Arne Duncan, the Secretary of Education, has just spoken to release his plan for teacher education reform. Once again the plan cites the McKinsey report : “only 23% of all teachers, and only 14% of teachers in high-poverty schools, come from the top third of college graduates.” OK, let’s go to the report. It in turn, cites these numbers and the source is “Derived from the US Department of Education, NCES, 2001 Baccalaureate and Beyond Longitudinal Survey.” Well, this does not shed much light on the methodology, so I asked them. But in the meanwhile, let’s look at the raw data to see how they found it out.
How do you actually identify the top third of college graduates? Note, they don’t cite the ETS report on SAT scores, also erroneous. We certainly do not rank students in our graduating class, unlike some med schools do. The survey has only a few categories related to performance:
- Overall Undergraduate GPA
- GPA in Undergraduate major
- Graduated with academic honors
- Received incomplete grade
- Repeated class to earn higher grade
- Withdrew from course due to failure
What do we see? Graduates who work in Education (which is mainly school teachers) have the third highest average Cumulative GPA, the highest GPA in the major, the second highest number of people graduated with honors. They are in comfortable thirds and second quartiles on the three other quality measures. This is consistent with our data for just one institution. I don’t know if my data table will re-run for you, but you can easily rebuild it, or see my exported table.
Anyway, as Russian say “Either I am stupid, or these skis don’t slide” (Don’t ask). But I really would like to know how is it that the secretary of education finds is acceptable to cite a non-refereed source in a major policy speech, and how that source can publish data without at least explaining its methodology.