"Heroic" charter school stories are the stock-in-trade of the "reform" industry. These tales take the same basic form:

1. Present some a-contextual data point -- usually having to do with a score on some sort of standardized test -- as proof of a charter school's "success." 
2. Posit a causal relationship between the structure and/or practices of the charter school and the data point presented. 
3. Imply or state outright that the many factors outside of school classrooms that we know for a fact contribute to student outcomes can be overcome simply by following the model above.

There are two reasons these stories make me nuts. The first is the unwarranted braggadocio inherent in the genre: "We rock, you suck!" There's almost a twisted negative correlation between a charter school's leaders' humility and the unacknowledged factors -- attrition, extra funding, peer effects -- that contribute to its "success."

But there's another, more pernicious effect that comes from telling tall tales about charter "success": they keep us from having the conversation we need to have about American schools as engines of social replication.

Case in point:

I think it’s time that the narrative in our country began a fundamental shift – away from “How is it possible to succeed amid all these challenges?” to “How is it possible to remain failing when there are so many good solutions?” I don’t care how stubborn poverty is — failing schools can never be an acceptable status quo. 

So here’s another story the press hasn’t been quite so aggressive about covering. It’s about educators who — tired of the lone wolf Stand and Deliver mythology — decided to build something different, and to do it together. 

Instead of a “sorry, but,” it’s a story of “yes, and.”

Did it take better training and support for teachers? Yes, and — a character-building discipline system? Yes, and — a curriculum that challenges students to think at the highest levels? Yes, and — regular follow-up with students even when they are in college? Yes and...I think you get the idea. 

The thing is, there are schools out there right now — as you read this and as I write it — that are giving lie to the implication that school improvement needs to wait for the country to heal poverty.

We can’t wait, and we don’t have to. [emphasis mine]

This comes to us from Stephen Chiger, the "Director of Literacy" for Uncommon Schools. Uncommon is the charter chain where the current SecEd, John King, cut his reformy teeth. Uncommon's charters are the quintessential "no excuses" schools, boasting very high student suspension rates (in contradiction to current USDOE policy).

You'll note that Chiger has already taken care of points #2 and #3 from my list above. His schools get their gains from "better training and support for teachers." Those high suspension rates are part of a "character-building discipline system." His schools "challenge students to think at the highest levels." All we have to do is do what he and his colleagues are doing, and we can skip addressing poverty. Simple, right?

Before we continue, let me state the obvious for the willingly obtuse who look for any excuse possible to misread my work: yes, schools can and should improve in the absence of larger societal changes. Yes, children from disadvantage can achieve at high levels. Yes, there can be a place for "choice" in school systems. No one disputes any of this.

But, as I said above, when the authors of these stories cherry-pick their facts, they keep us from having the conversation we need to have:

Take the most recent PARCC exams in New Jersey. About 41% of the state’s 11th graders met or exceeded expectations on the test. 

In Essex County, high-income Millburn High School (2.2% economically disadvantaged) saw 57% of students scoring proficient or advanced on the assessment. The juniors at Livingston High School (1.5 % economically disadvantaged) earned 56.5%.

A few miles away, the juniors at Newark-based North Star Academy (83.7% economically disadvantaged) earned an 80.6% pass rate.

The school achieved it through more systems and strategies than I could possibly recount here. The good news is that I don’t have to. 

Uncommon Schools, which manages North Star, publishes books — and books and books – to share its practices. It regularly films teachers to “show” and not just “tell.” It opens its doors to hundreds of visitors every year. It runs professional development for external audiences and sells trainings so they can turn-key them locally. 

And, there’s good evidence it’s working. [emphasis mine]

Whoo, boy. Where to begin?

Let's start by acknowledging something my research work, and the work of Dr. Bruce Baker*, has acknowledged for some time: compared to the rest of the state, North Star does, indeed, "beat the odds" on test based outcomes. They are not nearly the standout you'd think they were from their reputation in the credulous press; many schools in NJ "beat the odds" regularly. But if we are to judge North Star simply by its test-based outcomes, we should acknowledge their performance is very solid.

The real question, however, has always been: "Why?"

This shows the "cohort attrition" for the Class of '16, the class referenced in Chiger's post. Livingston and Millburn are affluent communities with reputations for good schools; it's no wonder the size of each cohort remains stable as they travel from grade to grade. Newark, like most communities with large numbers of families in economic disadvantage, is prone to see more transience, and cohort shrinking due to a larger drop-out rate.

But as we have shown time and again, North Star has a particularly high cohort attrition rate. Clearly, they don't "backfill" -- bring in students to replace ones who've left -- much. Is this attrition selective? Are they counseling out the lower performers? We can't say for sure from the data, but we do have some clues:

I've shown this graph many times: as North Star's cohorts shrink, their test scores rise. No, this is not direct proof of selective attrition. Of course, neither is this:

North Star has a very high suspension rate -- quite typical for the Newark charter schools that enroll large proportions of black students (and not typical for the Newark district schools that enroll lower percentages of black students -- anyone OK with that?).

Again: the effects of attrition, lower enrollments of special education & LEP students, greater resources through philanthropic giving, "no excuses" discipline, etc. on student test outcomes at certain charter schools have been well-documented by me and others.

But what Chiger does here is something I've started to see creep into the rhetoric of the "reform" industry lately: a direct comparison between "successful" charter schools and suburban public schools.

For those of you not from Jersey, let me assure you that Chiger did not just pull Livingston and Millburn out of a hat. Livingston is the alma mater of Governor Christie, and Millburn is consistently rated as one of the top high schools in the nation. Considering the vast differences in the schools'  student populations, Chiger is setting his sights very high.

Which is fine... so long as we have an honest discussion about whether North Star is "giving lie to the implication that school improvement needs to wait for the country to heal poverty." Is it? Let's start by looking carefully at Chiger's data point:

There is no doubt: North Star beat the wealthy 'burbs on the Grade 11 ELA test. But did every student at each school take the test?

  

Millburn and Livingston -- and Newark -- all had much lower rates of test participation than North Star.** We all heard the stories about high school students blowing off the tests; in fact, many who did take the test probably just filled in nonsense answers. Why? Because they knew it didn't matter.

The PARCC had zero effect on the grades or graduation status of high school juniors last year, which is why so many didn't even bother taking the test. But what happened when the juniors took a test that actually counted for something?

We don't have last year's SAT scores yet; these are from two years ago. But they make the point: when the test matters, the outcomes change.

Let me be very clear about something here: in no way do I think SAT scores are a reflection of the effectiveness of a school. SATs are highly correlated with socioeconomic status, and there is little reason to believe they are a valid instrument for assessing school performance. Which is exactly the point.

It is absurd to think that North Star, or any school that serves large proportions of disadvantaged students, can fully overcome the disadvantages of economic and social inequity that are reflected in educational outcomes.

I have nothing but respect for people, including those who work in charters, who teach disadvantaged children. Without question, we as a society have tended to set low expectations for children from traditionally disadvantaged communities, consigning them to the lower rungs of the economic ladder. We most certainly should be demanding better from our urban and poor rural schools. As I said above, we should always be demanding better of all schools.

But anyone who makes the case that reading Doug Lemov's book (more on that in a bit) is what we need to ameliorate the disadvantages urban students and schools face is telling a whopper. The advantage that affluent suburban schools and students enjoy is structural. It will not be wiped away by "choice." It will not be wiped away by "no excuses." It will not be wiped away by a "character-building discipline system" and "better training and support for teachers."

The structural inequities in our schooling system will only be addressed when we start having an honest conversation about what is really happening in and out of our schools. Poverty, segregation, and resource inequity do matter. But heroic charter school stories are designed to keep us from addressing this reality.

However, let's turn that notion on its head, and use Stephen Chiger's piece here to delve further into the issue. What is the reality of the difference between the schooling a Millburn student receives and that of a North Star student? Those PARCC participation rates and SAT scores give us a clue, but they are just the beginning.

Stand by...

                     

* As always: Bruce is my advisor in the PhD program at Rutgers' GSE.

** I had to fudge this a bit, because North Star's test taker data was suppressed in the NJDOE files. The denominator for the North Star figure is the enrollment number for Grade 11 at the school. For the others, it's the registered student number from the assessment file itself. I think it's more than fair to assume that the differences here, which are very large, couldn't be explained by using two different but related data sources.