Although evidence generally suggests that parental involvement is beneficial for the academic success of students, much less is known about the determinants of this involvement. The study examined the associations between parental characteristics, beliefs and two types of parental involvement (school-based and home-based) in secondary education in three countries in Europe. Data from 1757 parents from nineteen public secondary schools in three European countries (six in Luxembourg, six in France and seven in Belgium) were analysed using multiple regression analyses. Models predicted overall significant but relatively low portions of variance in the home- and school-based involvement of parents. Different patterns of results emerged from the three very different samples. When controlling for family background and school environment, parental self-efficacy--associated for example with the role of parents at school--appeared to be central indicators of parental involvement. Results are discussed in terms of research on parental involvement and family-school communication, as well as in relation to school practices involving parents.