More thoughts on Charter Punditry & Declarations of Certainty

I’m a little late in pouncing on this one. JerseyJazzMan beat me to the punch with some relevant points.  A short while back, the Wall Street Journal posted an op-ed by Deborah Kenny, CEO of New York based charter chain Harlem Village Academies. Kenny’s op-ed purported to explain why charter schools are successful.  Of course, we could spend all day on that contention alone, since it is relatively well understood that charter results have been mixed at best. Indeed, I have explained in my published work and in blog posts that the track record for certain charter chains and in certain settings seems stronger than in others.

Here is how Deborah Kenny explained why charters succeed (implicitly where traditional public schools do not):

Critics claim that charter schools are successful only because they cherry-pick students, because they have smaller class sizes, or because motivated parents apply for charter lotteries and non-motivated parents do not. And even if charters are successful, they argue, there is no way to scale that success to reform a large district.

None of that is true. Charters succeed because of their two defining characteristics—accountability and freedom. In exchange for being held accountable for student achievement results, charter schools are generally free from bureaucratic and union rules that prevent principals from hiring, firing or evaluating their own teams.

As is par for the course of late in such arguments, Kenny’s chartery punditry is completely void of any data or contextual information that might provide insights as to why, or even whether charter schools “succeed.” Yet, while bafflingly void of substantiation, Kenny’s punditry is disturbingly decisive & hyper-confident.

It is yet another case of declaring to know absolutely what we absolutely don’t know!

For the moment, let’s accept Kenny’s proposition that at least in New York City, many charter schools affiliated with high profile management organizations have posted solid test scores (not entirely the case… but let’s accept that proposition…).

So then, let’s compare New York City charter schools from these CMO chains to traditional public schools in the city on a handful key parameters – a) how much they spend and b) which kids they serve – each relative to the schools which they supposedly far outshine.  These are things that actually matter. Now… if they do spend the same as NYC traditional public schools and serve similar student populations, we might be able to make the case that their “success” is a function of something different that they are doing with the same dollar – more bang for the buck. A relevant question… but a hard one to distill. But, if they serve very different student populations, then it’s even harder to distill what the heck is really going on.[1]

Further, if they are outspending NYC public schools that do serve similar populations, their access to resources may be what allows them to do different stuff… which may then explain their supposed “success.”  It would certainly be hard to make the above claims without looking at any of this, wouldn’t it?

So, here’s the stat sheet:

For each of these comparisons I have used a three year panel of data on NYC Charters schools and all NYC traditional public schools, from 2008 to 2010. To compare spending, I have used the estimates generated in our recent report on charter school spending:

  • Baker, B.D., Libby, K., & Wiley, K. (2012). Spending by the Major Charter Management Organizations: Comparing charter school and local public district financial resources in New York, Ohio, and Texas. Boulder, CO: National Education Policy Center. Retrieved [date] from

Further discussion of the spending comparisons for NYC can be found here:

In short, each of these charter chains spends more per pupil than NYC public schools that serve similar student populations. Some, like KIPP and UnCommon schools spend a lot more!

Further, when compared against same grade level schools citywide, each of these charter chains serves fewer children with disabilities (and I lack data on the type of disabilities, which may also matter).

Finally, when compared against same grade level schools in the same zip code, each of these charter chains serves far fewer low income children and FAR fewer children with limited English language proficiency.

These substantive differences in resources and student populations make it difficult if not impossible to assert that these charter school chains operating in New York City have somehow identified a magic formula for success that is neither resource dependent nor dependent on serving very different student populations than city district schools.

There is certainly no basis whatsoever for asserting that accountability and freedom – specifically freedom from bureaucratic and union rules – are necessarily the determinants of charter success. In fact, these broad principles apply similarly to all independent charters, but while some are good, others suck – and many are allowed to persistently suck despite supposed heightened accountability. Indeed, the upper half is better than average! And the lower half… is not!

It’s hard to suggest that either of these factors – accountability or freedom – are the determinants of charter success when success varies so widely across charters. What does tend to vary across charters is a) access to philanthropic resources and b) student populations served. AND… it may also be the case that some charters have adopted unique strategies…… some of which may actually come with additional costs!

There may be some cool stuff going on in some of these schools, just as there may be some cool stuff going on in NYC district schools.  It may well be that freedom from bureaucratic rules permits schools to do cool stuff.  It would certainly seem advantageous in the context of New York State moving forward to be able to skip out on complying with new, ill-conceived teacher evaluation legislation.

We need to figure out what works and for whom, whether those ideas come from traditional public schools, charter schools or private schools.

We need to figure out the costs of doing these things. Ken Libby, Kathryn Wiley and I discuss these issues in our recent policy brief (read it! It’s not some anti-charter propaganda. It’s an actual study of spending data… with detailed documentation & extensive lit review).

Unfortunately, the tendency among charter “defenders” is to simply deny, deny, deny… ignore costs (make bizarre, unfounded excuses, present half-assed, back of the napkin estimates, or sidestep them)… ignore substantive contextual issues, etc., etc., etc. (certainly, the tendency among the attackers is to declare all charter operators/supporters to be union-busting privatizing profiteers – also an unhelpful characterization for a diverse array of institutions).

It’s time to start digging deeper into what makes schools tick and for whom and how to provide the mix of schooling that best serves the largest share of children.

[1] As I explained in a recent post, even in a lottery study – of students lotteried in/lotteried out – those lotteried out likely attend schools with substantively different classroom peers than those lotteried in, and it remains difficult if not impossible to distill school/teacher effect from peer effect since both operate at the classroom level.






    1. Stuyvesant HS is NOT a “magnet school”. By federal regulation, magnet schools are prohibited from “screening” applicants except insofar as the magnet school has an already existing “zone” or catchment area. In that event, magnet schools are allowed to grant preferences to those within the zone.

      Stuyvesant is among approximately eight “specialized schools” where admissions are exclusively determined by the outcomes on a standardized admissions test. Thus, there is no “subjective” factor in determining who gets in. the same is true for new admissions into Hunter High School (which is run by CUNY)

      The only two outright “elite” schools subject to openly subjective admissions criteria are LaGuardia (an audition school) and Townsend Harris – although admittedly, Bard and Beacon also can be charged with that stigma. Ironically, LaGuardia, Bard and Beacon are all MUCH more diverse than the “specialized” schools.

      That DOES leave the ongoing scandal of the District Two “screened” High Schools which are egregiously segregated. The worst example being the HUMONGOUS differences in ethnic composition of two “high performing” schools located less than a half mile apart in a common commercial area – viz., Baruch and Manhattan Village Academy. THAT is an undeniable disgrace which no-one seems at all interested in.

      As to the policy questions raised by the existence of “specialized” or “screened” schools, I’ll leave that to another day. As to magnet schools, which are deliberately designed to REDUCE “minority group isolation”, their positive impact is far less in doubt. Using precise language matters – especially in a blog of this high intellectual calibre.

      1. You may be adding more to the delineation of “magnet” schools, “specialized” schools and “elite” schools than is generally acknowledged, but I certainly appreciate the necessity for establishing precision in this delineation. It is a much needed clear, delineation/typology (and one that has perhaps a few more dimensions).

        Magnet programs in cities like Hartford and Kansas City have been established to promote diversity by recruiting suburban students, but have also done so by providing elite academic programs and specialized programs (arts). Other “magnet” programs in places like Montgomery County, MD are focused on within-county, cross-neighborhood diversity goals, providing themed curriculum, but using lottery-based assignment systems. “Magnet” schools in Hartford in particular, both because they offer more rigorous academic programs and because they do successfully recruit in some suburban kids, have populations that are hugely different from Hartford city schools. See: In short, yes… the hartford magnets do succeed in integrating high ability Hartford city kids who happen to be minority (but less poor than their minority peers in the district) with other less poor, white suburban kids… but they also have the effect of segregating kids within the city.

        It would indeed be a fun exercise to establish a complete, precise classification system regarding public school selective/limited admissions systems and explore the relationship between those systems and segregation. It seems to me that some of the “fuzzier” selection processes might be subject to greater segregative effect – even though that should not be the case (one would expect race/poverty disparities in test-based admissions). It seems that this is what you are implying above? Screened schools that just-so-happen to end up really segregated?

        There are certainly numerous mechanisms in play in the New York City system. In my time teaching middle school at an elite private school in NYC, we had many parents who seemed to be banking on spending the tuition on elite private education up front to give their kids an edge on gaining access to Stuyvesant or Bronx Science – incl. nailing the test. I wrote many recommendations each year. Clearly, that pattern of sorting doesn’t help diversity in the highly selective schools.

        Fuzzy screening processes are more interesting to explore – with clearly greater potential for political manipulation/corruption. Certainly, that type of gaming (which I have seen/noticed elsewhere) is offensive and worth exploring.

  1. How about comments on the ongoing scandal of quasi-private public schools that are allowed to screen out students via admissions test and auditions? You call them “magnets in New York,and the NY Times points out one of the best known has had cheating scandals for years:
    Does this bother you, or are you ok with quasi private “magnets”?
    Joe Nathan, parent of 3 children who attended St. Paul, Minnesota public schools open to all, k-12.

    1. I’ve commented in previous posts that much of the student segregation that occurs in larger urban districts is a function of districts’ own practices of providing magnet schools etc. That providing the best system for the largest share of kids is indeed an issue of determining the correct role for magnet schools, charter schools and “other” schools.

      I think that it’s equally or even more problematic to overdo magnet schooling – leading to extreme segregation. That said, pundits don’t typically pretend (for the most part… ) that magnets are creating the outcomes they do with “the same students.” Further, we also often acknowledge that magnets – because they serve the elite students and provide elite programs – often represent a resource drain on districts – an inequitable allocation of resources to those already most advantaged. We acknowledge that their students are different and for that reason their results are not necessarily scalable to “all kids” (kids unlike those in the magnets). My problem with charter rhetoric is that we rarely make this same acknowledgement. We instead accept – unfounded – that the kids are the same… because the entry mechanism is randomized… even though the sign-up for the randomization is non-random, as is the attrition…. and certainly the resulting populations in many cases are very different. Any reasonable conversation about scalability must recognize this.

      As for the public/private question… that’s a somewhat different question and one that I addressed here. In most cases, Magnet schools are still governed/operated by districts. They are limited public access (admissions tests, etc.), publicly operated/managed schools (see: That means that employee and student rights issues are substantively different than if these were privately governed/managed institutions. That’s a totally separate issue.

    2. Why don’t you write something Mr. Nathen? Or are you just trying to excuse the lucrative charter industry by pointing out highly specialized public schools called Magnets?

      On balance, while Magnet schools are more like charters in terms of being able to exclude students, they are unlike privately managed charters (or quasi-public if you like) in that they are district schools and typically have publicly elected school boards (Bloomberg’s fiefdom in NYC is a notable exception). Moreover, they are members of the same district (sharing funding) to which rejected students will end up, unlike charters, who have no consequences for discriminating.

Comments are closed.