CONTEXTUAL VALUE ADDED MEASUREMENT
Not quite as robust as some claim
The Sykes Review of qualifications added to the growing criticism of contextual value added (CVA) measurement, increasingly used in league tables and by the SSAT quango to measure the performance of the state schools its supports (over 90%).The Sykes review noted that while the use of CVA is an attempt to “respond, within the league table culture, to a genuine and important issue; namely the greater challenge that some schools face, compared to others, in improving the academic achievement of their pupils….the implied precision of these measures is, in fact, spurious, and the measures are therefore unfair to schools and teachers, and to pupils and parents making use of them.” So, the Review recommended that “Wide performance measures (value added, CVA, average point score) should not be calculated unless it can be demonstrated that there is real underlying validity to the methodology, including the methodology for valuing different qualifications and different subjects within the same groups of qualifications.” Significantly, there is a range of different types of value added measurement. A recent US report by the National Research Council and National Academy of Education, found, to nobody’s apparent surprise, there is not one dominant Value Added Measurement. It noted too that “there are many technical and practical issues that need to be resolved in order for researchers to feel confident in supporting certain policy uses of value-added results”. Indeed, it went further concluding that no one value-added approach (or any test-based indicator, for that matter) addresses all the challenges to identifying effective or ineffective schools or teachers. Each major class of model has shortcomings, and there is no consensus on the best approaches, and little work has been done on synthesizing the best aspects of each approach. There are questions too about the accuracy and stability of value added estimates of schools, teachers, or programme effects. The overall conclusion was that much more needs to be learned about how these properties differ, using different value-added techniques and under different conditions.