Defending the No Child Left Behind Act, on the Hoover Institution’s online Uncommon Knowledge show with Peter Robinson, President George W. Bush argued that parents need to be able to see measurements of their school districts’ achievements in order to hold them accountable. The point is well taken, but there are reasons conservatives at the time were suspicious of the enthusiastic support of the late “liberal lion,” Senator Ted Kennedy (D, MA).
Even apart from the urge to teach to the test, measurements run the risk of being obscured in order to argue for increased funding. If a school does poorly, administrators and union organizers blame the lack of resources (and the local population); if a school does well, the same people declare success and argue for rewards.
The latter was recently the case in Tiverton (where, full disclosure, I’m running for school committee). Justifying a three-year contract extension that included various forms of raises, despite the uncertain economy and annual budget fights, Superintendent William Rearick picked from among the RI Department of Education’s (RIDE’s) school report card results for evidence that the town’s schools are “top performers.”
Of particular note is the ranking of one elementary school, Fort Barton, at the very crest of RIDE’s list, among the 17 “commended” elementary schools. Tiverton has two other schools for children of the same age group, one of which, Pocasset, landed at the next level, “leading,” and the other of which, Walter Ranger, was graded “typical.” Familiarity with some of the demographic differences across this economically diverse town led me to wonder how the scores are calculated.
According to the department’s fact sheet for its accountability system, elementary and middle schools achieve their scores based on the following categories:
- 30 points for proficiency: Averaging reading and math scores, what percentage of students are “proficient” on the New England Common Assessment Program (NECAP) tests?
- 5 points for distinction: Averaging reading and math, what percentage of students are “proficient with distinction” (i.e., highly proficient) on the NECAPs?
- 30 points for gap-closing: Combining ethnic and economic categories into one group and special education and English-language learners into another, how much of a gap is there between such students’ scores and those of students not in either?
- 10 points for progress: Does the school appear to be on track for targets set by RIDE for 2017?
- 25 points for growth: Are students progressing from one year to the next compared with their academic peers? This measure also breaks students into three groups: “all students” and the groups described in “gap-closing.”
A conspicuous shortcoming of this methodology is that it effectively gives extra credit to schools that don’t have subgroups to compare for gap-closing purposes. If a school doesn’t have a certain number of minority, English-language learner, special education, or economically disadvantaged students, those categories simply don’t count in the average, and the school gets 30 points toward its total. That’s potentially the difference between “commended” and “typical.”
A review of the report cards for the 17 commended schools shows that nearly one in four has no subgroups at all — four of them, or 23.5%. Another 10 (58.8%) only have enough students in the “economically disadvantaged” category to count, and just three (17.6%) have data in the disability category. (See here for all schools’ scores.)
A closer look at “economic disadvantage” is justified. That category is defined by participation in the National School Lunch Program, by which the federal government subsidizes meals for school children. All students can participate, but those from households above 185% of the federal poverty level (FPL) pay a “regular price.” In Rhode Island, 60% of participants receive reduced prices or free lunches.
For RIDE’s assessment purposes, however, the wide range in circumstances covered by this program should be acknowledged. For a family of two parents and one child, 185% of FPL is income of $35,316.50. The same family with another child could make up to $42,642.50, and a two-parent, three-child household could earn up to $49,968.50 and still count as “economically disadvantaged.”
It doesn’t detract from the difficulty that such families can face to suggest that there’s quite a difference between growing up at the top of that scale and growing up in the grinding poverty more common in urban areas.
For context, consider the 17 schools at the bottom of RIDE’s ranking: all but one have scores for either black or Hispanic students; 11 count both. Fourteen schools (82.4%) have enough students with disabilities to count. Seven (41.2%) count English-language learners. And every single one of them (100.0%) has economically disadvantaged students.
None of these findings is all that surprising, and none of them detracts from the possibility that Rhode Island schools are making tangible progress, both institutionally and with individual students. But for parents and other Rhode Islanders invested in our education system (which means all of us), it is critical to understand what is being assessed and what rankings actually mean.
As opponents suggested back when No Child Left Behind entered the scene, inside interests have much greater incentive and access to understand and manipulate any measurement that the government might contrive.