While the recent landmark Vergara v. California court case focused on teachers who are or should be dismissed, the greatest portion of teachers are still those who don't stick around to collect their gold watches (er, engraved pencil boxes), but leave voluntarily.

Annual teacher turnover leaves principals scrambling to fill classrooms and get new staff up to speed. This turnover costs school districts money (roughly an estimated $9,000 per teacher according to a study by Gary Barnes, Edward Crowe and Benjamin Schaefer) and hurts students' academic achievement. While no principal knows for sure whether the teacher being interviewed will stick it out or not, new research by Dan Goldhaber and James Cowan suggests that tracing teachers back to where they were prepared may contain a few hints. Using Washington State data, they found that the variance in teacher attrition by preparation program was about 5-7 percent each year.

Put in clearer terms, at the 10-year mark, 73 percent of teachers who graduated from the prep program with the highest "survival rate" were still teaching, while only 34 percent of teachers from the program with the lowest survival rate remained. That's remarkable.

It's unclear if this variance is due to what the prep programs are teaching future teachers, or because of differences in who attends these programs. But it does suggest districts and principals ought to be collecting and using such data for decision making.