What does affect education outcomes?

by Michael Jacobs, UNC professor of finance, published in News and Observer, November 10, 2013.

North Carolina’s teachers are outraged over how public education fared in this year’s state budget. While it would appear to be common sense that putting more money into schools would improve our children’s educational outcomes, my anecdotal experience and the data indicate it is not so simple.

As a half-time professor at UNC, I am a public school teacher, and the education budget affected me this year. A new course I conceived in sustainable finance was not funded for budgetary reason. Moreover, my wife and I were both educated in public schools and strongly desired for our daughter to have a similar life experience. But after three years sending her to the top-rated public elementary school in one of North Carolina’s best-funded school districts, we were so disappointed in the quality of instruction that we moved her to a private school. This experience inspired me to more closely examine the assumption that dollars spent on education are strongly correlated to educational outcomes.

I analyzed two statistics people frequently point to when arguing that certain schools are superior: test results and dollars spent per pupil.

Simply comparing test results across school districts is a flawed approach to measuring educational effectiveness. If all the schools in the state were the exact same quality, the district that has the inherently brightest children would have the highest passing rates on end-of-grade exams.

Since we don’t have data on the IQs of schoolage children, we need a reasonable proxy. A useful statistic that is publicly available is the parents’ education. The percentage of adults with a college degree or higher is far greater in Chapel Hill than in any other city in the state. So it would be logical to assume either Chapel Hill is populated with brighter kids or parents who care more about education or both. In either case, this would inject a significant bias when comparing test results across school systems – which would make schools in Chapel Hill and other communities with highly educated parents look better than they are on a relative basis.

The second demographic bias is that in every school district in North Carolina, white and Asian students perform significantly better on EOG tests than do non-Asian minorities. Surprisingly, the racial performance gap in Chapel Hill, which spends the most local dollars on public education, is the second highest in the state, which we would not expect to see if it had superior schools.

Curiously, the only city with a worse performance gap in test scores of blacks and whites is Asheville, which is the only city that spends more per pupil than Chapel Hill (Chapel Hill spends the most from local property taxes; Asheville, the most overall including federal and state dollars.)

Chapel Hill schools have the second-lowest percentage of non-Asian minorities, which has a material positive bias on EOG passing percentages. Only one other major county besides Orange (home to Chapel Hill) saw a decline in its black population in the past decade. It was Buncombe, home to Asheville. These are the only two cities with a special school tax.

For the school districts of the state’s 24 largest cities, I examined the correlation between the percentage of students passing 2012 math and reading EOG proficiency tests and parents’ education, percentage of white/Asian students and spending per pupil. Not surprisingly, math proficiency is most strongly correlated to parents’ education. Reading proficiency is actually slightly more correlated to ethnicity.

Amazingly, there is no statistically significant correlation between the amount of money spent per pupil on public education and academic proficiency throughout North Carolina.

The “progressive” communities that are the most vocal critics of our elected officials clearly spend more on education, but their outcomes appear to be more a function of their demographics. While spending more may not improve our children’s education, higher taxes do improve educational performance metrics. This is because high local taxes push minorities and less-educated parents out of the school district, which drives up the passing rates.

There is a new normal in state and local government finance, though some communities may not realize it. Elected officials are having to learn to do more with less. I am not recommending further education spending cuts by the state. And I do believe the teachers have some legitimate gripes. But rather than demonizing state government leaders who are struggling with economic realities, perhaps we should channel our energy into figuring out how to do more with less in education, as virtually every business and nonprofit in America has. The answer to every multiple-choice problem is not A. ) more money.