It is commonly understood that democracies need actively engaged democrats and that adolescence is a significant period in life for educating engaged citizens. Whereas previous quantitative studies in the field have primarily focused on the relationships among participation-related variables, the research reported here aims to categorize secondary school students according to their civic orientations. Thus, the present study proposes a different strategy of analysing quantitative data, namely a person-centred statistical approach, which is well suited when the research focuses on heterogeneous populations. It utilizes attitudes towards the importance of citizenship behaviours and employed latent class analysis using two cohorts of the Australian "National Assessment Program: Civics and Citizenship". Analyses yielded four groups for both the importance of conventional citizenship and the importance of social movement-related citizenship. About one-third of all students were "political enthusiasts", as they were likely to endorse all kinds of citizenship behaviour. These patterns were stable across cohorts, but some latent class sizes varied between both cohorts. The findings of this innovative approach to the study of good citizenship are linked to previous research, and possible explanations for the differences between both cohorts--cohort, lifecycle, and period effects--and the potential of person-centred quantitative research for civics and citizenship education and policy are discussed.