Wishing for something does not make it true.

Fascinating and frustrating.  From Unlearning an Answer Charter schools deliver extraordinary results, but their political support among Democrats has collapsed. What will Biden do? by Jonathan Chait.  Chait is capable of fine insight on occasion but all his writing is heavily skewed by a strong partisan bias which manifests as an incapacity or unwillingness to deal with evidence as it is rather than as would wish it to be.  

In this long form essay, Chait is concerned that the Democratic Party apparatus is preparing to toss out the advantages of charter schools.  

I was intrigued by the claim that "Charter schools deliver extraordinary results" and totally unsurprised by the fact that "political support among Democrats has collapsed."  It was never high to begin with, dependent as the DNC is on education establishment and union donations.  The whole charter school movement has long had an ethos of giving parents a choice whereas the unions want parents to have no choice of where to send their children.

Chait's wife is apparently a Democratic convert to the value of charter schools and clearly Chait is convinced as well.

But what of these "extraordinary results?"  I have been a supporter of charter schools on a philosophical basis for years.  I was on the board seeking to found a charter school.  I once owned a company whose mission was to enable parents to improve the odds of their children becoming enthusiastic readers.  In Boy Scouts of America, I was a District Commissioner for a city with traditional big city urban poverty issues and conducted some reading programs for troops.  

I have never lost my enthusiasm for charter schools but my expectations became more grounded over time.  Some do wonders, some not.  But all of them demonstrate greater parental engagement with the education process and even if scores don't move much, there is value in all citizens more engaged with and positive about their local governance.  

In all my time, charter schools were very much affiliated with conservative or Republican initiatives in the face of Democratic Party opposition.  But apparently there has been some support in some corners, which Chait explores.

But the last time I did any deep dive into the performance metrics of charter schools, perhaps a decade ago, the evidence was reasonably mixed.  Better parental and student satisfaction, slightly higher college matriculation, perhaps a third of a standard deviation improvement in standard test performance.  Good, but not a solution to the full standard deviation gap which has exercised Democrats for decades.  

I was hopeful that Chait might have compelling evidence of a striking improvement.  But no.  There are snippets here and there but nothing supporting the claim that the Black-White performance gap has been closed.  Chait overclaims:

• Charter schools have produced dramatic learning gains for low-income minority students
• In city after city, from New York to New Orleans, charters have found ways to reach the children who have been most consistently failed by traditional schools. 
• The evidence for their success has become overwhelming, with apolitical education researchers pronouncing themselves shocked at the size of the gains. 
• What was ten years ago merely an experiment has become a proven means to develop the potential of children whose minds had been neglected for generations.  
• The rapid progress in producing dramatic learning gains for poor children, and the discovery of models that have proved reliable in their ability to reproduce them, is one of the most exciting breakthroughs in American social policy. 
• Many of those gains are huge, effectively wiping out the educational inequities that have persisted for the entire history of American schools.
• But that wouldn’t explain the fact that, as researchers have found, urban charter students are producing higher achievement on state high-school exams and SAT scores and are enrolling in more Advanced Placement courses and scoring higher on those exams. 
• It [CREDO] found urban charters on average gave their students the equivalent of 40 additional school days of learning in math and 28 additional days of learning in reading every year. 

OK.  I want to believe.  I would wish to turn my back on all the prior evidence I have read over the years.  So where is the empirical reporting of improvements?  Don't tell me they have improved, tell me by how much!

Chait doesn't provide much of that evidence.  He clearly is enthused about charter schools but he doesn't muster much evidence for their purported magnificent gains.  

He cites two case studies.

• As mayor of Newark between 2006 and 2013, Booker had overseen a major charter effort; his goal, he said at the time, was to make the city the “charter-school capital of the nation.” The project worked. A recent study documenting the gains found that Newark’s students, whose performance on statewide tests had once ranked in the 38th percentile, had vaulted nearly 40 points. Newark’s charter-school students now exceed the state average in math and language, an extraordinarily impressive result given their high poverty rate. 
• Recently, Tulane economist and education specialist Douglass Harris published Charter School City, a detailed study of what happened in New Orleans after Hurricane Katrina, when reformers turned the failing public-school system into a network of charter schools. The city’s mostly poor black students have closed the achievement gap with the rest of the state, vaulting over every possible measure: state tests, ACT tests, high-school graduation rates, and college enrollment. 

FOr Newark, Chait is skimpy on the numbers but does provide a link to A New Baseline: Progress in Newark’s District and Charter Schools from 2006 to 2018.  Going to the source, the evidence may be more nuanced than Chait's claims.  It is still hard to get a read on actual performance.  Improvement, yes.  Performance . . . ?  

• Between 2006 and 2018, when compared to other low-income cities and towns in New Jersey, Newark’s citywide average test score rank has improved from the 39th to the 78th percentile in both ELA and math. [this is both public and charter schools]
• The share of Black students in Newark attending a school that beat the state proficiency average in their grade has more than quadrupled, from 7% in 2006 to 31% in 2018. [Indicates improvement but not level of achievement]
• Newark’s charters have shown particularly strong test score gains, and in 2018, they beat the state proficiency rate for the first time in both math and ELA.  [Seems like clear evidence]
• In high school, the citywide four-year graduation rate has risen from 63% in 2011 to 77% in 2018 and closed the gap with the state by seven percentage points.

What we really want to know is whether they score on the statewide tests at the same level, do they score the same on SATs, do the graduate at the same rates, do they matriculate into higher education at the same rates.   

If any of that were true, you'd expect it to be trumpeted.  The very absence of this barking dog is suggestive.  Especially when you pay attention to their note:

• When compared to all cities and towns in New Jersey – not just those with similar populations – Newark’s citywide average test score rank is in the 14th percentile (though up from the 4th percentile in 2006).

OK - that seems clearer.  Newark schools (public and charter) are doing better but they are nowhere near closing the gap.  Charters are an unquantified contributor to that improvement.  

None of which sounds like the extraordinary results being championed by Chait.

What about New Orleans?  Regrettably Douglass Harris' Charter School City is not readily available digitally.  

But the claim

The city’s mostly poor black students have closed the achievement gap with the rest of the state, vaulting over every possible measure: state tests, ACT tests, high-school graduation rates, and college enrollment. 

seems hard to reconcile with this report from Compare 2019 New Orleans school ratings by Matt Jewson

Eight city schools received A’s and 11 earned B’s. But the vast majority of city schools earned a C or a D. Twenty-six earned a C and 24 received a D. 

Twelve schools got an F rating and four of those closed at the end of last school year. That means eight remain open this fall. Of the eight, only two are up for charter renewal. The district’s release states staff have already visited those schools. 

Many New Orleans school grades were brought up somewhat by a new formula adopted last year that incorporates year-over-year academic growth. Growth accounts for a quarter of an elementary school’s grade, 12.5 percent for high schools. The rest — called the assessment grade — is based on academic performance in a single year. 

Of the city’s roughly 80 schools, the majority got an F on the assessment portion of the state ratings.

[snip]

Overall, the NOLA Public Schools district received a C letter grade with a district performance score of 67.8. That was a slight uptick from its C rating of 66.2 last year. The district fell shy of this year’s state average of 77.1, which equates to a B. Schools are measured on a 150-point scale. 

So once again, improvements but overall performance still below that of the state and 45 of the 78 schools had a performance score based on objective tests of F.  

Yikes!  These are the bases of Chait's claims.  I clicked through on the other links as well and there were similar issues of nuance and clarity of absolute performance.

Again, I am not criticizing the charter schools.  I support them.  I am just trying to reconcile what I recall from a decade ago when I was last immersed in the data and the "clear" case Chait is making for their success.

The data doesn't seem to be there.  Charter schools do raise performance, they do engender parentaaml and student support.  But miracle machines?  Still not seeing it.

What I do see is the typical advocate's sin of cherry-picking.  There is a subsegment of charter schools which do have a program of behavior encouragement which does also seem to translate into improved performance.  Chait captures some of this (emphasis added):

The fact that charters can produce dramatic learning gains is no longer in serious question. Why they can do so is a matter of some conjecture, but the most successful urban school models share some basic practices. Charters tend to have less money than traditional public schools, and so they focus their resources on longer learning time — extending both the school day and the school calendar. They invest in intensive tutoring, and they don’t spend as much as traditional schools on administrative staff or gyms, cafeterias, and other amenities. They instill schoolwide cultures of respect for learning and orderly environments, so that one or two disruptive students can’t bring classes to a standstill. The best charters tend to focus on high expectations for students, driving home the expectation that every student will attend college. Schools in the Knowledge Is Power program network name each classroom after the teacher’s alma mater, name every class after its expected year of college enrollment, and conduct visits to university campuses — among other methods that might seem hokey if you grew up the child of college graduates.

The final element of charters’ formula is inescapably controversial. They prioritize the welfare of their students over those of their employees, which means paying teachers based on effectiveness rather than how long they’ve been on the job — and being able to fire the worst ones.

Yes, a small proportion of charters have these attributes and these attributes do drive improvements in performance.  If Chait were to have argued that charter academies with a sharp focus on traditional learning behaviors and priorities demonstrate both sharp improvements in relative performance and in absolute performance, he would be on good ground.  But that is not his argument.  He argues for all charters and there the evidence simply does not support his argument.  

But look at those behaviors and cultural priorities.  These are all conservative themes which are part of what makes this essay so intriguing but also so flawed.  

Chait clearly believes in charter schools and does not want to see his party turns its back on the single policy that has demonstrated predictive performance improvement - adoption of mainstream cultural values and traditional norms such as respect for authority, aspiration for education, intense and sustained effort, prioritization of student education over the interests of other possible stakeholders.  

I am afraid there is no way for Chait to square the success of such academies with support by traditional Democratic Party stakeholder groups.  

Chait seems to clearly know how precarious his position is.  Charters have long been championed by conservatives and/or Republicans.  But Chait starts his essay as if charters came into their own with the Obama administration.  

In the dozen years since Barack Obama undertook the most dramatic education reform in half a century — prodding local governments to measure how they serve their poorest students and to create alternatives, especially charter schools, for those who lack decent neighborhood options — two unexpected things have happened. The first is that charter schools have produced dramatic learning gains for low-income minority students. In city after city, from New York to New Orleans, charters have found ways to reach the children who have been most consistently failed by traditional schools. The evidence for their success has become overwhelming, with apolitical education researchers pronouncing themselves shocked at the size of the gains. What was ten years ago merely an experiment has become a proven means to develop the potential of children whose minds had been neglected for generations.

And yet the second outcome of the charter-school breakthrough has been a bitter backlash within the Democratic Party. The political standing of the idea has moved in the opposite direction of the data, as two powerful forces — unions and progressive activists — have come to regard charter schools as a plutocratic assault on public education and an ideological betrayal.

The shift has made charter schools anathema to the left. 

Reading only this, and it is true that the Obama administration supported charters carefully, you would never know that charters and education were a signature of the preceding Bush administration.  Bush increased charters 100% from beginning to end of his administration whereas Obama did so by only 50%.   There is no mention by Chait of No Child Left Behind under Bush.

There is this exceptionally odd paragraph:

The departure of Obama, who gave the movement some protective cover, accelerated the Democratic Party’s anti-reform lurch. And then, of course, the advent of Trump presented unions with the ideal foil. His secretary of Education, the arch-conservative Betsy DeVos, backed charters. (Prior to her appointment, in Michigan, she had aggressively lobbied for a charter sector that was one of the country’s most loosely regulated and poorest performing.) DeVos also championed private-school vouchers, making it easier for reform opponents to package “school choice” as something that was entirely wicked. Liberal supporters of charter schools immediately saw DeVos’s appointment as a catastrophic blow to their cause.

Unions gleefully announced plans to make DeVos the face of charter schools and drive education reform out of liberal politics. 

DeVos has been the most openly and consistently supportive public leader of charter schools and education choice, at least back to the 2000s lending her philanthropic support to the cause and then her political support.  She is supporting the outcome Chait wants but he never acknowledges her contribution.  Indeed, he appears to fault her for engendering Democratic party opposition to the policy he wishes to advance.

Chait occasionally writes some fine pieces.  This, however, is not one of them.  Partisanship, mean  spiritedness, cherry picking of data, unsupported claims, failure to represent reality, are all on display.  He finds himself at odds with his party on something he holds to be critically important and suffers strained cognitive dissonance.  

I had hoped to find that my decade of inattention provided the evidence for charter effectiveness which Chait heralds.  But it is not there in what he has written.