With the prospect of teacher bonuses to be paid on the basis of gains in student achievement the measurement of “value added” by teachers has become a critical issue. Just how easy is it to measure the value added of each individual teacher?
This issue is being much debated in academic circles in the United States, especially in the light of President Obama’s Race to the Top program which requires the states to introduce performance pay based on student achievement. The answer so far is that while it sounds great in theory, there are insurmountable barriers to effective and reliable measurement of teacher performance.
The problem is that there are a myriad of factors which influence student performance in the classroom over a year or several years. Therefore, it is extremely difficult to attribute the gain in test scores from one year to the next, or over several years, to individual teachers.
An article in the latest issue of the Journal of Research on Educational Effectiveness (July/September) points to one of the problems. The study by researchers from Mathematica Policy Research shows that the quality of curriculum programs chosen by schools has an effect on student achievement. This suggests that the quality of each curriculum program a teacher is using needs to be considered when measuring their “value added” for bonus payments.
In the study, four different elementary school math programs were randomly assigned to 1st grade classrooms in 39 schools. The schools all served students from low-income families and were located in four different school districts. The same standardized test in mathematics was administered to all students.
The study found that students who were taught using two of the programs achieved test scores that were 9 to 12 percentage points higher than student achievement scores using the other two programs. This suggests that teachers of equal ability would have achieved different results depending on which maths program they used in their classroom. Teachers using the first two programs would have had a much higher value added score than teachers using either of the second set of programs.
More generally, such results make it difficult to hold teachers accountable for student achievement without regard to the programs they are required to teach. Some programs can help raise student achievement in the hands of an average teacher whereas other programs will not be as successful even for an excellent teacher.
Another possibility not considered in the Mathematica research study is that some curriculum programs may work better for some groups of students than for others. This would further confound the measurement of value added by teachers using gains in student test scores.
The introduction of the national curriculum will not obviate the need to distinguish between curriculum and teacher effects on gains in student test scores. The national curriculum will not require teachers to follow specific programs. This will remain a matter of choice for education departments, school districts and schools.
Thus, the quality of each curriculum program used by teachers will need to be considered when student achievement scores are used to award bonuses to teachers. However, it will be virtually impossible to determine whether a particular curriculum program has a large or small effect on increases in students’ test scores.
This will mean that some teachers could be awarded bonuses because they are using a better curriculum program than other equally good teachers.