61

This article offers a critical analysis of the performance indicators for French lycées [high schools] published annually on the internet by the Département des Études et de la Prospective. In a study of the 103 lycées in the Bordeaux educational area, school performance was considered in relation to pupils’ starting academic level (n = 18 000) as measured by grades on the middle school written leaving exam. The analysis identified four main points. 1) Schools assumed comparable in terms of type of pupil pool in fact educate pupils of highly diverse academic levels, which distorts “value added” calculations. 2) This phenomenon constitutes a systematic bias that leads to under-evaluating lycées attended by lower-status class pupils and overevaluating lycées more privileged in terms of pupil pool. 3) The systematic evaluation bias in turn results from localized schooling markets which affect who enrols in what lycée and thereby accentuate inequalities among schools. 4) The author proposes an experimental method of calculating a “corrective adjustment” of expectations about individual schools by taking into account initial academic level of first-year lycée pupils.