To measure pupils’ self-confidence, the authors conducted a survey of 2,034 sixth-form pupils during the three weeks leading up to the deadline for submitting university applications via Parcoursup in 2021. The authors used a measure of self-confidence that is commonly employed and validated in the experimental economics literature (Niederle and Vesterlund, 2007; Buser, Niederle and Oosterbeek, 2014; Dargnies, Hakimov and Kübler, 2019; Mobius et al., 2011).
They then linked their survey data with administrative data from Parcoursup, which includes, in particular, the choices submitted by students, the offers received, and the final choice made. As the relative rank perceived by pupils is measured in the three weeks leading up to the deadline for submitting applications, these data enable an analysis of the link between self-confidence and the choices made.
The authors examine several characteristics of the choices made, in particular whether a student applied for a preparatory class (CPGE), a useful indicator for detecting self-censorship amongst the top-performing students. They also examine the selectivity of the courses ranked by applicants, as measured by the average marks obtained in the baccalaureate by students admitted to these courses.
Finally, the authors devised a simple intervention consisting of informing pupils of their actual position in the national distribution of final-year average marks. For pupils who initially underestimate their rank, this information adjusts their perception of rank upwards. Conversely, for pupils who initially overestimate their rank, it revises their perception downwards. The pupils who received this information were selected at random in order to measure the causal effect on their course choices.
