assessment, accountability, and other important stuff

Archive for September, 2019

A good day ruined

Charlie DePascale

After a wonderful late summer day spent enjoying a rare weekday afternoon baseball game in Boston, I sat down last night and looked at my Twitter feed.  There among the trending items was this headline

The University of Texas’s Secret Strategy to Keep Out Black Students

Without even clicking to look at the article I knew without a doubt that this ‘secret strategy’ had to involve standardized testing.  Is this how gun manufacturers, drug companies, and those e-cigarette people feel?

Standardized tests don’t kill people …

Where will this end?

Will NCME soon be thought of in the same way as the NRA – but you know, without the money, membership, or political clout? The president of AERA fired a direct shot at testing last spring with her presidential address, “An Inconvenient Truth About the New Jim Crow of Education” – a catchy title.

Will San Francisco lawmakers be next to set their sights on NCME – labeling it a domestic terrorist organization?  Will there be a resolution to block NCME from holding its 2020 conference in San Francisco?  If so, will AERA support it?  After all, it wouldn’t be the first time AERA moved a conference in California in the name of supporting a social cause.

What about New York, already regarded as the Lexington and Concord of the opt-out revolution?  Will the governor and legislature take aim at NCME and deny state funds to people doing business with the testing industry?  What will the pineapple say?

Where do we go from here?

Is there a way to stem this anti-testing tide and restore the shine to this field of ours? (dare I say, to make testing great again. no, I think I’ll pass on that.)

I can support increased background checks for all test users.  I am more skeptical about federal- or state-mandated bans or limits on state and local testing.  Perhaps those efforts can reduce the damage caused by high-stakes census testing, but an ill-conceived and poorly-developed teacher-made test in the hands of an inexperienced teacher can still cause a lot of harm one child at a time.

Standards, lists of best and fair testing practices, and policy statements are necessary, but not sufficient.  I am pretty sure that I already read somewhere that it’s not acceptable practice to base a high-stakes decision on a single test score.

Improving the assessment literacy of all involved in the testing process including students, teachers, policy makers, the media, the general public, and psychometricians is a good place to start.  I know some folks in New Hampshire who are doing some nice work in that area. (sorry, no names or links.  need to keep that personal/professional firewall intact.)

Improved assessment literacy, of course, won’t stop people who want to use testing for unsavory or evil purposes from doing their thing.  It might, however, make others more aware of when testing is being used to do harm; and make them more likely to speak out; and make them less easily persuaded to opt out.

The trend toward moving the locus of assessment from the state house to the classroom also seems like a step in the right direction. Not only will that put actionable information in the hands of teachers, where it belongs; that shift will help eliminate problems like the one faced by Massachusetts last spring. As we have known for a long time, passage-based triggers are much easier to avoid in the classroom than on a state assessment.

Improving local assessment is a step, however, that will require tremendous investment in the infrastructure of teacher preparation programs and schools.  Remember that one of the big advantages of large-scale assessment is that it’s cheap and doesn’t require much training of teachers and school administrators.

Advances in instructional and assessment technology, personalized learning systems, and modeling based on a much broader base of data than test scores and student attendance also has a great deal of promise, but will not be without its own technical and social challenges.

I look back fondly on the days when we were accused of trying to peek into family life with our student survey questions on how much time was spent watching television each night or whether the student had a part-time job; of trying to brainwash students with our questions about the environment; or of simply trying to get students to fail with our trick questions that contained plausible distractors.  So, yes, tracking and developing psychometric models for a student’s eye movement, heart rate, and other signals of student engagement might face some resistance.

For now, however, I will go back to Twitter and read about climate change.  That ‘inconvenient truth’ reference aside, I am pretty sure testing isn’t being blamed for climate change – well, not yet – or maybe I’ll just listen to some music.