Ballatore, Rosario Maria Paccagnella, Marco Tonello, Marco
Year of Publication:
GLO Discussion Paper 116
Using census data on three cohorts of 5th grade Italian students we investigate how the ordinal rank in the within-school age distribution affects the probability of being bullied. Identification is achieved by exploiting within-school between-cohort variation in the age composition of different school cohorts, and through an IV strategy based on the discontinuity in the probability of enrolling in a given school year generated by an end-of-year cut-off rule. We find that being in the upper part of the school age distribution reduces the probability of being bullied: a one-decile increase in the within-school rank decreases the probability of being victimized by about one percentage point. The effects are stronger for females, children from disadvantaged backgrounds, and children spending the entire day at school; they do not depend on the choice of the reference group, as defined according to socio-demographic characteristics.
bullying ordinal rank relative age school violence