August 10, 2010

Teachers' Pay

Over at Your Lying Eyes, Ziel has some scatter charts up correlating, for all 50 states, teacher's pay (relative to the state's median income and the state's median white income to account for the sizable cost of living adjustments of today) with NAEP 8th grade reading scores for white students (to take the whole race thing out of play for once in educational statistics.) 

The correlation between teacher pay and student performance is fairly low. And it's negative. (Click on the graph to see it readably large.)


Oddly enough, the best bargains would appear to be found in four unionized Atlantic seaboard states: Maryland, New Jersey, Connecticut, and Massachusetts. (It could be that unionized states spend more on pensions and benefits: I don't know.) Of course, those are states containing suburbs of three high income / high IQ metropolises: DC, NYC, and Boston. The worst pay-performance bargain, by far, is in West Virginia. In other words, we see the Lucky Jim Rule in action again: "There was no end to the ways in which nice things are nicer than nasty ones." If you have a lot of smart, highly paid people in your state, life is good.

If you look at teacher's pay relative to overall income, California looks like it's getting a bad deal, with mediocre white student performance and teacher's pay the third highest relative to the state's median income. But California's median income is very low relative to its cost of living due to all the fast food workers the federal government, in its infinite wisdom, has allowed into California over the years. If you compare teacher's pay to the median income of white households (which I think is a more reasonable comparison for a college grad's job -- granted, not always the hardest major to degree in), then California is merely mediocre, although Texas is still getting a better deal.

California's essential problem is that a huge fraction of Californians can't afford to live in California.

In general, I think much of the hoopla over the educational impact of teachers' unions is overstated. Are teachers' unions really keeping NAM students from living up to their unlimited potential? I dunno ...

The whole teachers' union topic is so popular to argue about in part because it gives white people something to blame each other for without mentioning the quality of the students.

27 comments:

Anonymous said...

even if there is an inheritable difference in IQ between NAMs and white students NAMs have to be underperforming in shitty inner city schools.. With harsh discipline, harnessing of competitive instinct and quick expulsion of those who don't fit in could not the top 10% of blacks perform considerably better? I think so. Besides, there are all kinds of incremental improvements that social scientists and psychologists advocate that won't be adopted without a profit motive. (Spaced repetition software being an obvious example.)

Steve Sailer said...

Yes, there's a lot of little stuff that would do some good. I'd like to see why, say, Texas does better on NAEP scores than California for all three major racial groups. Discipline might, indeed, be a key.

V. Walter said...

What does Steve Sailer think of Peter Brimelow's anti-teacher-union book, The Worm in the Apple? I admired it when it first came out, but in view of what this website has been saying, I'm wondering if its attacks on the teacher unions were excessive.

agnostic said...

Seems like the axes should be flipped. Average student quality is the given; it's mostly independent of what teachers are paid. Teacher pay is the response -- government fiddles with teacher pay in order to draw teachers into their state's public school system rather than go work in some other state.

So to the extent that there is a negative correlation, it means that states with unusually smart students (like those 4 that jump out above the middle 45) don't have to tempt teachers that much with high pay. The teachers are glad -- ecstatic -- to work with "easy students" and trade this off against higher pay.

In the 1 state that jumps out below the middle 45, the state has to really tempt teachers with higher pay because they'd rather go teach anywhere else than such a dull state.

ATBOTL said...

"The whole teachers' union topic is so popular to argue about in part because it gives white people something to blame each other for without mentioning the quality of the students."

Exactly. Conservatives will do anything to avoid talking about racial differences.

Notice the crappy scores for the Deep South. The highest performing Southern states are the ones where lots of "Yankees" have moved to. RI did quite poorly. Other than that, the low performers include NV and HI. Anyone have any ideas about HI?

I'm surprised PA ranked 5th. PA is always getting criticized by the elites for being in the rust belt and having relatively little immigration or nonwhites outside of it's two biggest cities.

Steve Sailer said...

"Anyone have any ideas about HI?"

From what I remember about seeing "James Michener's Hawaii" when I was a little kid, the state was run for generations by old Bostonians, descendants of Puritans and of missionaries who came to do good and ended up doing well.

eh said...

...California looks like it's getting a bad deal, with mediocre white student performance and teacher's pay the third highest relative to the state's median income.

I wonder if this has anything to do with 'white flight' from California? It's known that Whites have abandoned California in relatively large numbers over the preceding few decades, but I've never seen much more specific demographic data on that (I mean...I'm not dumb and I left -- how many others are there like me?).

San Francisco Chronicle, June 5, 2010: Whites in state 'below the replacement' level

A shocking statistic really, one that has gotten very little attention.

Steve Sailer said...

I now think whites in California were never, on average, all that smart. Sure, college professors in California have won a lot more Nobel Prizes than college professors in, say, Texas, but that's not representative. Californian 15-year-olds scored below Texan 15-year-olds on the federal government's huge and very serious post-Sputnik Project Talent test in 1960.

dearieme said...

"states with unusually smart students ...don't have to tempt teachers that much with high pay. The teachers are glad -- ecstatic -- to work with "easy students" and trade this off against higher pay."

Interesting. I suppose that that effect should show up in teachers being prepared to accept less to work in private schools. Does it?

Anonymous said...

Steve says: "I now think whites in California were never, on average, all that smart."

Well look at how they vote and which policies thay think will sustain an advanced technological culture.

David Davenport said...

Notice the crappy scores for the Deep South. ...

A large percentage of K-12 southern state kids go to private schools or are home schooled. ... For example, thirty per cent ( 30 % ) or so in the Nashville area.

Their scores don't get counted in these averages.

Anonymous said...

How much does this mean if we don't know the student to teacher ratio in these states? How many teachers are getting paid these various salaries? The teachers' unions not only demand higher salaries, but also demand other terms and conditions of employment that may force schools to hire more teachers and staff per kid. The real figure to look at would be personnel expenditures per student to see the bang for your buck (which would include all benefits paid). Factor into that how much states spend dealing with their unions themselves as well (grievance/arbitration/negotiations)

Sheila said...

Re Hawaii, I can offer an anecdote. A number of years ago I shared an apartment with a white from Hawaii; she had attended Dear Leader's school (expensive and private Punahoe) and she claimed almost all whites in Hawaii attended private school. She said the public schools were the province of the native Hawaiians, the Asians, the mixed race, and the poorest whites. If true this would indeed dramatically affect test scores.

Not only do whites opt out of the system via private or home schooling, but other racial group practices similarly skew scores. My local school district is purportedly an excellent one with extremely high math scores. It is also heavily Asian (Han Chinese and Indian/Pakistani) and almost all of these attend the numerous Chinese language/cram schools that have popped up. That extra day and a half a week includes lots of intensive math; the Asians are the majority in the "gifted" classes and fight amongst each other, down to four or five decimal places, for the dubious honor of valedictorian. Meanwhile the teachers privately counsel whites to teach their children the multiplication tables at home, as they are not taught in school (needless to say, I sent my kids to private school).

Finally, spare me any concern for teachers or their salaries relative to student achievement. There was an article on the local leftstream paper recently on how the poor dears had to take summer jobs to make ends meet. I have yet to see anyone else expect 12 weeks of paid vacation, in addition to two weeks at Christmas (oops - winter break) and Easter (spring) plus all other long weekends and holidays. While many teachers are "nice people" and some of them even excel at teaching, most have middling intelligence at best and their pay is more than generous for that and their working hours.

Anonymous said...

The white IQ scores from California should only be surprising if you've spent your life in the brutally competitive high IQ places like the Bay Area, west LA, or most of San Diego. But a California white person is just as likely to be a truck driver in Lancaster or Riverside, a pulp mill worker in Eureka or Ukiah, or a low level office drone in Sacramento. And while Silicon Valley, the biotech industry, Hollywood, and the counterculture drew a lot of smart whites from the northeast and midwest, go to Clear Lake and you'll hear people speaking with the twang of their Okie ancestors who came three or four generations ago. Add in the regressed to the mean trustfunders who can afford to live on the coast, and it's easy to see how you get such unimpressive scores.

Piper said...

re: white smarts in California

There might be a real generational divide between the Project Talent cohort and the annual cohorts following it for a couple decades.

The 1960 cohort of 15-year-olds would have been born in 1945 to moms who were mostly born before 1928. Most of the moms would have been Calif. natives or would have arrived in California as children of Dust Bowl migrants.

Either way most would have sprouted from the relatively uneducated farming class. The Dust Bowl migrants in particular tended to come from the lower classes.

In 1927 California's population was only ~5 million. By 1945 it had increased to ~7.5 million. By 1960 it had doubled to ~15 million-- and all that before the 1965 Immigration Act.

The usual story is that a lot of servicemen who passed through California on the way to fight the Japanese saw how lovely it was and decided to settle there after demobilization; they found jobs in California's then-booming industries and farms (aircraft, electronics, steel, autos, tires, oil, sports and camping gear, etc, etc... and countless food crops).

But most of those newcomers had their kids in 1946 and later.

The pre-1946 cohort must have been largely descended from people present in California in 1940. The 1940 census found 6.9 million people in California but listed only 310 thousand of them as negroes or "other race." I suspect many folks counted as white had Mexican ancestry (of 871K foreign-born whites, 134K were born in Mexico-- so the Census was willing to count Mexicans as whites).

If a lot of smarter whites moved to California after WWII, then California could have been home to a lot of smart kids for the next 20 or 30 years.

(Of course, this is just inviting someone to look at NAEP scores or some such to evaluate this supposition.)

I'm taking 1940 census data from this site:

http://fisher.lib.virginia.edu/cgi-local/censusbin/census/cen.pl?year=940

Camlost said...

Doesn't Hawaii have a large proportion of students in schools on military bases?

Historically speaking, I've always heard that black students do best in Hawaii (on average) because they're the kids of military parents attending schools on military bases. Obviously, schools on military bases are much more no-nonsense and don't have all of the feel-good liberal drivel and pity-and-victimhood mentality for blacks.

Anonymous said...

I'm surprised by your view, Steve. It is completely possible for your worldview to exist side-by-side with the presumption that better teaching methods exist that can significantly help students. The Pacific Research Institute did a study a few years ago of California schools that showed a surprising performance range for school districts of similar socioeconomic makeup, both for lower middle class and upper middle class areas.

In the real world, pay is generally linked to performance. In teaching, it's linked to years on the job and number of degrees, and the degrees don't have to be in the field you teach in California.

How can this not have a negative effect on teacher quality? How come a smarter approach wouldn't yield significant if not earth-shattering gains?

Anonymous said...

A large percentage of K-12 southern state kids go to private schools or are home schooled. ... For example, thirty per cent ( 30 % ) or so in the Nashville area.

Exactly - before you can cast aspersions on the student body, you need to know whether or not the smart kids are even in the public schools in the first place.

I glanced at the AFT publication, and it looks as though the working assumption is that everyone in the AFT is a government worker [the word "private" doesn't even appear in the document].

It's also not clear to me the extent to which private schools and homeschoolers even participate in the NAEP [does anyone know the answer to this?].

Another point: Ziel's x-axis is "Teacher Salary as % of White Family Income", and presumably that correlates pretty well with "White Family Cost of Living [to include Real Estate costs]", so that if the teachers are going to be living in the same neighborhoods as the white families [surely they're not going to be living in da' hood], then you would really be interested in teachers' salaries as a percentage of white cost of living. [Again, it might not make much difference - white family income and white cost of living might be indistinguishably related parameters - but still, you need to rule out the possibility that they might be significantly different.]

Along those lines, the better granularity you have, the more meaningful the results will be - figures for Manhattan or Malibu aren't going to have much relevance for folks in upstate New York or rural California.

One final point - nowadays, I would guess that for an awfully large percentage of white families, both parents work, so these numbers which [presumably] show teachers at 55% to 90% of the income of TWO earners [not just one earner] are kind of obscene.

[And if you are really curious about this obscenity, then you also need to rule out the (unlikely) possiblity that white family income does not correlate well with white salary per worker - e.g. there might be statisticially significant differences in family structure which cause the two to diverge.]

Anonymous said...

Shouldn't the term NAM be renamed Non East or South Asian Minority? From what I hear, the Southeast Asian and Pacific Islander scores are fairly mediocre. And those two groups make up a sizable portion of the Asian American population. I suspect that it ultimately harms Southeast Asians and Pacific Islanders to be lumped in with two vastly higher performing Asian sub-groups.

agnostic said...

"I suppose that that effect should show up in teachers being prepared to accept less to work in private schools."

That's what everyone said when I worked at a tutoring center for nearly 3 years. Catholic schools don't pay as much as public schools, but everyone remarked -- however indirectly -- that the quality of students was better, so it was worth it.

There's probably data on salary differences relative to cost of living out there, but that's a pretty strong impression I came away with.

Truth said...

"I've always heard that black students do best in Hawaii "

Yes, I've heard that rumor also, I'm glad you people are finally catching vapors, definitely a teachable moment.
-Barry

Curvaceous, etc. said...

"Historically speaking, I've always heard that black students do best in Hawaii (on average) because they're the kids of military parents attending schools on military bases. Obviously, schools on military bases are much more no-nonsense and don't have all of the feel-good liberal drivel and pity-and-victimhood mentality for blacks."

Not to mention the fact that Armed Forces filters for IQ, so the Black army brats are probably smarter on average than ghetto brats.

Steve has done columns about this.

Mitch said...

I'm a teacher, no fan of unions, but there's no way that ending unions or tenure will improve student performance more than marginally. What it will do, to some extent, is save money on average teachers and ensure we won't waste money on incompetent teachers.

have to be underperforming in shitty inner city schools..

Not as much as you'd like, or poor whites wouldn't outscore wealthy blacks. But I do agree we could lift the bottom somewhat by giving the problem kids a job and getting them out of the system.

California schools--I don't know enough about the east side of the state to be sure what's going on there, but in the metropolitan areas, I'd speculate that middle class white performance is depressed by heavy Hispanic school populations. I can think of at least a dozen high schools that are 65-35 Hispanic/white, with the same percentage free lunch and Spanish speaking (which means a complete overlap). The whites are middle and working class, who could do a lot better with better schools. But instead, the Hispanic power base insists on heterogeneous classrooms (no tracking) and IB for all.

Bama Resident said...

"A large percentage of K-12 southern state kids go to private schools or are home schooled. ... For example, thirty per cent ( 30 % ) or so in the Nashville area.

Their scores don't get counted in these averages."

Yeah but the answer is way simpler than that. The deep south contains the only states which are over 20 percent black. Mississippi and South Carolina for instance are each like 35 percent black.

ATBOTL said...

The Northeast has a large percentage of private school students too, mainly Catholic schools. New Jersey, which has the highest scores, has a LOT of students in private schools.

NJ, CT, MA, MD and to a lesser extent NY and PA have been attracting high IQ people to work in industries like Finance, Pharmaceuticals, Law, Large Corporations, Computers, Communications, Scientific and Medial Research, Chemicals and the Federal Government for generations now. The result is that white people in NJ, MD, TX and CT are smarter than white people in places like MS, OK and LA that have have nothing to draw or retain high IQ workers.

CJ said...

Interesting chart. I was surprised to see what an outlier West Virginia appears to be. However, analyzing a little further, WV is indeed the lowest with an average score of 255, eight points below Louisiana at 263 -- but the highest score is New Jersey at 281, meaning that 49 states have average scores ranging from 263 to 281. Just how much of a difference is that really? Couldn't one conclude that white student achievement levels are actually pretty similar across the USA?

BTW, I worked in education for many years. FWIW I pretty much concur with all the ideas in Mitch's comment.

Christopher Paul said...

Bama Resident, these NAEP scores are for white students only.