With all the recent discussion nationally with the appointment of Betsy DeVos as Secretary of Education and at the state level with the election of Mark Johnson as Superintendant of Public Instruction, I have seen a lot of discussion regarding the merits of charter schools compared to traditional public schools. Advocates for charter schools claim they out-perform traditional schools and are generally better educational environments while those opposed claim they gentrify schools and are not accessible to all students and perform better because they have fewer enrolled students with special needs and impoverished students.
With all of the discussion, there is little in the way of actual facts and data being cited to support the position that charter schools are actually performing better than their traditional public school counterparts. It is clear that there are fewer low-income and special needs students enrolled in charter schools and this could explain the differences in performance metrics (Look here for that information). With this in mind, I wanted to look at the actual metrics and see what differences existed and how those differences might be affected by a student’s social economic status (SES). I used the publicly available 2014-15 state reports you can see here.
I was truly surprised by the results. The data clearly shows a lack of significant difference between the average charter school results and those of traditional public schools. I examined the eight different metrics listed below:
- Math Scores
- Reading Scores
- EVAAS Growth Scores
- Science Scores
- ACT Scores
- Cohort Graduation Rates
- School Performance Grade
- Overall Achievement Score
The only two metrics that showed statistically significant differences were reading scores and cohort graduation rates. Reading scores were about 6.6 points higher among charter school students and the cohort graduation rates were approximately 5.7 points higher in traditional public schools. All other metrics display no significant differences.
The graphs below show the side-by-side comparison of these metrics and the results from the statistical analysis for those interested:
Charter Schools have higher reading scores, but it is unknown whether these students have higher reading levels as a result of the schools or other environmental or social factors. SES status has significant effects on reading ability so this difference can be hypothesized to be attributed to the significant differences in SES between these school types. This would have to be tested and analyzed to determine; however, there is not enough evidence currently available to draw conclusions either way. Traditional public schools have better cohort graduation rates which may be a result of students leaving charters prior to graduation for a myriad of reasons. Regardless, both of these significant differences need to be investigated further and there is not enough evidence to determine the cause of these differences.
The question remains: Do North Carolina Charter Schools Perform Better? According to the data used in this analysis, no. There is a difference in reading scores that favors charter school students, but all other factors have no significant differences. Given the advantages charter school students generally start with (parents who can get them to and from school, not requiring the school to provide meals, the motivation and support to apply in the first place, etc.), I was expecting North Carolina charter schools to out-perform traditional public schools. The results were surprising and makes me wonder why there are so many people saying how much better NC charters are. The evidence does not support that conclusion and it appears as though there is no real difference between them.
I am happy to entertain data to the contrary, but my conclusion is that charter schools have not shown that they have a better way of delivering education. The only metric in which charter schools show better results can be logically explained with a cursory understanding of the factors that affect literacy rates among school-aged Americans.
Given this information, my question to charter school advocates is: Why should we continue to divert funds to charter schools if there are no discernable differences in student outcomes?
My analysis of the reason for the reading score differences can be found here.