For months we heard from union lackeys, their politician bootlickers, fellow-travelling ideologues, and substandard teachers about how the new COMPASS evaluation system for public school teachers in Louisiana was so deficient in erring, they asserted, about true teacher quality. They were right – for the wrong reason.
The party line from this crew was that the new system, which would base half of scoring on measured student learning progress and the other half on subjective evaluations similar to the past, would make good teachers appear substandard. Of course, we never heard complaints from these fiddlers that as the state continued to rank among the lowest in student achievement, and lower than many with similar demographic characteristics, about the fact that under the old completely subjective system that only one percent of all teachers were rated poorly enough to be subject to dismissal for incompetence and that only a little over two percent get fired for that reason.
Now the results are out and, contrary to all the hyperbole, the system computed that only four percent of teachers were judged “ineffective” and thus could be set up for dismissal in the next two years, while only eight percent fell into the category suggesting remediation before things got worse. That means more than seven-eighths of all teachers were found at least adequate, despite the state’s continued below-average showing – hardly the stuff of hyperventilating claims of a “war on public education.”
And while the state’s Department of Education noted there was rough congruence between student achievement and district teacher results – better-achieving districts tended to have a higher proportion of adequate-and-above teachers – some disturbing trends emerged. This relationship held most strongly where state standards were the basis of measuring student learning, but was much weaker where they were not; for the core subjects, such standards exist, but for those areas of instruction such as in the arts and physical education, local districts come up with their own, somewhat more subjective, criteria. Also, in some districts there was a much higher skew on the subjective observational portion towards judgments of adequacy that brought the distribution of those scores into stark contrast with the distributions for the objective testing portion that skewed much more to inadequacy in teaching – a pattern most evident among the lower-performing districts.
In other words, many districts – with some notable exceptions concentrated among the higher-performers – were gaming the system to pull some teachers out of inadequate ratings. For example, in Ouachita Parish, while 97 percent of teachers were ranked in the upper two categories according to the subjective half of scoring, only 79 percent were from the objective half based on student outcomes, and in the City on Monroe schools, 94 percent of teachers were in the upper categories by observation, but just 82 percent were on the basis of student progress.
However, those results include all classrooms. Limited only to the 21 to 43 percent in each district across the state only who teach in the core areas – those where districts do not impose their own standards – in Ouachita only 43 percent of teachers had student performance in the upper categories while for Monroe just 34 percent did that well. Ineffective ratings were assigned among these teachers on this standard to seven percent of them in Ouachita and 24 percent in Monroe.
There were worse districts of skew in this direction, and some among higher performing districts actually in the opposite direction, but as a whole the lesson learned is, through the observational component and when they set their own achievement standards, the lower-performing the district’s schools, the weaker the relationship between that and teacher evaluation because they are more likely to engage in behavior that pumps up teacher overall scores. That’s not unexpected, since almost all local administrators themselves once held teaching certificates and want to put performances, on which their success may be judged, in the best possible light.
So the naysayers were correct – except instead of the system being needlessly harsh on judging teachers, it is too lenient and thereby reduces its value as a diagnostic tool. To fix that, Louisiana needs to do what nearly half of all other states presently do: evaluate teachers in part on their actual subject knowledge competency, in addition to the other things.
As a college professor, I encounter some students who cannot spell well or use grammar consistently correctly and/or cannot figure out simple math such as computing a course’s overall grade in progress. No student graduated from high school and admitted into college should display this level of ineptitude, and the only way they would be able to get this far is that teachers passed them along, either out of indifference or because of their own ineptitude in teaching these skills. A subject area exam taken every couple of years or so by teachers, developed by the state aligned to learning expectancy standards of students at that grade level to ensure rigor, would show decisively where the breakdown in knowledge transmission is occurring that shortchanges some students.
Adding this to the COMPASS instrument not only provides a very significant indicator of actual teacher quality, but it also would dilute the subversive motives that subjective evaluators may bring to the task. This addition ought to serve as the next iteration in education reform in Louisiana.