We seek to quantify the role of education as a mechanism through which family background affects earnings. To this end, we propose a generalisation of statistical 'mediation analysis'. In our approach, the treatment and mediator can be multidimensional. This allows us to directly and flexibly...

We use rich data on a cohort of English adolescents to analyse the long-term effects of experiencing bullying victimisation in junior high school. The data contain self-reports of five types of bullying and their frequency, for three waves of the data, when the pupils were aged 13 to 16 years....

We investigate the relationship between peers' abilities and educational outcomes at the end of high school using data from the rich Longitudinal Study of Young People in England (LSYPE) matched to the National Pupil Database of children in state schools in England. In particular, we focus on...