According to the reporting of Penn Live, Republicans in the state legislature are threatening to hold hostage next year’s state budget - due this June 30th - due to the Governor’s veto of House Bill 805, the so-called teacher furlough reform bill. As if the budget doesn’t have enough problems already!
While I support giving school boards the option of furloughing professional staff for economic reasons, and while this legislation was marketed to appear reasonable (who doesn’t want to get rid of bad teachers?) this would have been terrible education policy and a disservice to students.
The key problem is that for the foreseeable future the authority to furlough teachers would have been tied to a deeply flawed evaluation system. A substantial portion of Pennsylvania's evaluation formula depends on standardized student test results that are neither statistically valid nor reliable, and which do a better job of measuring community wealth than student learning.
Let me repeat that: Pennsylvania's current teacher evaluation formula is currently tied to standardized student test results that are neither statistically valid nor reliable.
Furthermore, even the more 'valid' component of the evaluation system - principal observations - is still in its infancy. In many cases, evaluators have received only minimal training, and most still have very limited experience. Attaching 'high stakes' to these evaluations would have negated their potential value for improving overall teacher effectiveness. (Isn’t that the goal?)
And, by the way, we would also have inadvertently compounded the challenge of recruiting teachers in economically challenged districts. Why? Because part of the calculation of a teacher’s score is the overall school score. As a result, a teacher’s score is automatically lower if they teach in a struggling – i.e., poor – school. Which, in effect, reduces their job security. I am not making this up.
There have been some proposals for tweaking the evaluation formula. But as long as the legislature is dug in on using these test scores – which were designed for completely different purposes - changing the mixture of the "garbage" that goes into the formula won't improve the validity of the result: garbage-in is still garbage-out. Claims that the public wants to use standardized tests scores to evaluate teachers is a belief not based in reality. Increasingly, the public wants no such thing, and the more they learn about it, the less they like it.
Perhaps our legislators can't be expected to understand this. Few of them have had any direct experience with public education since they were last in school, in most cases, decades ago. And like most Americans, few have a good grasp of the statistical concepts of reliability and validity. Even fewer would understand the real-life dilemmas a principal faces when considering whether to rate a teacher 'unsatisfactory' - such as, will the replacement be an improvement?
This was another example of our legislature proposing a simplistic answer to a complex problem. The proposed legislation would have done nothing to improve teacher effectiveness. But ironically, there is actually a lot of potential in the relatively new Danielson observation model, particularly if it’s implemented in a climate of trust and greater peer-to-peer accountability. Just maybe we should give that a chance to work.