Age of tracking and the smoothness of upper secondary school transitions
|Title||Age of tracking and the smoothness of upper secondary school transitions|
|Publication Type||Journal Article|
|Year of Publication||2012|
|Authors||Latina, J, Ramirez, J|
|Journal||University of Geneva Working Papers|
|Keywords||education, relative educational quality, tracking, transitions, treatment effect|
Middle school tracking is a common feature of many counrties' school system. In this paper, we examine if a one-year delay in middle school tracking influences the probability of track changes at the upper secondary level, using panel data for Geneva (Switzerland) over the period 1993-2007. We tyke advantage of a natural experiment, whereby a group of middle schools tracked their students a year later than the rest of schools. Because of specific conditions, we argue that selection into treatment is independent of potential outcome and, therefore, use a regression design to identify the average treatment effect. Both linear probability and panel data models estimate a small but significant negative impact of delayed middle-school tracking on the smoothness of upper secondary school transitions. However, we find that this negative effect of a later tracking decreases with the level of performance. We explain this result as a by-product of the "big-fish-little-pond" effect.