There are 1000s of more important things for us to be talking about but Ravitch's outsized role in so many conversations is one of those things I hope makes it way up to the top of the to-do list.

Lots of people are going to think this line is true because she wrote it. https://twitter.com/AdriaRHoffman/status/1356594042092597251
Diane Ravitch, the historian, is one of my favorite authors. She wrote the OG education "war" book about the rise of parochial schools in NYC and it's fantastic. Ravitch, the activist, is complicated and messy. Which isn't a slight! The topics she writes about are messy!
And I know lots of teachers find inspiration in her work and advocacy! Which is great! But also! All our faves are problematic and she's no exception. Which is to say, her writing on assessment is... less than reliable.
This paragraph, in particular, is frustrating.
She's using "standardized tests" in this sentence to mean "norm-referenced standardized tests." The challenge is that driver's tests are also "standardized." Standardized means all test takers are tested under the same conditions, their work scored with the same key/rubric.
(If driver's tests weren't standardized, the person administering the test could pass or fail someone based entirely on their opinion of the driver or their skills. But since it's standardized, they use the same checklist for all drivers. Standardized isn't inherently bad.)
The reason it matters, IMO, is that teaching is a profession; Assessments are professional tools. The way we talk about the tools of the profession matters. Psychometrics isn't just about the SAT. It's about teachers' daily efforts to collect evidence of student learning.
The challenge is that in one paragraph, she dismisses "standardized tests." In another, she advocates their use. A teacher-design test is a "standardized" test. It doesn't become more or less standardized if it's for 25 or 25,000 students.
Lots of people use the phrase "standardized test" as shorthand for "federally-mandated, large-scale, norm-referenced tests that consist primarily of recall and short product tasks." Which is fine for laypeople.

She is not a layperson.
And also! I'm 99% sure NCLB mandated criterion-referenced tests; that is, the tests states use for accountability purposes aren't about norms. The data may fall into a bell but scores aren't calculated based on a curve. AFAIK, cutpoints aren't based on aggregate performance.
You can follow @JennBinis.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.