Home > Uncategorized > What Could Replace Standardized Tests as A Metric?

What Could Replace Standardized Tests as A Metric?

January 27, 2015

Several years ago (over 35 to be more precise), I recall pondering to move as I sought a new job in a larger school district. To help us decide where we might want to move, my wife and I had purchased  Places Rated Almanac which complied reams of data sets to help families like ours decide where we might want to move. As I recall, the data sets included items like housing costs, availability of transportation, recreation alternatives, medical services, weather, taxes… and schools. The metrics used to assign “stars” for each of the data sets had some flaws (e.g. recreation included the number of golf courses and bowling alleys, neither of which interested me) but the one I found most frustrating was the data set for schools. The ratings used some combination of per pupil spending, class size, and standardized achievement tests to evaluate “quality” and they were particularly inadequate in metropolitan areas like Hartford CT where the city school indices were blended with the nearby suburban districts… and I knew from my experiences in the Philadelphia area that there were VAST differences between the city schools and suburban schools and among the suburban school districts as well.

It is therefore not surprising that after relocating to Western MD in the late 1980s I was enthusiastic about the State Department of Education’s notion of developing “Report Cards” for each school as part of an accountability initiative and I welcomed the opportunity to participate in the development process as one of the 24 Superintendents in the State. Under the leadership of State Superintendent Nancy Grasmick we developed a Report Card that included standardized test scores, drop out rates, attendance, and a host of demographic data that included race, socio-economics, special education, and ESL. As the report cards evolved we worked to upgrade the Report Cards. We changed the test data reported, moving away from the use of minimum competency tests to the “Maryland School Performance Assessment Plan” (MSPAP) assessments that were administered in grades 5, 8 and 11. We made certain we all used a common methodology for defining “drop outs”.  We reported the data in a disaggregated format to make certain that high-performing students in affluent schools were not masking the deficiencies of, say, special education students in less affluent schools. The MSPAP Report card was far superior to the blunt measurements used by Places Rated but still fell short of capturing the elusive qualities the separate a “good” school from an “excellent” school.

I was saddened when I learned that Maryland had to abandon its tests with the advent of NCLB because they were not given at the end of each grade and they also needed to abandon the format of the Report Card. Standardized Test data became the primary metric for determining school quality. Race To The Top, as noted repeatedly in this blog, only made matters worse by raising the stakes of standardized test results thereby making them the de facto exclusive metric for school quality. Ironically, the depth of the data available to parents in the age of the internet was more school specific than the Places Rated data, but it was far less helpful to parents and far more punitive to schools.

But, as Anna Kamenetz reported in an NPR post earlier this month, the reauthorization of NCLB may result in the use of a different set of accountability metrics. In “What Schools Could Use Instead of Standardized Tests” Kamenetz offers a long list of possibilities:

  • Samplingwhere tests like NAEP would examine random samples of students in a school instead of stopping everything to administer tests to all students simultaneously.
  • “Stealth testing”, which suggests formative assessments like NWEA or Khan Academy dashboards could be used to systematically determine individual student progress
  • Multiple Measures, where routinely collected data, like “…graduation rates, discipline outcomes, demographic information, teacher-created assessments and, eventually, workforce outcomes” could be used to measure a school district’s effectiveness. Kamenetz also suggests that emerging metrics like social/emotional skills surveys, game-based assessments, and portfolios could be included in the “multiple measures” provided to students and/or their parents.
  • Inspections, conducted by well-funded state departments of education.

Kamenetz has written a book on this topic, which provides a more expansive description of each of these ideas and offers others…. and anyone who thinks this is “unaffordable” should look at how much we are spending on standardized tests. The billions spent on those tests could easily be redirected to the metrics described above… and the results described above would be far more beneficial to teachers, parents, and students than the results we are getting today.


%d bloggers like this: