The Education of the Educators
David Glenn at the Chronicle reports on the use of longitudinal databases to evaluate teacher preparation. Loiuisiana’s is considered one of the better ones.
Using a new longitudinal database, it had analyzed the standardized-test scores of fourth- through ninth-grade students and matched those scores to the institutions that had trained their teachers.
This isn’t an unexpected development. Once we start looking at student academic achievement it follows that we’ll look at teaching and then at teacher preparation.
To many observers, these databases are a long-overdue step toward a real understanding of the quality of higher education. “This kind of information can be extremely powerful,” says Grover J. Whitehurst, a former official of the U.S. Education Department who directs the Brookings Institution’s Brown Center on Education Policy. “For the first time, you’re getting data on whether training is actually affecting teachers in the way that you would expect it would.”
Researchers in Louisiana note that improvement in teacher preparation isn’t possible if schools of education think they’re doing fine.
“Everybody said the same thing to us,” Mr. [George H.] Noell [professor of psychology at Louisiana State University at Baton Rouge] says. “‘We’re fine. Why are you bothering us? We have a great product. We feel like we produce great teachers, and when we survey principals, they say that our teachers are strong.’
“And that’s true. The survey data is uniformly positive. And yet when you look at the level of achievement of schoolchildren in Louisiana, things are not good. So there have to be issues somewhere. We came to the realization that without some sort of objective metric, we probably couldn’t make any progress, because people were convinced that they were fine.”
The Louisiana system is based on value added measures and not on raw test scores.
Teachers are assessed according to how well they build their students’ skills from one year to the next, so they are not penalized for having students who arrived in their classrooms with weak skills. That means, in turn, that teacher-training programs are not punished as a result of a large proportion of their graduates’ choosing to work in high-poverty districts.
In addition, the analyses cover a three-year range, so they do not suffer from the statistical volatility that often afflicts single-year measures of teachers’ effects. And the Louisiana reports include only students who stay in the same classroom throughout the academic year, so teachers are not unfairly penalized (or rewarded) for midyear transfers.
Still, not everyone is happy about the use of standardized test scores, raw or not:
“There’s much, much more going on that we’d like to know about,” says Sandra L. Robinson, dean of education at the University of Central Florida. “What about the progress of the students toward graduation? What about referrals to counseling? There are so many more impacts that a teacher has in the classroom that aren’t captured by standardized tests.”
Others voice concern that the making of a good teacher reflects more than college of education programs.
Some observers are skeptical even of relatively sophisticated models like Louisiana’s. “Any accountability system worth its salt will not only tell you who’s doing better or not, but why,” says Barnett Berry, president and chief executive of the Center for Teaching Quality, a nonprofit organization in North Carolina.
He doubts that measures based only on high-stakes standardized tests can capture the full complexity of teachers’ work. He also worries that the analyses will fail to take into account school-level variables like the mentoring (or lack thereof) that new teachers receive, or the camaraderie and stability of a school’s faculty.
Despite these concerns, we can expect more and more states to use longitudinal data systems:
Since 2005 the federal government has given states more than $400-million to build long¬itudinal education databases. Before long, nearly every state should have the capacity to do the kinds of analysis that Louisiana has done, tracing schoolchildren’s academic performance back to their teachers’ teachers.
