Tenure may be the new panacea for solving teacher quality problems. Researchers like Tom Kane, Robert Gordon and Eric Hanushek are all making a steady and convincing case that a system predicated on predicting a future teacher's effectiveness is akin to chasing fool's gold. They argue that the real hope for raising the quality of the teaching force is to put up few barriers at the point when teachers enter the profession but then to take tenure decisions (usually taking place after two to three years) much more seriously.

Economists Dan Goldhaber and Michael Hansen (who happens to be a finalist in NCTQ's upcoming collective bargaining research competition) contributed an exceptionally useful paper on the topic at CALDER, which is becoming the can't-miss annual education research conference.

Looking at value added data on some 1,300 novice fifth grade teachers in North Carolina, Goldhaber and Hansen perform a simple but smart exercise to see if those teachers' value added before they earned tenure predicted their value added after they earned tenure.

The answer? It depends on the subject being taught and the teacher's initial strength or weakness. For purposes of denying tenure, one third of the teachers who performed in the bottom quintile in their pre-tenure years of teaching also performed in the bottom quintile in their post-tenure years. When you expand that definition of weak to accommodate teachers in the lower 40 percent of performance, a sizeable number are still at this level a few years later.

While these results may be fodder for both sides of the debate over using test scores to judge teacher performance, we argue that the results are pretty good for having used a single measure to judge performance. Coupled with the additional knowledge gained by multiple evaluations over the same period, there is every reason for districts to make harder and more meaningful decisions about which teachers should earn tenure.