(Guest Post by Matthew Ladner)
So the 2017 NAEP is underway as we speak. As the NAEP released all six 2015 exams a cornucopia of good news spilled out for Arizona as a whole and (especially) for Arizona charters. Statewide Arizona has been leading the nation in 4th to 8th grade cohort gains since 2009. That’s a pretty good measure of overall school quality for those grades, as the demographic profile of the cohort isn’t likely to change much for a cohort between (for instance) the time when they were 4th graders in 2011 to when they were 8th graders in 2015. Kids will come and go of course, but absent a DC level of gentrification the academic ability of those coming and going should not skew heavily in a particular direction. You also have a measured amount of sampling error in both the 4th and 8th grade measures, but these are quite modest.
Arizona’s charter schools rocked all 2015 NAEP exams in the fashion of a New England state. This is quite impressive given the very modest level of per-pupil spending in Arizona charters, the majority minority student population etc. Sampling error is a bigger issue with regards to charter schools, but as you can see in the figure above, the state’s AZMerit exam tells us a story very similar to NAEP, and does not involve sampling. AZMerit in short provides backup to the findings in NAEP.
The figure above should lend confidence to Arizona charter leaders that they are likely to rock the NAEP again in 2017. The figure shows 4th and 8th grade NAEP scores from 2015 (proficiency rates) and then the proficiency rates for districts and charters from 2016 on AZMerit. These tests are not scaled exactly the same so you would not expect them to match up exactly, but the similarity in the pattern provides confidence that 2015 was not a fluke. Some of the differences between Arizona charters and districts can be attributed to differences in student demographics, although not nearly so much as sometimes are imagined, but New Hampshire doesn’t have any excuses for losing out to Arizona charter schools.
On the final set of columns on the right, for perspective, Arizona charters scored just a smidge below the highest performing state (MA) while Arizona districts would be right around the national average. Drawing random samples of students leaves room for goofy variation in subgroup scores, but there isn’t any reason to believe such goofiness would skew the same way in six different samples unless one wants to believe the NAEP is stacking the deck for charter schools in Arizona for some nefarious reason. While they were at it, the same conspirators would have to have infiltrated the AZMerit as well.
Absent goofiness and based on what we see in the AZMerit, I’m expecting Arizona charter students to CeleNAEP Good Times again in 2017. The statewide trend will be of even greater importance. Let’s see what happens next.
Jay P. Greene's Blog 