With the CMAS results being released today, it is worthwhile to take a few minutes and consider an example of how the data made available by standardized tests is being used by education researchers. For several years now, Stanford University’s Center for Research on Academic Outcomes (CREDO) has been publishing reports on the performance of charter schools across the United States. This year, CREDO’s focus has additionally been on the overall performance of city-wide school systems in America’s major metropolitan areas.
CREDO’s city-level studies all follow a similar format and model in comparing average one-year learning gains on standardized test scores to statewide averages in their respective states. In additional to an overall comparison of students in a particular city to their peers across the state, each study also features a breakdown by school characteristics (charter, innovation, or traditional public), and by student characteristics (gender, race, poverty status, ELL status, and special education status); each student characteristic subgroup is then also broken down into the types of schools these students attend.
So far, the cities CREDO has examined are Baton Rouge, Camden, NJ, Indianapolis, Kansas City, Memphis, New Orleans, San Antonio, St. Louis, Washington, D.C., and—and, in all honesty, we probably wouldn’t be writing this if it wasn’t the case—Denver.
Overall, students within the city of Denver compare rather positively to their peers across the state and have posted greater learning gains in both reading and math assessments in the academic years 2014/15, 2015/16, 2016/17. When growth performance is broken down by school category, charter and innovation schools outperform traditional public schools when it comes to reading growth results, though in the latest year for which data is available the gap was lesser than in prior years. Within the category of charter schools, CREDO additionally compared independent charters to those affiliated with a network or management organization. Here, the study shows that growth numbers do not differ between the two categories in either reading or math.
Though this is only a minor portion of the study, CREDO’s report also provides distributions of school growth scores (which appear anonymously) in reading and math, subdivided by school sector. There, the graphs demonstrate that the range of charter school scores is much broader, with some schools outperforming the state growth average by close to one standard deviation in reading, while traditional public schools appear much more closely concentrated, though their results too fall mostly on the positive end of the scale.
Among the more noteworthy points of the study are the performance of charter school black and special education students, with both of these groups significantly outpacing state average growth numbers in reading. For a discussion of the study’s other results be sure to take a look at Chalkbeat’s article on this topic.
Finally, it is worth noting that Denver students outperform their state’s average at a rate higher than do students in any of the other cities examined by CREDO. Of course, absent any explanation of why this might be the case, this is little more than an interesting tidbit of information given the nature of the data that is being examined.