Studies Link Classroom Observations to Student Achievement
Two recent studies on Cincinnati's teacher-evaluation system provide some initial insights about classroom practices that seem to be linked to better student performance, and evidence that teachers improve as a result of the formal review process.
Published in the journal Education Next, the first study, by Tom Kane, Eric Taylor, John Tyler, and Amy Wooten, analyzes teacher scores on Cincinnati's evaluation system between the 2000-01 and 2008-09 school years. Under this system, teachers receive four observations and are graded on a 1-to-4 scale on several different standards.
The researchers grouped these standards into three main areas: overall classroom practices (which includes all of the standards), classroom management, and using questioning and discussion instructional techniques.
They found that a 1-point increase (from 3 to 4) on overall classroom practices led to a seventh of a standard-deviation increase in reading achievement and a tenth of a standard deviation in math achievement.
In addition, they found that, while overall teaching practice was the best predictor of student achievement, classroom management was more highly correlated with better math performance than the teachers' use of questioning. For reading, use of open-ended questions was more highly correlated with student performance than classroom management, though this finding was somewhat less robust.
The link between classroom practices and achievement is also a question being studied as part of the Bill & Melinda Gates Foundation's Measure of Effective Teaching project. (MET is also being overseen by Kane, who participated in this Cincinnati study.) The Gates folks have released some information from MET, but we've not yet had the analysis of the classroom observations.
Arguably, information on classroom practices are the most important of all, since it will help teachers know specifically what they can do to be better at their craft.
A second study on Cincinnati's evaluation system came out in March. That one, credited just to Taylor and Tyler, looks at the effect of evaluation on midcareer teachers.
Again using longitudinal student data, the researchers found that student performance improved in math—both during the year the teacher was evaluated and even more in the years following evaluation—compared with students assigned to the same teacher in the years before the teacher was reviewed. The study found that these results couldn't be explained to changes in the type of students taught by those teachers or to experience.
You may be wondering why all this research is coming out of Cincinnati. Well, it's one of the few districts that has used a teacher-evaluation system based on several classroom observations—and with four categories of teacher performance—for more than a decade.
In other words, it's essentially the basic model many states and districts are now trying to put in place to meet the requirements of federal competitions and newly passed state laws on teacher evaluation.
(There are, of course, some differences between the Cincinnati model and some of the new systems going into place, like the District of Columbia's IMPACT. For instance, Cincinnati does not currently use value-added, and most teachers are reviewed at Year 1, at the tenure year, and every fifth year thereafter, rather than annually.)
Cincinnati's system also connects to a career ladder of sorts. Teachers identified as being especially effective can move to "lead teacher" role and serve as peer-evaluators or in a variety of other roles across the district.
A plan to tie the system more broadly to teacher pay across the district, however, was voted down by the teaching corps in 2002.
Another important point: These findings are coming out in a teacher-evaluation system where the majority of teachers receive 3s or 4s on the 4-point scale. That's raised some questions about the rigor of the system. But in the Education Next study, the authors note that, despite the high ratings overall, "there is a fair amount of variation from teacher to teacher that we can use to examine the relationship between TES ratings and classroom effectiveness."