A Second Pass at the AzMERIT Scores


  • Courtesy of commons.wikimedia.org

Yesterday I posted about my first impressions of the newly released AzMERIT scores. Passing rates dropped significantly statewide from the 2014 AIMS test, which has everything to do with a tougher test and higher cut scores and nothing to do with Arizona student achievement going downhill. The most obvious observation is that the highest scoring districts in the Tucson area are in the high rent districts, and the lowest scoring districts are in the low rent districts. And I observed that the passing rates fell further in the Flowing Wells and Nogales districts than in others, which is interesting because the previous AIMS scores of those districts were frequently cited to show those districts beat the odds and scored higher than you might expect given the socioeconomic status of their students. The new scores may or may not indicate that those two districts were overrated in the past.

Since then, I've dug deeper into the daunting piles of AzMERIT data the Department of Education put into its downloadable spreadsheets—far, far more data, by the way, than was ever supplied in previous years—and I've compared them to the scores on the 2014 AIMS test. I'll give you some of the information I've pulled together and a few conclusions I've drawn from the data.

First, let's look at how the passing rates changed in the Tucson-area and Nogales districts from the 2014 AIMS to the 2015 AzMERIT. All of them went down, just like the rest of the state, but they went down at different rates. (Note: I had to approximate the passing rates of the districts on the 2014 AIMS test because I couldn't find that number in the data, but I used the same method of approximation for all the districts, so if the results aren't perfect, they should be fairly consistent.) How much did district passing rates drop from AIMS to AzMERIT? The largest drop was in the Nogales district: 48 points, from an average of 73 percent to 25 percent passing. That drop could be explained by a cheating scandal in the district which inflated its numbers. Three districts dropped 40 points: Amphi, Flowing Wells and Sahuarita, followed by Sunnyside at 39 points and Marana at 38 points. Vail and Tanque Verde dropped 35 points and TUSD dropped 34 points. Catalina Foothills had the lowest dropoff, at 25 points.

What does all this mean? You can play with the numbers all day, but here's one conclusion I arrived at. Even when the number of points the districts dropped were similar, the lower income districts with lower passing rates actually fell farther. I'll use Vail and TUSD as an example. Vail fell one point more than TUSD, 35 points vs. 34 points, which makes it look like they took a similar hit. But in 2014, Vail's passage rate was 28 points higher than TUSD's, 89 percent vs. 61 percent. So while they fell about the same amount in absolute terms, TUSD's passage rate fell by 57 percent while Vail's fell by 41 percent, a 16 point difference. (As an example, think of two people whose incomes are $100,000 and $60,000. If both incomes drop by $40,000, it's a far larger hit for the person at $60,000.) The same pattern follows elsewhere. The districts with lower income students, which also had the lowest scores on the 2014 AIMS, lost more ground than the districts with higher income students, even when the actual point drop was similar, because the high rent districts started with higher passing rates than the low rent districts. 

Months ago I predicted that the drop in scores this year would be greater in low income than high income areas, which is what has happened. I also said that it doesn't mean the achievement gap between the groups has increased. It's what happens automatically when you raise the cut scores. It would have happened even if they kept the same AIMS test and raised the number of points it takes to pass.

I went through the 2014 AIMS and the 2015 AzMERIT passing rates at individual TUSD schools and compared how far the schools dropped. Almost without fail, the TUSD schools that started with the highest passage rates on AIMS had the smallest drop on the AzMERIT—less than 50 percent—while the schools with the lower AIMS passage rate had the greatest drop—generally more than 60 percent and sometimes as high as 70 percent. Here are a few examples. On the high end, Borman Elementary had an overall 83 percent passage rate on AIMS and a 55 percent rate on AzMERIT, a drop of 34 percent. On the low end, Maldonado Amelia Elementary had a 49 percent passage rate on AIMS and a 15 percent rate on AzMERIT, a drop of 69 percent. On the high end, Annie Kellond Elementary fell from 80 percent on AIMS to 46 percent on AzMERIT, a 42 percent drop. On the low end, Cavett Elementary dropped from 44 percent on AIMS to 14 percent on AzMERIT, a 68 percent drop.

The increased standardized test score gaps between the educational haves and have nots have nothing to do with an increased achievement gap. But be ready for the faux gaspings which are sure to come from conservatives: "Oh my God, it's even worse than I thought! And it keeps getting worser and worser! Just look at this years' scores!" Which means, of course, they'll be even more adamant that we need to give less money to those "failing schools" and more to the "successful schools" in high rent areas, and, of course, more money to charter schools and private school vouchers.

Comments (13)

Showing 1-13 of 13

Add a comment

Add a comment