Skip to main content

View Diary: Education: Why data-driven instruction does not work (98 comments)

Comment Preferences

  •  OK, I'll agree with that (8+ / 0-)

    And I do use tests and quizzes that way. But these 60 question, 4-hour benchmark tests every 6 weeks are overkill. They are often on material the students have not even had presented to them in the classroom yet, and the kids get so sick of the constant testing they don't give any efffort on the benchmarks, so the data has even less value than I indicated above. Adding another 110 hours of testing into a 180 day school year is a lot and the value of the data is so minimal, it really is a bad idea.

    •  Well, I'll agree to your agreement. (0+ / 0-)

      The benchmark tests sound stupid. There's a world of difference between pretest/posttest -- which is not intended as a learning activity, but nonetheless clearly functions as one, the psych research is not really ambiguous on this -- and tests that are random samplings of what a student knows, taken from a large "population" of target knowledge. Anyone who knows the first thing about stats knows that you might get something meaningful as an aggregate measure for the test group, but the individual student's results will tell you next to nothing.

      To put the torture behind us is, inevitably, to put it in front of us.

      by UntimelyRippd on Sat Apr 27, 2013 at 07:47:03 PM PDT

      [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site