A Web Paper Stating That Student Achievement Should Not Be Used to Measure Teacher Effectiveness
Is student achievement the measure of teacher effectiveness?
The short, and the long answer to this is no.
Of course I framed the question using the phrase “the measure of teacher effectiveness.” Why do I bring this topic up for discussion on this blog. In America there is a lot of talk about educational reform, especially from President Obama and the new Secretary of EducationArne Duncan (read Duncan’s commentsgiven at the NSTAConference). There is not only interest, but there is a lot of Federal money being directed toward the nation’s schools, and a good deal of the funding will be to focus on educational reform.
The U.S. Department of Education’s vision of educational reform is embedded in the No Child Left Behind Act, which has created an environment in which student test scores are used to measure teacher and school effectiveness. It seems perfectly reasonable to link how well students do on tests to the effectiveness of teachers and schools. But is really?
Many pundits (and many educators) like to infer that teaching and learning are are like any business, which that business is manufacturing and selling automobiles, selling antiques, making refrigerators, manufacturing matresses, selling paper clips, and I should add, selling securities such as stocks and mortgages. The basic idea is of course to reward the high performers (those with high sales records, or improvement from the last quarter) in these businesses with high pay, or some other measure of reward. In fact in today’sAtlanta Journal, Mr. Dean Alford, a former Georgia House member, vice chair of the State Board of the Technical College System of Georgia, and a business owner, believes that “paying teachers who show improved student achievement” will attract more dynamic individuals into teaching. It really galls me when I read a statement such as this.
These individuals really believe that there is a cause and effect relationship between teacher behavior (or teaching skill, if you wish), and how many questions students answer correctly on an end-of-the-year high stakes test. Further they believe that you can use test score change (improvement, hopefully) as a way to calculate teacher pay. It is very difficult to carry on a conversation with individuals like Alford because they honestly believe the work of a teacher in a classroom is very much like the work of the person trying to sell you a car, or a paperclip. These individuals are trapped in a 19th Century model of work, and learning. Most companies have multifaceted performance evaluations, and tend not to rely on one variable to make a decision on the worth of an individual to a company. This has to be true, otherwise how do you justify the enormous bonusesbeing given to AIG executives in departments that caused the company to tank.
Given this angst, I do want to say that there is research evidence to support the idea that effective teaching results in increased science achievement for all students. Johnson, Kahle, and Fargo in a studypublished inScience Education, report that effective teachers positively impact student learning. In fact, you will find a lot of supporting research in the journal Science Education related to teacher effectiveness. Follow this link, and you will find 152 studies that deal in some way with teacher effectiveness in science.
However, science achievement is not the measure of teacher effectiveness. There are wide variations in the way in which effective and outstanding teachers perform. To use a test score to validate teacher effectiveness is not only short-sighted, but not supportable in terms of research design. In order to tie student achievement to teacher performance we would need to be able to support thehypothesis experimentally that gains in students achievement are the result of teacher performance. This is an unreasonable request in that it would require assigning random groups of students to teachers, and then pre- and post-testing with experimental and control groups. Indeed, most studies attempting to tie student achievement and teacher performance use “quasi-experiental designs,” and in these studies there a many variables that could contribute to the “acheivement gains.”
The environment that will foster student learning is a humanistic learning environment. Its environment that encouages teachers to be creative and inventive in working with students, and designs a workspace in which risks can be taken. Students need to be challenged, but more importantly need to involved in making decisions about their own education. We know that learning science is NOT akin to passing out bricks, and then measuring how many bricks were attained be students in our classes. Humans construct knowlege, and learning environments that are contextual, relevant, and of value provide the setting for meaningful learning.
We need to support the notion that teaching is an art, and can not be reduced to a student achievement test score.