Good day, CharterFolk.
Earlier this week, Andy, Derrell and I hashed out reactions to NHJ’s recent piece in the NYT …
… which, as I wrote in the hours after its publication …
… inexplicably failed to even mention charter schools, an omission I found particularly glaring considering how well charter schools are performing academically in NYC …
… including for the students for whom Hannah-Jones purports to have been fighting.
Earlier this week, the NYT, perhaps partially in response to the NHJ column, published another article …
… in which, in an omission I find even more egregious, they again left out mention of charter schools altogether, aside from noting that at least a few of their national experts thought charter schools should have been on the list.
Such is the state of the broader world’s recognition of the unique potential of the national charter school movement to fix our nation’s public schools.
Not even included in the top 30.
Below requiring students to read whole books …
And paying all teachers more regardless how well they teach …
And reporting student attendance better …
… when performance data keeps amassing that charter schools have far more potential for lasting change than any other idea on the list.
A couple of weeks ago …
… I highlighted the stunning progress that Indianapolis’s charter and innovation schools have made.
The state’s latest release of performance data makes what is happening even more plain – the charter and innovation schools excelling, the district-run schools, sadly, completely imploding.
At least from a grades 3–8 perspective, these graphs show it all.
Again, this is regression analysis. We give schools a predicted outcome based upon the students served, and then arrange results by performance bins, performance under prediction on the left, over on the right. Saturated colors reflect 2025-26 results, faded ones the year before.
Here is the story reported by halves.
District performance going in the wrong direction, innovation and charter schools going in the right.
Here it is by quartiles.
And by deciles.
For those of you on phones, you’re reading that right: nearly 40% of district students attend schools performing in the bottom tenth, with the percentage more than doubling from the prior year.
Meanwhile, nearly 60% of charter school students are attending schools performing in the top quartile, with the percentage having grown by 17 percentage points from the prior year.
Some of you may have seen last week that someone on X accused me of engaging in “causal salad” by showing the regression analysis for NYC.
Another accused me of ignoring “selection bias.”
So let me be clear.
I didn’t say in my NHJ piece, nor am I saying here, that anything I present demonstrates causality.
Nor am I claiming that this analysis can rule out selection bias.
But what we can say is that charter schools are doing far better than you would predict given the students that they serve.
And other studies in the past …
… have seen performance patterns like the ones we observe in NYC today and established causal effects.
Many have looked for evidence of selection bias over the decades and have found none.
In terms of presenting an immediate analysis of very recent data, this is the best we can do.
So let’s acknowledge what we’re seeing:
The overperformance of charter schools in NYC looks much like patterns other studies have found to be causal, with selection bias ruled out as the explanation.
Could this year’s pattern, unlike those examined in earlier studies, turn out to be the first to be found non-causal and primarily a function of selection bias?
Well, CharterFolk, I’ll let you reach your own conclusions on that one.
From where I sit, nothing impedes my central contention, which is that results are certainly strong enough for us to move forward with confidence that if we grow more high-performing charter schools, thousands if not millions of better student outcomes await on the other side.
Because these patterns of performance are not unique to either NYC or to Indy.
Perhaps no better example presents itself than New Orleans right now.
Though the quartile bins skew more toward the third quartile than the fourth than is found in Indy …
… with overall representation in the top half being 71.6%, and progress toward the top half year over year being nearly 11% …
… New Orleans is vying with Indianapolis for which city is seeing the fastest improvement of urban education in America.
And is there any possibility that these results in New Orleans are a function of “selection bias?”
Nearly every public school is a charter school!
Who is possibly being selected out?
And the one school that isn’t a charter is struggling mightily, leading the school board to consider holding that school accountable to the exact same standards that are applied to charter schools …
… which is exactly what should have happened where NHJ’s child languished in a clearly underperforming, unaccountable district school.
And while there is certainly variation by city, the broader pattern is very clear.
Again, no “causal salad” is asserted here.
But look at Newark.
Ninety-one percent of charter students attend schools performing in the top half in ELA. The district has six times the share of students in bottom-quartile schools as charter schools do.
Even in places where charter schools have been subjected to some of the most withering criticism, a similar pattern emerges.
Look at Detroit.
Charter schools’ inspiring representation in the top quartile nearly matches the district’s woeful over-representation in the bottom.
And even in places like Denver where all boats have clearly risen, charter schools have led the rise …
… while helping nurture the emergence of innovation schools whose performance also clearly exceeds district schools.
It’s against this backdrop that we find ourselves in a circumstance where the largest daily publication in the country has chosen to leave us off a list of 30 ideas for fixing our public schools.
Would I call that publication’s omission “causal” for all of society getting it wrong about the progress that charter schools have made and our potential to do even more?
Well, that would obviously be a true salad offense.
But CharterFolk, we have much further to go in getting the word out.
My sense is that it begins with we ourselves knowing our own story better.
Thus this post.
But it also requires getting off this ridiculous belief that we can’t say anything comparative about the performance of our schools relative to district schools.
We have tons of problems. No doubt. Not all of our schools are great. In many places where comparative analyses are favorable, we still have overall proficiency rates that are woefully too low.
But the performance of a movement that has grown to serve four million students deserves not merely to be among the 30 best ideas for fixing our nation’s public schools, but deserves to be the idea at the very top of that list.
Which puts it back on us to find a way to make our case.
It begins with data, and more compelling ways to present it visually. And ways to succinctly talk about what our results do and do not confirm without getting bogged down in the tedium of causal salad.
And nothing here suggests this is the only way to do it.
But it’s a way.
And the faster we can get to many, many other ways of talking about our patterns of over performance, the faster we can bring to an end the pattern of omission by many in our country who do the great, great disservice of not giving our movement the fair shake it deserves.
And that our kids and communities so desperately need.
I leave you with just a couple more.
Here, CharterFolk, is all schools in New Orleans before and after Katrina by halves …
… and by quartiles.
If you had said 21 years ago that this is what is within our potential to achieve, not many would have believed it.
And yet here we are.
Not perfection. Still much further to go, obviously.
But so, so much promise.
What we’re well on our way to bringing about in many other cities across the country.
And what we could achieve in every city in America over the 21 years if we get our acts together.
CharterFolk?
Let’s go!