Private School is Stupid

You hear the public schools are in shambles so you do what any reasonable parent would do and start shopping for private schools right? You would do anything to help you kid get ahead and are willing to sacrifice in order to make it happen. How amazing!! Maybe it would be amazing, if that’s actually what you were doing. But you’re not. You’re just throwing money down the toilet. Don’t get me wrong, there may be some legitimate reasons to send your child to a private school, if all boys or all girls, or religious education are worth millions of dollars to you. But if you think that your children are receiving better instruction at a private school you are probably wrong. And if you think that the instruction that they receive is worth the additional $34,000 per year that you spend on tuition then you are almost certainly wrong. Let’s talk about why private school is stupid:

(Note: This article has been reposted from my personal finance blog: https://richdadrichdad.com/private-school-is-stupid/)

Value Added vs. Biased Samples

When comparing the quality of instruction/support at two schools it is important to judge them on the value they add, and not differences in their student bodies. Consider the following example:

Which is the better school? If you answered School A, then your parents probably spent too much money on private school. The correct answer is “I don’t have enough information”. You do not know if the higher test scores coming out of School A were the result of superior instruction or the result of inherent differences between Student Body A and Student Body B.

Now lets consider another situation:

In this example we actually do have enough information to draw a conclusion. Both Student Body A and B perform better at School B than in School A so you can determine that School B does in fact add more value that School A. Unfortunately, the real world does not lend itself to this type of experimentation. However, we do have a great little study from the Fordham Instutute that attempts to do the next best thing. The Fordham Institute is a huge proponent and architect of school choice programs across the country aimed at removing children from “failing public schools” and putting them into schools with “higher quality instruction”. They looked at samples of children of similar communities, some who stayed in public and others that took vouchers to go to public schools and guess what happened? The students that took vouchers to go to private schools actually did WORSE than the kids that stayed, and the difference was especially pronounced in math. All those tuition dollars for worse performance.

Private school is stupid.

NOTE: So why does the school choice system still exist if it actually reduces the chances of student success? Because it is not actually about improving student education it is actually about subsidizing failing PRIVATE schools. But that’s a story for another day.

SAT Scores

Let’s say that you live in Northeast Ohio and are considering a choice between my alma mater Cleveland Heights High School and Gilmour Academy ($20k-$34 per year). The average SAT score at Cleveland Heights was 949 while the average SAT score at Gilmour was 1183.

“A 234 point difference in ACT scores ?!?!?!?! Clearly anyone trying to give their child the best chance to get into college, succeed in college, and thrive afterward would bite the bullet, pay the money and send your kid to Gilmour… right?“

If that’s how you think please re-read the section above over and over again until you realize that the difference in SAT scores IS NOT and indication of a better school, at least not until we have more information.

Income Effects

We may not have the ability to take the Gilmour student body and force them to attend Cleveland Heights High School to see what their SAT scores would be. But we do have some demographic information that can help. For example, we know based on numerous studies that every $20k of additional f household income results in about 12 additional points on your SAT score (Link). According to the ODE website the median income of students attending Cleveland Hts High School was $37,752. While Gilmour does not readily provide the median income of its student body, the median income of Gates Mills, where it is located is $123k per year. It is highly likely that the median income at Gilmour is substantially higher than the median income of Gates Mills, but we will use that number to stay conservative. Therefore, based on income alone you would expect at least a 51 point difference in scores between the two schools even if the quality of instruction was identical.

“See that leaves 183 points of SAT score unexplained!!! Private school does add value!!!!“

Slow down, there’s more:

Racial Effects

17% of Gilmour’s population is non-white, compared to compared to 83% at Cleveland Heights. Although it upsets me greatly that the achievement gap persists even after income is accounted for, it is a fact of life in America and one that has to be considered when comparing the effectiveness of schools. As it turns out, white students at Cleveland Heights average 1152 on their SAT’s. Gilmour does not have an equivalent breakdown of SAT performance by race which makes direct comparisons difficult, but if we assume that the 17% of the non-white population at Gilmour scores somewhere between the Gilmour average SAT score and the average SAT score for non-white students nationwide, then we can estimate that the white population tests somewhere between 1180 and 1230. Having a whiter student body than public schools biases your test scores up regardless of the quality of instruction.

“See that still leaves 30-80 points unaccounted for!!! Private Schools Are Better!!!”

Hold on. Still not done.

Selection Bias

There are also at least two examples of sample bias that have not yet been accounted for. First of all, Cleveland Heights accepts all students regardless aptitude at the time of admission. Gilmour has entrance exams that are used to filter out kids that are likely to score low on the actual SAT’s later in life. This again biases their SAT scores up and has NOTHING to do with the quality of instruction or support at the school.

Second, at Cleveland Heights, the entire student body takes the SAT, including students that have no intention of attending college. At Gilmour only 62% of the students take the SAT. When given a choice, who are the students most likely to take the SAT? Kids that plan on going to college, and are therefore more likely to score high than the general school population. This again biases the scores higher regardless of the quality of instruction at the school. What would Gilmour’s SAT scores look like if 100% of students had to take the test like they do at Heights? Well… They probably wouldn’t be higher.

It is difficult to quantify just how much this shifts the scores, but it is probably a large effect that at least makes up for the score discrepancies between the white students. (Note, since the minority population is so low at Gilmour and they do not have a racial performance breakdown, the uncertainty in minority SAT scores is too large to draw conclusions in that group.)

The income discrepancies, racial discrepancies, and selection biases of the two schools are such that even if the quality of instruction in the schools was identical, the expected performance of the populations would be dramatically different. STOP ATTRIBUTING DIFFERENCES IN TEST SCORES TO THE QUALITY OF THE SCHOOLS!!! Life is just not that simple.

What Does This Have to Do With Personal Finance

So far this article may seem to be more about public education or statistics than personal finance, which is the focus of my blog, so I want to get back to dollars and cents. The cost of sending a child to Gilmour from Kindergarten through high school ranges from $20k per year to $34k per year. A family with three kids that elects to send their children to Gilmour from kindergarten to college will spend $1.1 Million doing so over the course of about 17 years. Had you instead invested the tuition and earned a 7% ROR, then at the end of that 17 years you would have $2M in the bank. If you continued to invest that money until retirement age you would have nearly $6M. If you continue to invest it until you die at age 80 it will have turned into $31 Million. Let’s say, even in the best case scenario, that you find a school that truly does offer substantial value above your current public school option. How much is it worth to you? Go up to your kids right now and ask them if they’d rather go to Gilmour or inherit $10M a piece. Even if your entire family earns half-off scholarships, ask the same question to your kids with a $5M inheritance.

Private School is Stupid.

Summary

People tend to assume that private schools are better than public schools because private schools tend to have better test scores. This is a basic statistical fallacy that fails to take into account differences in the incoming student bodies. Most private schools service a wealthier and whiter population than public schools. Private schools also have entrance exams and do not require the entire student body to take college entrance exams, which results in selection bias. All of these factors improve their standardized test scores, but NONE OF THESE FACTORS HAVE ANYTHING TO DO WITH QUALITY OF INSTRUCTION. If your local high school was allowed to cherry pick test takers the way private schools do, their aggregate scores would be just as high as private schools. At the individual level most of the kids scoring well at private schools would have scored just as well had they gone to public school. Their parents just have to tell themselves something to help them sleep at night while they are ruining their financial futures.

Even if there is some benefit to going to a top notch private school it must be weighed against the cost, which in many cases is astronomical. Even if your child received a 50% scholarship to attend Gilmour and you were convinced that it would increase his SAT score by 50 points, you would still be sacrificing millions of dollars in opportunity cost (quite literally) to make that happen.

Why are people willing to pay so much money to send their children to private schools despite little to no evidence of higher quality? Perhaps quality of instruction is not what is being purchased. Perhaps making your child part of a hand-picked, wealthier, and whiter student body is what is being purchased, even if that means your child receives inferior instruction and ruing your financial future. How much are you willing to pay to escape the demographics of your community?

Private school is stupid.