I will protect your pensions. Nothing about your pension is going to change when I am governor. - Chris Christie, "An Open Letter to the Teachers of NJ" October, 2009

Monday, March 19, 2018

The Facts About NJ Charter Schools, Part III: Segregation By English Proficiency

This week, I'm going into detail about a new report on New Jersey's charter schools I wrote with Julia Sass Rubin. In the last post, I showed conclusively that the charters enroll proportionally fewer special education students. In addition, the classified students charters do enroll tend to have less costly learning disabilities. This puts both fiscal and educational pressure on public district schools, which are forced to subsidize charters at the same time they must provide an education to students with special needs.

One of my more tenacious commenters keeps trying to make the case that the reason charters don't enroll as many special needs students is that they declassify special education students at higher rates than public schools. But there is no empirical evidence I am aware of to support this claim. Further, as I've showed before, NJ public district schools spend much more on the support services special education students need than charters. In addition, there are more support staff per pupil in the public schools than in the charters. All the evidence suggests the student populations of charters and public district schools are different.

I don't know why anyone would be surprised by this. The entire theory of charter schools is that they will enroll students who are a good "fit." Why, then, would we be surprised that the charter student populations aren't like the public school populations? Isn't that the entire point?

Keep this in mind as we now look at the differences in English language proficiency between public and charter school students.


Year after year, New Jersey's public district schools enroll many more Limited English Proficient (LEP) students proportionally than the charter schools.

Again, you can try to make the case that this is because the charters remove LEP classification more than public, district schools. But there's no evidence to back up that claim. Further, there is a significant incentive for charters to have students retain their LEP classification, as charter schools receive more funding if they have more LEP students.

The idea that charters are so much better than public district schools at teaching LEP students that they can immediately remove their status, even at a financial disincentive, flies in the face of logic. It's also contradicted by one of the other arguments charter cheerleaders often try to advance: that the difference in LEP classification in some cities is due to the location of the charters.

It is certainly true that the charters often tend to cluster in neighborhoods with smaller Hispanic populations  -- that is likely the explanation for the difference in LEP populations in Newark. But so what? The charters chose to locate in those neighborhoods -- now the district has to pay the costs of educating a concentrated LEP population. Considering that a district like Newark has been underfunded for years while the charters are "held harmless," this remains a serious problem.

Finally, let's consider individual communities, and how their charter sectors differ from public school districts:


As I've noted before, the racial profile of Red Bank Boro -- where the disparity in LEP percentage is the greatest in the state -- is very different than the profile of the area's charter schools:



The idea that the huge disparity in LEP rates between Red Bank Boro and the students attending charters* can be explained by either LEP declassification or location of the school is very hard to defend when it's clear that far more white students proportionally attend the local charter school. The much more plausible explanation is that "choice" has led similar families to "choose" the same schools. This lines up with a growing body of evidence that shows that parents rely on their social networks to make navigate a "choice" system.

All this said, look at some of the districts at the bottom of the table. In North Plainfield and New Brunswick -- communities with large rates of LEP classification -- the charter schools, as a group, actually enroll more LEP students.

As with special education classification rates, the data here show that the charter sector could be enrolling more LEP students. But why doesn't it? If charters are serving more LEP students in New Brunswick and North Plainfield, why aren't they serving at least a similar rate of LEP students in Jersey City or Morris or Passaic or Trenton?

I would suggest the data here shows that it's at least possible that charters could enroll more LEP students. Where then, has the state been during the last decade? Why aren't they demanding better from the entire sector? 

I'll talk about disparities between NJ charters and public district schools in socio-economic status next.


* To be clear: the disparity chart does not only include students who attend the local charter school; it counts all students who reside in the district but attend a charter anywhere in the state. So the "Charter LEP %" figure will not be the same as the local charter school(s) percentage.

3 comments:

JSS said...

The same issue with higher/lower cost special education students applies to English learners as well. Districts/schools with more beginners would require more resources to adequately serve those students than places with a higher proportion of advanced students. It would be great to know if charter schools enroll as many beginners (levels 1-2) as intermediate/advanced ESL students (levels 3-4). This would certainly explain differences in EL performance as well.

Stephen said...

JJ: "One of my more tenacious commenters keeps trying to make the case that the reason charters don't enroll as many special needs students is that they declassify special education students at higher rates than public schools."

I'm not aware of anyone who has asserted that as a full, inevitable, invariable explanation. As you know, a study of Boston schools showed students attending charter schools attaining English proficiency far faster than their counterparts in district schools and, not surprisingly, charter school LEP declassification was much quicker. The degree to which that may also be true in New Jersey seems worth careful examination, but I'm not at all clear how well your state's public data permits you to assess that. Can you determine LEP rates at initial enrollment? The degree to which such rates vary from grade to grade among a particular cohort? Is it possible to compare ELL/FLNE ratios between district and charter schools in NJ? I don't know. But meanwhile, I don't contest the plausibility of additional or alternative contributors, such as NJ charters historically locating more frequently in predominantly low-income, African-American neighborhoods relative to locales with large Hispanic populations.

JJ: "As with special education classification rates, the data here show that the charter sector could be enrolling more LEP students. But why doesn't it?"

Here in Massachusetts, key to increasing ELL enrollment was the education reform effort reflected in the 2010 "Act Relative to the Achievement Gap".
https://malegislature.gov/Laws/SessionLaws/Acts/2010/Chapter12

That required that recruitment and retention plans of charter schools must include "a detailed description of deliberate, specific strategies the charter school shall use to attract, enroll and retain a student population that, when compared to students in similar grades in schools from which the charter school shall enroll students, contains a comparable or greater percentage of special education students or students who are limited English-proficient of similar language proficiency as measured by the Massachusetts English Proficiency Assessment examination [...] If a school is or shall be located in a district with 10 per cent or more of limited English-proficient students, the recruitment strategies shall include a variety of outreach efforts in the most prevalent languages of the district. The recruitment and retention plan shall be updated each year to account for changes in both district and charter school enrollment."
[...]
"(g) To ensure that a commonwealth charter school shall fulfill its obligations under its recruitment and retention plan, the school district or districts from which the commonwealth charter school is expected to enroll students shall annually provide, at the request of a commonwealth charter school, to a third party mail house authorized by the department, the addresses for all students in the district eligible to enroll in the school, unless a student’s parent or guardian requests that the district withhold that student’s information; provided, however, that the department may require the charter school to send the mailing in the most prevalent languages of the district or districts that the charter school is authorized to serve."

laMissy said...

Here in Massachusetts, key to increasing ELL enrollment was the education reform effort reflected in the 2010 "Act Relative to the Achievement Gap".
https://malegislature.gov/Laws/SessionLaws/Acts/2010/Chapter12

This legislation was written by Marty Walz, then a state legislator, now working for the charter school industry.

She claimed that in order to increase enrollement of ELL and SPED kids, the charters needed more effective recruitment tools. The change in the law allowed charters to harvest student information from the public schools. Parents must opt OUT of the collection of their child's information to the charters, and the public schools must use their resources to submit the data.

Nonetheless, our charters still way underserve these populations, eight years later.