Regarding the discrepancy between the census and CDE estimates, I see the ACS question includes "home school" as part of the private school enrollment answer:
That might explain at least part of the discrepancy if tens of thousands of students are homeschooled in CA.
The density variable is interesting as perhaps a "supply-side" explanation, if it's seen as easier to find an audience for a new private school in a large city, and thus the larger and more varied supply of private schooling options creates more interest in the population for private schooling.
That said, that SF by itself drives most of that relationship is reason for skepticism. In particular, the fact that SF is right on the regression line is not that reassuring: if you have an observation that is an outlier both in the predictive variable (density) and the response (private school %), then that "high-leverage" observation is going to pull the regression line strongly towards itself.
I typically think of three categories of schools -- public, private, and parochial. There are a lot of neighborhood parochial (mostly Catholic) schools where the majority of families are not poor, but are making a financial sacrifice to pay for their child's education. In my experience, most families are choosing a Catholic education based on faith and tradition rather than as a rejection of public education.
Parochial schools are usually significantly less expensive and have fewer bells and whistles than a private independent school.
This dynamic may be changing as the population of San Francisco becomes less Catholic overall. I'm wondering if you can see a shift in your data over the decades between low-cost parochial schools to high-cost independent schools?
It is important to distinguish between K-8 and High Schools. Enrollment in Catholic K-8 schools is down around 50% since the turn of the century. Enrollment in Catholic high schools is up.
Yeah and that tracks with your money theory. Working and middle-class families who traditionally enrolled in K-8 Catholic schools have been priced out of the city, and families that remain can afford higher-end high schools. Lower price high schools have either closed (Mercy), merged (Riodan), or changed their mission (Immaculate Conception Academy)
I know about Mercy and Riordan. Tell me more about ICA's changed mission. I'm not really familiar with it despite driving past it multiple times a week.
Immaculate Conception converted from a regular Catholic girls high school to a Cristo Rey school more than 10 years go (can't remember the exact year.) https://www.cristoreynetwork.org/about
Cristo Rey network schools combine full-time school with 1-day a week work experience in local businesses. Businesses sponsor the mission of the school by offering work experience to students that offsets their tuition.
Mercy closed. The site is the new home for CAIS. Riordan used to be an all boys school. When Mercy closed, Riordan went co-ed and now has more students than Mercy and Riordan together had before.
K-8 schools also have fewer discipline problems and usually smaller attendance boundaries (high schools often have large enough attendance bounties to include poor neighborhoods). It's common for a city to have some decent K-8 public schools but few good public high schools.
I've done this analysis for San Francisco but not for the rest of the state. Private school enrollment figures for 2024-25 should be published over the Summer so I'll write it up early in the new school year and will try to include this data.
Just got my Florida school vouchers approved yesterday.
Since 2020 there is also people who want a different kind of schooling for their kids. We were in a blue ribbon rich suburban public school and I was still dissatisfied with it.
“In the absence of evidence that private schools in the Central Valley are systematically failing to report their enrollments to CDE, I’m going to trust the CDE’s numbers.”
I don’t have an opinion on the thesis, but keep in mind that CDE’s oversight of private schools is extremely light and light-handed. You fill out an online “affidavit” once a year that honestly could be a SurveyMonkey form. It requires some extremely basic information like school name, custodian of records, contact information, and some very light demographic info like current enrollment, student age range, and how many seniors graduated high school the past year.
On top of that, enforcement is nearly nonexistent. No one goes around forcing private schools to file. If you have filed, then you get an email reminder to file annually. Follow-up if you don’t is next to nonexistent. Finally, I can’t remember exact details off the top of my head - it was a policy quietly executed by CDE and never really documented - but CDE doesn’t issue you a registration number until you report more than (5? 7?) students. So I wonder if their numbers include school populations without registration numbers.
Also, there is zero obligation to update your affidavit for accuracy. If I had 500 students the week before filing, and 350 left the day before filing, the correct affidavit answer is 150 not 500.
I once did a legal summary of applicable CA private K-12 law for my company’s exec team. I didn’t fill a whole page.
That's an interesting one. As a society, we want the high school graduation rate to increase, even if those students have no intention of going to college, because high school graduates earn much more than those who drop out.
It's okay to be a high school graduate but not meet the SBAC requirements. Proficiency on the SBAC is meant to indicate preparedness for post-secondary education. It was never intended to be a high school graduation requirement.
At the same time, even if they're not reaching SBAC standards, we want to know that HS graduates genuinely earned the grades they need for graduation and are not just being given "C"s to get them through the system. More generally, we'd like evidence that they are actually learning. I don't know if that evidence exists. Ideally, we'd be able to compare how students did on the 11th grade SBAC with how they did on the 8th grade SBAC. If they've improved by more than the state average, I'd celebrate that as a success even if they are still scoring below standard.
Unfortunately a student can fulfill all their A-G requirements with a grade of D and still graduate. Also, some SFUSD schools use equity grading. The lowest grade a student can earn is a 50. This happens at Presidio Middle school. Why anyone with means would send their child to SFUSD is beyond me.
I appreciate the nuanced response but I disagree with what I *think* is your overall conclusion.
I think you're saying that:
- high school graduation is a useful measure of future earnings, AND
- that relationship is due to things people learn in high school.
I'm not sure that's true! An alternative hypothesis is:
- anyone who shows up consistently will graduate high school, irrespective of learning or achievement, and
- people who don't show up consistently in high school have some reason(s) for that, and those same reasons affect their future earnings.
It may be true that Math SBAC is designed to measure academic readiness, but is the content knowledge expected for 11th graders beyond that which most people will need in life? If so, it's probably *also* true for all the other subjects covered in high school.
So I find it hard to understand someone holding these two positions simultaneously:
A) high school is a useful for people who do not plan to go to college, and
B) 11th grade math SBAC does not matter for people who do not plan to go to college.
Do you believe both A and B, or am I reading too much into what you wrote?
Interesting as always. Curious how much of the observed result is explained by income level alone. Though I certainly get why density matters - people will only drive so far for a school, especially if there is another one nearby.
Thanks. Another way to couch this is, "Richer people tend to live in cities, and are more likely to spend on private schooling for their kids." It's not objectionable and doesn't tell you much about local public schools - it's just the way things are.
That said, it would be interesting to look at why SF has the lowest percentage of school age kids of any US city. Has that changed over time? That seems a much bigger driver of SFUSD enrollment than the "going private" thing.
The biggest factor is probably that San Francisco is more densely populated than other cities. Women are having fewer kids and having them later in life everywhere. I know that the average mother's age in SF is particularly high now but I haven't checked if it has always been higher than other cities. It has also always been the case that families leave the city after their children are born (the kindergarten class size is much lower than births 5 years earlier).
Hi Paul,
Regarding the discrepancy between the census and CDE estimates, I see the ACS question includes "home school" as part of the private school enrollment answer:
https://www2.census.gov/programs-surveys/acs/methodology/questionnaires/2022/quest22GQ.pdf
That might explain at least part of the discrepancy if tens of thousands of students are homeschooled in CA.
The density variable is interesting as perhaps a "supply-side" explanation, if it's seen as easier to find an audience for a new private school in a large city, and thus the larger and more varied supply of private schooling options creates more interest in the population for private schooling.
That said, that SF by itself drives most of that relationship is reason for skepticism. In particular, the fact that SF is right on the regression line is not that reassuring: if you have an observation that is an outlier both in the predictive variable (density) and the response (private school %), then that "high-leverage" observation is going to pull the regression line strongly towards itself.
Interesting topic!
I typically think of three categories of schools -- public, private, and parochial. There are a lot of neighborhood parochial (mostly Catholic) schools where the majority of families are not poor, but are making a financial sacrifice to pay for their child's education. In my experience, most families are choosing a Catholic education based on faith and tradition rather than as a rejection of public education.
Parochial schools are usually significantly less expensive and have fewer bells and whistles than a private independent school.
This dynamic may be changing as the population of San Francisco becomes less Catholic overall. I'm wondering if you can see a shift in your data over the decades between low-cost parochial schools to high-cost independent schools?
It is important to distinguish between K-8 and High Schools. Enrollment in Catholic K-8 schools is down around 50% since the turn of the century. Enrollment in Catholic high schools is up.
Yeah and that tracks with your money theory. Working and middle-class families who traditionally enrolled in K-8 Catholic schools have been priced out of the city, and families that remain can afford higher-end high schools. Lower price high schools have either closed (Mercy), merged (Riodan), or changed their mission (Immaculate Conception Academy)
I know about Mercy and Riordan. Tell me more about ICA's changed mission. I'm not really familiar with it despite driving past it multiple times a week.
Immaculate Conception converted from a regular Catholic girls high school to a Cristo Rey school more than 10 years go (can't remember the exact year.) https://www.cristoreynetwork.org/about
Cristo Rey network schools combine full-time school with 1-day a week work experience in local businesses. Businesses sponsor the mission of the school by offering work experience to students that offsets their tuition.
What happened to mercy and riordan?
Mercy closed. The site is the new home for CAIS. Riordan used to be an all boys school. When Mercy closed, Riordan went co-ed and now has more students than Mercy and Riordan together had before.
Cheers.
K-8 schools also have fewer discipline problems and usually smaller attendance boundaries (high schools often have large enough attendance bounties to include poor neighborhoods). It's common for a city to have some decent K-8 public schools but few good public high schools.
How has the number of private schools changed over time, specifically in San Francisco, but also across the state?
I've done this analysis for San Francisco but not for the rest of the state. Private school enrollment figures for 2024-25 should be published over the Summer so I'll write it up early in the new school year and will try to include this data.
Just got my Florida school vouchers approved yesterday.
Since 2020 there is also people who want a different kind of schooling for their kids. We were in a blue ribbon rich suburban public school and I was still dissatisfied with it.
“In the absence of evidence that private schools in the Central Valley are systematically failing to report their enrollments to CDE, I’m going to trust the CDE’s numbers.”
I don’t have an opinion on the thesis, but keep in mind that CDE’s oversight of private schools is extremely light and light-handed. You fill out an online “affidavit” once a year that honestly could be a SurveyMonkey form. It requires some extremely basic information like school name, custodian of records, contact information, and some very light demographic info like current enrollment, student age range, and how many seniors graduated high school the past year.
https://www.cde.ca.gov/sp/ps/psainstructions.asp
On top of that, enforcement is nearly nonexistent. No one goes around forcing private schools to file. If you have filed, then you get an email reminder to file annually. Follow-up if you don’t is next to nonexistent. Finally, I can’t remember exact details off the top of my head - it was a policy quietly executed by CDE and never really documented - but CDE doesn’t issue you a registration number until you report more than (5? 7?) students. So I wonder if their numbers include school populations without registration numbers.
Also, there is zero obligation to update your affidavit for accuracy. If I had 500 students the week before filing, and 350 left the day before filing, the correct affidavit answer is 150 not 500.
I once did a legal summary of applicable CA private K-12 law for my company’s exec team. I didn’t fill a whole page.
"If you come across some education-related issue that might be addressable with data, please let me know. Fresh ideas are always welcome."
I wonder whether you have any interest in comparing high school graduation rate and student achievement?
I looked at one data point: SFUSD's high school graduating class of 2024.
83% graduated even though, just a year before, only 45% met state standards in math (in the grade 11 SBAC).
I wonder whether this issue has changed over time and/or varies across districts.
That's an interesting one. As a society, we want the high school graduation rate to increase, even if those students have no intention of going to college, because high school graduates earn much more than those who drop out.
It's okay to be a high school graduate but not meet the SBAC requirements. Proficiency on the SBAC is meant to indicate preparedness for post-secondary education. It was never intended to be a high school graduation requirement.
At the same time, even if they're not reaching SBAC standards, we want to know that HS graduates genuinely earned the grades they need for graduation and are not just being given "C"s to get them through the system. More generally, we'd like evidence that they are actually learning. I don't know if that evidence exists. Ideally, we'd be able to compare how students did on the 11th grade SBAC with how they did on the 8th grade SBAC. If they've improved by more than the state average, I'd celebrate that as a success even if they are still scoring below standard.
Unfortunately a student can fulfill all their A-G requirements with a grade of D and still graduate. Also, some SFUSD schools use equity grading. The lowest grade a student can earn is a 50. This happens at Presidio Middle school. Why anyone with means would send their child to SFUSD is beyond me.
https://www.sfusd.edu/school/independence-high-school/learning/graduation-requirements
I appreciate the nuanced response but I disagree with what I *think* is your overall conclusion.
I think you're saying that:
- high school graduation is a useful measure of future earnings, AND
- that relationship is due to things people learn in high school.
I'm not sure that's true! An alternative hypothesis is:
- anyone who shows up consistently will graduate high school, irrespective of learning or achievement, and
- people who don't show up consistently in high school have some reason(s) for that, and those same reasons affect their future earnings.
It may be true that Math SBAC is designed to measure academic readiness, but is the content knowledge expected for 11th graders beyond that which most people will need in life? If so, it's probably *also* true for all the other subjects covered in high school.
So I find it hard to understand someone holding these two positions simultaneously:
A) high school is a useful for people who do not plan to go to college, and
B) 11th grade math SBAC does not matter for people who do not plan to go to college.
Do you believe both A and B, or am I reading too much into what you wrote?
Interesting as always. Curious how much of the observed result is explained by income level alone. Though I certainly get why density matters - people will only drive so far for a school, especially if there is another one nearby.
Income by itself has an R-squared of 0.71. Density by itself has an R-squared of 0.55
Thanks. Another way to couch this is, "Richer people tend to live in cities, and are more likely to spend on private schooling for their kids." It's not objectionable and doesn't tell you much about local public schools - it's just the way things are.
That said, it would be interesting to look at why SF has the lowest percentage of school age kids of any US city. Has that changed over time? That seems a much bigger driver of SFUSD enrollment than the "going private" thing.
The biggest factor is probably that San Francisco is more densely populated than other cities. Women are having fewer kids and having them later in life everywhere. I know that the average mother's age in SF is particularly high now but I haven't checked if it has always been higher than other cities. It has also always been the case that families leave the city after their children are born (the kindergarten class size is much lower than births 5 years earlier).