Anticipatory effects of curriculum tracking

Curriculum tracking, the separation of secondary school students into academic and vocational tracks, correlates positively with pretracking achievement in both British and international data. I argue that this correlation is caused by the incentives emanating from the track placement decision. Using test score data collected in TIMSS 1995 and 2003, and in PIRLS 2001 and 2006, I investigate the effect of tracking on the early achievement distribution empirically, amongst others by means of quantile regression. The evidence presented in this paper implicates that previous valueadded estimates of the net impact of tracking may be biased.