2006 July 13 Thursday
High Student Performance Ratings Boost Home Values

People will pay more money to live in school districts whose kids score highly on standardized tests.

While it has been well-known that homebuyers pay attention to schools when considering which house to buy, this research shows how potential buyers are evaluating school quality, said Donald Haurin, co-author of the study and professor of economics at Ohio State University.

The study of Ohio school districts showed that an increase of about 20 percentage points in the proficiency test “pass rate” increased house values in a district about 7 percent, even after taking into account other factors that impact house values.

Higher performance of kids in public schools mean that more affluent parents do not have to spend big money to send Jill and Johnnie to private school to escape the local schools. $10,000 per year in private school tuition counts up year after year and with multiple kids gets incredibly expensive. Paying more for a house makes much more sense. First, you avoid the cost of private school. Plus, the house will probably maintain its value. So the money you put in will come back to you if you sell. Not so with the money you spend on private school tuition. Plus, by living in a more upper class neighborhood you lower your risk of getting victimized by criminals. I wonder if these researchers controlled for crime rates when comparing housing prices.

Parents were not much interested in relatively higher performance of 9th graders versus 4th graders.

Another measure of school quality – how proficiency test pass rates improved between the 4th and 9th grades – didn’t have such a strong impact on house values.

But higher scoring 9th than 4th graders could be a sign that lower IQ people are moving into an area and having more kids.

Home buyers are interested in the bottom line.

The results suggest that, when evaluating school districts, homebuyers are looking at the end result of education – overall test scores – and not a value-added approach that considers how well schools do in improving students, Haurin said.

These findings add another factor to consider in the debate about whether proficiency tests are good for students, schools and communities.

“If parents and residents are paying attention to test outcomes, and not a value-added approach, that means school boards need to pay attention to outcomes too,” Haurin said. “But focusing on test scores may not be the best way to achieve the goal of educating students to the best of their abilities.”

You can understand from this data why the middle and upper classes have so opposed the libertarian dream of school vouchers. What good is the premium the buyers paid for houses in better districts if the rabble can send their kids to schools in good neighborhoods using vouchers? The libertarians are trying to undermine the ability of the middle class to defend their territory. That's just not going to fly politically.

“In Ohio, there are districts with 20 percent pass rates and some with 85 percent pass rates, so based on our findings that would result in about a 23 percent difference in house values solely because of the schools. It is not trivial amount,” he said.

One problem for school districts is that the value-added approach is difficult for researchers to measure, and difficult for the public to understand. Proficiency test scores, however, are readily available and easy to understand, which makes them more influential with the public, Haurin said.

These results suggest that some school boards may have a difficult time convincing residents, as well as potential homebuyers, that they have good schools in their district.

“The disadvantage that some school districts have is that they may be doing very well in terms of adding value to their students’ education, but still may not be among the top scorers in the state. And the reason may not be because of the schools but because their students don’t have the parental advantages that students in other districts have.”

That "parental advantage" translates from dishonest politically correct liberal-speak into English as "genetic advantage".

How much is the Hispanic influx depressing housing prices in some neighborhoods?

Share |      By Randall Parker at 2006 July 13 05:26 PM  Economics Demographic

gcochran said at July 13, 2006 7:01 PM:

School quality has almost nothing to do with the test scores: those home buyers are mistaken if they're buying on that basis.

Half Sigma said at July 13, 2006 7:05 PM:

It's not clear which way the cause and effect works here.

Because people who can afford more expensive houses are richer, and because there is a correlation between wealth and IQ, and because IQ is an hereditary trait, it follows that there will be a positive correlation between the home values in a school district and the test scores of the students in the school district.

I think that the school issue is an important component of why people don't want to live in nieghborhoods where people with lower incomes than themselves live. It's undoubtedly a virtuous circle. Rich people move into the neighorhood, making the schools better, encouraging even more rich people to move in, making the schools even better than before... etc.

Randall Parker said at July 13, 2006 7:43 PM:


Test scores correlate to some degree with IQ. If you have a high school where the average IQ is 85 or 90 then forget about classes in calculus, physics, and other mentally demanding topics. In mixed race schools the classes for mentally gifted minors get shut down because fewer Hispanics and blacks qualify for entrance.

Also, in lower income neighborhoods - especially ones with blacks and Hispanics - you increase the odds your kid will get beat up and have fear in his gut as he rides the school bus to school every morning. The odds of stabbings and shootings in schools go up as scores go down.

My parents debated sending me to a private school after my high school had race riots - really just blacks beating up whites.

ziel said at July 13, 2006 8:03 PM:

I think most parents who seek out school districts with high test scores really have their sights set on a top notch college, and feel that being in a high-performing school district will give their kids a leg up. They convince themselves (no one wants to admit to being that cynical) that their kid is going to get a really good education in the school district (as evidenced by the high scores), but they're really thinking good school district -> good college -> good job. My guess is that a student with mediocre SATs who gets good grades in a 'high-performing' district does indeed fare better in the admissions game than if he got the same grades in a medium-performing district. Now whether or not he will eventually have a more rewarding career as well is a lot more dubious.

gcochran9 said at July 13, 2006 8:17 PM:

Physical safety is worth thinking about: but as for how well you do on the SAT, school attended matters almost not all. For that matter, the college you attend seems to have no effect on your GRE scores and next to no influence on income after graduation.

If you want you kids to do well in school, pick a smart wife. Avoid closed-head injuries. I think that's about it. That could change: I have some ideas for smart pills, and if I ever get anywhere on that, the optimal strategy would also involve being really nice to me. The time to begin that strategy is _now_.

ziel said at July 13, 2006 8:32 PM:

Well I don't think we could be any nicer - do you Randall?

John S Bolton said at July 14, 2006 12:52 AM:

Parents care about the environmental portion of IQ, which is said to be especially subject to influence by peer standards.
They're also concerned with the learning which doesn't add to conceptual ability itself, but which is preparatory to further learning.
There are all the factors, such as drug use, which tend to depress the correlation of IQ and future incomes.
Parents want to avoid populations which have been passaged towards greater propensity to damage others.
When subpopulations get housing and positions by driving better-behaved groups out, this naturally selects for elements which do the most damage.

John S Bolton said at July 14, 2006 1:00 AM:

Family chain immigration is a most efficient means of obtaining such a result.

Rob said at July 14, 2006 5:50 AM:

Dr. Cochran, do you want a guinea pig for smart pills? As I'm sure you've figured out, I need them.

A Berman said at July 14, 2006 6:57 AM:

My wife and I played this game. Our criteria were high quality school, a Jewish community, and of course things like no real crime rate. When I was managing a Chinese immigrant, he and his wife picked our town even though it was way out of their price range. His wife was willing to sacrifice lots in order to ensure that her children went to a high quality school.

Our town also has a large number of Korean immigrants. In the North-East, the Koreans who have 'made it' often move into Jewish communities, again because of the school systems.

Of course we're not interested in "raising the level" betwen 4th and 9th grade. We want the level to be high already at the 4th grade.

The real difference is how much the family cares about school systems. Some ethnic groups don't really care about educating their kids.

Antero Kälvä said at July 14, 2006 11:36 AM:

A few regression coefficients that show the
effect of hispanic, black and asian enrollment
on median home values in California school districts:

Coeff SD
Intercept 250.840 9.9657
hispanic -1.576 0.1948
black -2.695 1.2739
asian 7.142 0.6677

One percentage point of hispanic enrollment decreases
expected median home value in a school district by
1576 dollars.

The median of school district median home values in
California in 2005 was 162285 dollars. The mean was
219118 dollars and the standard deviation was 181378 dollars.

John S Bolton said at July 15, 2006 12:33 AM:

Great work to get numbers like that, AK, but your SD needs correction, a decimal point or something is incorrect.

John S Bolton said at July 15, 2006 12:39 AM:

Home values means single family homes, presumably, yet the latino negative effect occurs in spite of their putting more incomes into such a unit. The negative influence from a Hispanic influx, could be greatly understated, and the price-increasing Asian one, overstated; when there is not adjustment for such increase of population density's effect on land prices.

Antero Kälvä said at July 15, 2006 7:06 AM:

The SD is correct, there are many really expensive districts.
The 90th percentile of median home value is 434201 dollars.

I estimate from California school district data that
mean white household size is less than 2.5,
and mean hispanic household size is greater than 4.

While we're on the subject of schooling, I repeat here some observations on California
school districts I posted in the comments at Sixteen volts:

  • total spending per student has no effect on academic performance
  • instruction spending per student has no effect on academic performance
  • teacher experience has no effect on academic performance
  • the number of students per teacher has a positive effect on white and
    asian students' scores, and is insignificant for hispanic and black students
  • teacher education level (% of teachers with a Master's degree) has
    a significant, but not very large, effect on white and asian students' scores,
    and is insignificant for hispanic and black students
  • income per capita and hispanic, black and asian enrollment account for 80% of
    the variance in the Academic Performance Index between school districts

crush41 said at July 15, 2006 6:40 PM:


What are the p-values on your regression results?

On the national level, total spending per student does have a positive effect on whites (statistically significant) and asians (at around 90% confidence), but not blacks or Hispanics. Hispanics, however, do seem to benefit from lower student-to-teacher ratios. Of course, aggregating all races vanquishes any statistically significant relationship.

Once standard-of-living is controlled for, however, the effect disappears. That's in line with the assumption that school districts with wealthier residents spend more in real terms on education.

Antero Kälvä said at July 16, 2006 2:19 AM:


I used control variables such as:
income per person, median home value, enrollment of economically disabled students,
enrollment of disabled students, enrollment of English language learners,
% of single-parent households with children, population density,
% of adult population with high school diploma,
% of adult population with bachelor's degree,
enrollment of hispanic, black and asian students.

Income/home value variables are significant beyond any doubt in every regression.
Enrollment of hispanic and black students always has a significant effect
on the overall performance of school districts.

The significance of teacher education level depends on what other
variables you include. For example, with just two other variables,
students per teacher and income per person, and white scores as the
dependent variable, it gets a p-value of 0.3 .

I reran some the regressions for white scores and it seems that
teacher education level actually does not have a significant effect.
The p-value of teacher education level for asians is 0.0026,
but the effect is small.

In California, the students-to-teacher ratio has a significant positive correlation
with the scores of all four major racial groups, and it gets small p-values
when predicting whites' and asians' scores even with lots of other variables included.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Web parapundit.com
Go Read More Posts On ParaPundit
Site Traffic Info
The contents of this site are copyright ©