Teacher training programs need a reboot

Teacher training programs should be designed on the medical model, writes Jane Dimyan-Ehrenfeld in the Washington Post.

I went to a highly ranked liberal arts college and graduated with a special major in sociology, anthropology and education as well as an elementary teaching certificate. I immediately found a job teaching breathtakingly underprivileged students in a persistently failing elementary school in Prince George’s County. I wasn’t prepared to teach my students how to tie their shoes, much less to make up for years of institutional neglect, hunger, poverty, family transience, isolation and other ills. My first year was a nightmarish blur; my second was only slightly less awful. My third had its highlights but was still a daily struggle.

She enrolled in a one-year master’s program at Teachers College, Columbia to learn how to teach special-needs students. She learned a lot about Lev Vygotsky’s theory of the zone of proximal development, but nothing of practical use.

We can’t decide whether teaching is a “craft or a profession,” Arthur Levine said in the Post‘s story on National Council on Teacher Quality‘s report criticizing teacher education. “Do you need a lot of education as you would in a profession, or do you need a little bit and then learn on the job, like a craft?” asked Levine.

It’s a false dichotomy, writes Dimyan-Ehrenfeld. Medical students combine highly specialized education with clinical rotations, “learning the craft of patient care through observation and guided practice.” They also take “rigorous licensing exams that test their theoretical and practical knowledge.” Then comes on-the-job learning under master practitioners and more tests.

Why not adopt this model for education? Educators could be required to complete a period of schooling in which they learn the theories and ideas that will be most valuable to them as teachers and hone their skills at thinking and talking about education from an intellectual standpoint. Then, perhaps, one to two years of guided practice under the supervision of master teachers could be required, with lots of coaching and meaningful feedback. We could even throw in some rigorous exams.

If it took years of education, training and testing to become a full-fledged teacher, would we have enough teachers?

Dimyan-Ehrenfeld taught for eight years in Maryland and Boston public schools. She now practices education and civil rights law in Washington.

Crowdsourcing Sociology 101

Millions of students around the world are enrolled in hundreds of MOOCs (massive open online courses), reports the New York Times. To evaluate students’ progress, Princeton Sociology Professor Mitchell Duneier is crowdsourcing his Introduction to Sociology class, which enrolls 40,000 students.

 “It was really intimidating at the beginning to do these lectures with no live audience, no sense of who was listening and how they were reacting,” Professor Duneier said. “I talk about things like racial differences in I.Q., Abu Ghraib and public bathrooms, and I worried that my lectures might come across as examples of American ethnocentrism.”

Feedback came quickly. When his first lecture went online, students wrote hundreds, then thousands, of comments and questions in online discussion forums — far too many for Professor Duneier to keep up with. But crowd-sourcing technology helped:  every student reading the forum could vote questions and comments up or down, allowing him to spot important topics and tailor his lectures to respond.

Each student must score the work of five classmates to get their own score, which is an average of what classmates have given them. To see whether peer grading matches traditional grading, Professor Duneier and his assistants graded thousands of midterms and finals.

“I had to announce to the students that some had gotten scores that were higher than they should have been,” he said. “And as data, the midterm scores are useless. But it helped us learn more about writing rubrics.”

. . . So far, he has found an impressive correlation of 0.88. The average peer score was 16.94 of 24 possible points, compared with an average teaching-staff score of 15.64. Peer graders give more accurate scores on good exams than bad ones, they found, and the lower the score, the more variance among graders.

About 3 percent of students copied from Wikipedia.

Princeton doesn’t offer a certificate of completion for MOOCs and less than 5 percent of sociology enrollees took the exams. That added up to 2,200 midterm exams and 1,283 final exams, still a heavy scoring burden without crowdsourcing.

Grading exams: The staircase method

Daniel Solove, a law professor, offers A Guide to Grading Exams on Concurring Opinions. It starts with a stack of exam papers. Then comes the toss down the stairs, which provides a spread for the grading curve.

This is an example of a toss of considerable skill — obviously the result of years of practice.

Exam-Grade-2a.jpg

Solove believes the papers that travel the farthest deserve the highest grades because they obviously have more heft. But an outlier that requires the professor to walk too far should be downgraded to a B.

Is he joking? Yes, he is. Or so he writes.

Via Instapundit, who also is a law professor.