In yesterday’s USA Today there was an article with the enticing title “Charter schools’ ‘thorny’ problem: Few students go on to earn college degrees” This article was shared widely by the pro-teacher, pro-public school crowd. And though the title does seem to support what many of us have been saying over the years, based on what they say in the article, I see it as something that can easily be quoted by the pro-charter crowd as evidence that charters are, in general, working.
Statistics for charter schools as a whole are hard to come by, but the best estimate puts charters’ college persistence rates at around 23%. To be fair, the rate overall for low-income students – the kind of students typically served by charters – is even worse: just 9%.
So if the rate of college completion for low-income students who attend charter schools is really 23%, that does sound like a big improvement over the non-charter rate of 9%.
The article goes on to highlight two charter chains who claim to have, respectively, a 45% and an 87.5% college completion rate. The 45% chain was KIPP. I remember a few years back when they first started saying this and I argued with Richard Barth, a co-CEO of KIPP, that you really can’t compare the rate KIPP publishes with the 9% statistic since the KIPP rate only applies to students who graduated KIPP and ignores the KIPP students who leave the school before reaching 12th grade. Statistically speaking, the KIPP students are a ‘biased’ sample.
He wasn’t really interested in debating this, here is the exchange:
The real heroes of the article are the Democracy Prep charter chain. They claim an amazing 87.5% college completion rate. (There is not mention in the article about the recent incident where a Democracy Prep student threatened another student at gunpoint over a dispute about a Chicken McNugget.)
Having the KIPP numbers and the Democracy Prep numbers really make this into a pro-charter piece. It basically says that some charters are struggling to get kids ‘to and through’ college, but the really good charter chains are doing well with this. So the conclusion isn’t to slow down charter proliferation, but to only expand the really good charters like KIPP and Democracy Prep.
New York State has a pretty good public data system, so I investigated the numbers for Democracy Prep’s first cohort, the ones that 87.5% of their graduates are on track to graduate from college. What I found was that in 2006-2007, they had 131 6th graders. According to their testing data from that year where 127 students were tested, there were 63 girls and 64 boys tested. Also, of the 131 students, 80% were Black while 20% were Latino.
Six years later they had 50 12th graders. This represents just 38% of the original 131 students. Of those 50, 13 were boys and 37 were girls. So they went from 50% boys to 33% boys. Also of their 50 students, they went from 80% Black in 2006 to 66% Black in 2013.
The public New York State data page for Democracy Prep’s 2006-2007 data is here and the 2012-2013 data can be found here.
So Democracy Prep does not deserve to held up as a model for how to get low-income students through college when they can’t even get them through high school. And USA Today, if they want to write an article about how Charter Schools are not a silver bullet for education, they should not publish misleading statistics that support the argument that they are.
Gary Rubinstein's Blog 
KIPP and so-called reformers lying. Yet again.
“Statistics for charter schools as a whole are hard to come by”. Their most impressive characteristic, clearing the way for invented stats which amounts to lies. It does not happen by accident.
KIPP= kids in prison programs
Pingback: Gary Rubinstein on the Miraculous Data from Charter Schools that Is Miraculously Distorted | Diane Ravitch's blog
Has anyone ever looked at charter school student performance vs public school student performance, controlling for things like diet, parent income or sleep?
It seems to me like all we do in these debates is look at the outputs when the inputs may arguably be more important.
That’s not possible because No Excuses!
Assuming that all 50 12th graders graduated, I can almost guarantee you that the grade 6 to 12 completion rate is much lower than 50/131. It’s doubtful that all 50 12th graders were in the original 6th grade cohort, esp. since the transition from middle to high school (gr8>9) is when lots of transfers occur. Easily 10 of those 50 were “replacement” students. Then, there were the students who were admitted after the 6th grade count and left before the 12th grade count. So, 50/131 is probably more like 50/150.
I would agree, Gary, that the USA Today article was in important respects misleading, but I fear you’ve added somewhat to the confusion.
You quote that article as stating:
“Statistics for charter schools as a whole are hard to come by, but the best estimate puts charters’ college persistence rates at around 23%. To be fair, the rate overall for low-income students – the kind of students typically served by charters – is even worse: just 9%. For low-income, high-minority urban public schools, most comparable to charters, the rate is 15%.”
I think typically, in academic literature both “college completion rates” and “college persistence rates” are relative to the number of those who had enrolled in a postsecondary school not to relative to all high school graduates. (Though as this article https://www.insidehighered.com/views/2011/10/10/completion-rates-contex points out it can be misleading to exclude from community college completion those who have transferred to Baccalaureate Institutions).
However, the NSC research center in its data (https://nscresearchcenter.org/hsbenchmarks2015/) reflected in the USA Today article seems to use “college completion rate” as relative to all high school graduates, with “persistence” relating just to those who have enrolled in a postsecondary institution. The USA Today’s references in the quoted paragraph to 23% and 15% seem to derive not from “persistence rates” but instead to the figures found at the very bottom of each of these spreadsheets
https://nscresearchcenter.org/wp-content/uploads/TablesandfiguresPUBLICCHARTERSCHOOLS.xlsx
https://nscresearchcenter.org/wp-content/uploads/TablesandfiguresPUBLICNONCHARTERSCHOOLS.xlsx
If I understand NSC correctly, the charts found there show:
23% of the graduates of the studied charter schools completing 4-year programs and 5% completing 2-year programs within 6 years
with
15% of the graduates of low income, high minority, non-charter urban public schools completing 4-year programs and 7% completing 2-year programs.
And, as you suggest, the reference to 9% in that USA Today paragraph is highly misleading as it is calculated not in reference to all who enroll in postsecondary institutions or to all HS graduates but instead to a greater pool that includes all dropouts.
http://www.pellinstitute.org/downloads/publications-Indicators_of_Higher_Education_Equity_in_the_US_45_Year_Trend_Report.pdf#page=33
Two pages later, table 5b provides an alternative perspective that, like most college completion rate analyses, starts with a base of those who have enrolled in a post secondary institution rather than all HS graduates. In neither case, is there an appropriate basis for comparison with the cited NSC data.
But, Gary, your attempt to derive a “High School Completion Rate” by dividing the number of seniors by the number of 6th graders seems faulty to me as you not only merge high school with middle school but don’t consider how many students at each grade may be repeating the year, how many may successfully complete school but take an extra year or two to do so. And Erich Martel is correct to suggest that incoming transfers could further muddy your waters.
I discussed these issues with Jersey Jazzman in this series of postings:
http://jerseyjazzman.blogspot.com/2016/09/massachusetts-charter-schools-and-their.html
http://jerseyjazzman.blogspot.com/2016/10/charter-school-attrition-in-ma-reader.html
http://jerseyjazzman.blogspot.com/2016/10/more-about-attrition-rates-in-boston.html
My three-part response to the last of those is in the comments section
Would welcome your thoughts if you have a chance to read that exchange.
– Stephen Ronan
Pingback: CURMUDGUCATION | Mister Journalism: "Reading, Sharing, Discussing, Learning"
Just in case you missed it when it first came through, I thought I’d mention that my comment submitted March 16 is still “awaiting moderation”.
Ah, well, it survived Diane Ravitch’s moderation here: