It takes time for charter schools to help under-educated students

I have written before about the often overlooked, but most important, finding from a study by a group called CREDO from Stanford University. That organization’s 2009 report shows that by the time students have spent at least three years in charter schools, they notably outperform their traditional public school counterparts.

Now, research from another college group, this time headquartered at Vanderbilt University, points to the same conclusion.

According to the National Center on School Choice at ‘Vandy:’

“…the data show that students who switched from a traditional public school to a charter school and stayed in a charter school for longer than one year experienced greater annual gains in mathematics than if they had stayed in the traditional public school. Compared with findings from previous studies of charter schools, these effect sizes are relatively large in magnitude.”

The ‘Vandy’ group also found charter school students did slightly better in reading, though the difference from public school performance was not statistically significant.

The new findings from Vanderbilt add to those from at least two other major research efforts that indicate any study on charter schools that ignores effects over time probably should be discounted as biased.

The evidence is becoming clear: charter schools need at least two to three years to turn around the impacts of bad teaching on incoming students, and it is simply unreasonable for researchers to ignore that rather common-sense situation.



The Vanderbilt study only looked at a three-year period of data from 2002-03 to 2005-06, so it is possible that even stronger impacts would be evident if the study looked at students who stay in charters for longer periods.

Certainly, stronger impacts over time have been found in other studies such as Stanford University researcher Caroline Hoxby’s examination of charter schools in New York City (See discussion starting on page IV-6) and a study from The Boston Foundation.

Hoxby found that students in the ‘Big Apple’ who spent their elementary school and middle school years in charters continuously moved ahead of their traditional public school counterparts in every grade from grade 3 to grade 8, winding up by the entry to high school with a huge educational advantage.

The Boston Foundation Study included this graph, similar to ones found in Hoxby, that show charter students get further and further ahead each year they stay in charters.


The Boston report says this graph plots the scores of lottery winners [shown by the green line for English Language Arts (ELA) and by the gray line (math)] to lottery losers scores (represented by the constant zero horizontal black colored axis line) over grades in middle and high school charter schools. The report concludes:

“The relatively steep upward slopes of the lines suggest that Charter School impacts increase over the course of school.”

In other words, charter middle schools in Boston provided stronger educations for students, and the amount of that extra education increased as students remained in the charter schools for longer periods of time.

These findings are very important to the current discussion about charter school impacts. Overall, a number of studies have shown mixed results for charter schools, but it is important to note that many of the studies are poorly done, and a number of the negative findings were based on student samples dominated by students with little time in charter schools. In fact, another ‘Vandy’ report says:

“…many of the studies that examine charter schools have weak designs.”

That sounds like me, but Vanderbilt researchers said it.

Thus, the findings of those negative studies may be due to faults within the research methodology rather than under-performance in charter schools.

As the battle over Kentucky’s charter school bill (Senate Bill 3) moves to the Kentucky House, I expect some of those poorly done studies are going to be cited by charter critics. We’ll try to let you know about the problems with those studies if this situation takes place.

But, there is enough evidence now from CREDO, Hoxby, Boston and now Vanderbilt to say that any study on charters that ignores student impacts over time is dubious.