Testing Performance Review
Last week, the New York City Department of Education released a database containing the historical standardized math and reading test scores for teachers whose students complete these tests. While hailed by many (see New York Times) as an heroic first step in teacher accountability, others see this as more political posturing. A stormy debate is raging over teacher performance and accountability; a debate with some worthwhile insight for corporate performance management!
Measure what Matters
Do test scores matter? Whether the tests referenced are useful predictors of future success or not, they do guide the schools, programs and opportunities available to pupils. For this reason alone, it makes sense to evaluate teachers, at least in part, on the scores their students attain. Should teachers, then, “teach to the test”? There are probably a lot of reasons they shouldn’t, BUT (yes- a big BUT), if teachers are evaluated on the basis of test scores, then they will "teach to the test". Wouldn't you?
Accountable for Controllable
Teachers rarely pick the students they teach. Teachers rarely work with a student group for more than a single academic year. Shouldn't the metric be the delta or 'change in score' year over year rather than the students' raw scores? If Johnny scored in the 49th percentile, and we hold Mrs. Apple accountable for that score, we've got to assume that Mrs. Apple is less competent than 51% of the teachers, BUT (another big BUT) if Johnny scored in the 40th percentile the year before, then Mrs. Apple had a positive impact of +9 percentile points in one year. Looking at it this way, Mrs. Apple may well be a star teacher!
Informed for Action
The rationale for publishing the scores is empowerment of parents and taxpayers. If public funds are being used, shouldn't we know how well they are being used? Well yes, we should. The average test scores of students taught by a teacher is, however, a great deal different than an Assemblyman’s voting record. The Assemblyman’s record is entirely ‘in his control’ and (a big AND) voters who have this information can take decisive and meaningful action with this information (vote for or against in the next election). How will parents and taxpayers use the test score information? Parents and taxpayers aren’t empowered to hire and fire teachers, (and, as we all know, even principals and School Boards are vexed in this regard by the UFT J). Should parents be allowed to decide which teacher will teach their kids? Can you imagine the administrative requirements? Assuming it could work (parents selecting their children’s teacher), it is reasonable to assume that kids with parents who get them into a good classroom are likely to score higher than kids with parents who don’t bother. In the end, test score differences are likely to say little about the students or the teachers, and most about the parents. This is not conjecture, but supported by research: involved and interested parents raise kids who perform better academically.
Good Performance Management
I’m not opposed to making these scores available, just as I’m not opposed to sharing a customer accolade or complaint at the workplace. It’s good practice to share point-in-time results to spur discussion and exploration. It’s just not a viable performance management approach. The public school system exists to build good citizens, and that requires a good math score and a whole lot more. Alice may score in the 99th percentile but be dealing drugs in the playground. We need Alice’s teacher to be focused on her math class and the playground (nice if her parents are engaged as well). A good performance management system looks at results, behaviors and skills applied across all relevant job requirements. The New York City Department of Education has allowed the public insight to only one small component of each teacher’s performance. It’s shocking for me to agree with the UFT, but, in this matter I believe the City has unleashed an unwieldy force that is likely to make managing teacher performance and accountability a whole lot harder in the weeks and months ahead.
No comments:
Post a Comment