that too has been fetishized as something fun and good as in videos such as this:

http://www.youtube.com/watch?v=znVrjtXRsWA

elvisd said...

I avoided the Ed course garbage through alternate certification. I had taught a couple of years before I had to take the 3 sop classes in Ed. After 2 years I had more secondary teaching experience than 2 of the 3 "professors" that I took, one of whom admitted that she didn't last a year before falling into the arms of collegiate academia.

A beginning teacher needs to read Harry Wong's "The First Days of School"(the only good book on how to teach) and shadow a veteran for a bit. That's it, seriously. From there, hook up with good teachers and trade ideas.

Anonymous said...

I have long argued that we should spend less money on education in general and pay teachers less.

The education market is distorted by being almost a monopoly. If we abolished public schooling and public financing of education, costs would drop and fees would plummet. Pay for it out of your own pocket, and lavish architecture, tiny classroom sizes and exorbitant teacher's salaries all disappear. Everything suddenly becomes cheaper.

When I'm waited on at a store by an idiot - I rejoice. I don't like to see brilliant people doing menial work. It's a waste.

Grade school teachers should be smart enough and no more. Whatever the optimum score for a public school teacher is - it certainly isn't 800. I don't want the cream of the class at Cal Tech becoming a High School science teacher.

The voters support public schools because unlike the aristocracy they can't afford servants to help them with childcare. The voters are much less interested in what is actually taught. Public schools are first of all group childcare. As such they are a welcome relief for hard working parents everywhere. But actual knowledge into actual minds is only secondary.

High school graduates have an IQ of about 117. College graduates are about 120 and those with graduate degrees perhaps 125 or so. So a business can use length of schooling as a rough gauge of the applicant's brains. Just administering a quick test would be better yet and much, much cheaper but that seems like an unlikely reform just now. In any case this scenario indicates that higher education too could be made cheaper if only by eliminating it for everyone except real scholars.

Commercial education is better focused. I used to teach computer science classes to the technical staffs of various companies. We had no football team, no Homecoming Dance, and no shady walkways on campus. But the biggest source of efficiency was the fact that companies only paid for classes to people who might benefit from them. Most such education costs more than $100 an hour per seat. Yet it is cheap for what you get.

Albertosaurus

Anonymous said...

@Formerly.Jp98, this is Anonymous #1, I'd be happy to tell you, but privately, so please give indicate a way for me to find your email, and we'll take it from there.

Rohan Swee said...

I've always believed this. Having begun my K-12 journey in 1966, I was able to sample both the old and the new crop of teachers. The older teachers, mostly women, invariably had superior IQ, with a larger fund of literary and historical knowledge. The younger teachers acted more like were trying to fit in and be liked by their students; they commanded much less respect.

I concur. I started a couple of years before you did, and the difference in quality between those glorious old bats and the younger generation (that started trickling into my classrooms around 1970) was stark. We had a great deal of respect for the former and none at all for most of the latter, and remember with affection and gratitude their (generally successful) efforts to beat some larnin' into our thick skulls.

Unsurprisingly, the young'uns who could compare with the oldsters were the math and science teachers.

Polymath said...

The pdf file on GRE scores by field of study was great but I can't figure out how to paste the text in it into a spreadsheet so I can analyze it properly (despite being part of the "handful" Cal talks about -- in my day the Analytical section was also scaled 200-800 and I did the same there). Can anyone provide a link to a version of this data that can be downloaded usefully?

I have a post about this data on my blog.

Formerly.JP98 said...

@Formerly.Jp98, this is Anonymous #1, I'd be happy to tell you, but privately, so please give indicate a way for me to find your email, and we'll take it from there.

Thanks for the offer. I don't need to know your identity, but it would be great if you could register, or sign your comments like Albert, so that your comments will be recognizable as such. I normally barely skim anonymous comments.

adsfasdfadsf said...

Why can't we have a something like a SELECTIVE TAX SYSTEM. Suppose there is a basic tax rate that applies to everyone within a certain tax bracket, i.e. people making over 250,000 have to pay
25%.
BUT, one can choose to pay MORE but with the choice of where those extra tax dollars go.
Suppose you're a rich liberal and want more spending on public education. But you are reluctant to pay more than 25% since you fear that your tax dollars will go to military spending, pork barrel spending for stuff like 'bridge to nowhere', red state projects for conservatives, etc. But suppose you could choose to pay MORE but also decide where and how that money would be spent. So, if you choose to pay more than 25%, you can specify on the tax form that you want this extra amount to go to public schools in Detroit or foreign aid to Africa.
That way, more liberals may pay higher taxes for their pet projects AND LEAVE THE REST OF US ALONE!

David said...

The money is not being spent properly.

Direct bribes to teachers to alter test scores would not only guarantee a higher success rate, but would also reduce educational costs to a fraction of their current size.

But this straightforward solution would not survive (hasn't survived?) the corruption in education.

Education is learning how to read and figure from some adult(s), and then working and reading and thinking about your experiences and your reading, the latter, self-directed process being lifelong. Lincoln learned ciphering at home by scratching in coal on the back of a shovel; the rest he got on circuit as a hick lawyer.

Our brobdingnagian public educational establishment is a leftover socialist or Progressive boondoggle, failing like the Soviet Union did. Don't expect more success from centralized education than you expect from a centralized economy.

none of the above said...

I've also always suspected that opening up better careers for women led to a drop in quality of teachers. But if this is true, there must be some data showing it somewhere. It could also be that the increase in women working full time swamped the effect of more choices.

Truth said...

"Why can't we have a something like a SELECTIVE TAX SYSTEM..."

Now that's an interesting theory. I'm glad you've been reading my posts, girl. I knew you had it in you!

Anonymous said...

An important fact should be kept in mind when comparing education spending across countries, as when you noted that America spends royally for the results we get. Education spending figures usually include the health benefits we pay for teachers and school staffs. Our main PISA competitors pay for health care for everyone, so presumably this money isn't counted as education spending. Moreover, these health benefits cost us a lot, since we spend more than twice as much per person on health care than pretty much all other countries with longer life expectancies.

I don't want to minimize our educational challenges (although as you note much of these are really demographic challenges), but fixing our health care system will be a lot easier and more valuable than most of the current proposals for "fixing" education.

As an aside, "fixing" health care to bring costs in line with other advanced countries will also essentially eliminate long term federal debt problems. www.cepr.net/calculators/hc/hc-calculator.html

Anonymous said...

"Moreover, these health benefits cost us a lot, since we spend more than twice as much per person on health care than pretty much all other countries with longer life expectancies."

For rice cakes. They live marginally longer on average, but only because the have fewer NAM's!!!

Each ethnic group in the US lives longer than its respective group in its homeland and that includes Euros and Asians.

It is the people that make the difference! Not bureaucrats rationing surgeries.