Transition to secondary school is one of the critical periods in a student's life, occurring as it does at the junction of childhood and adolescence. In mathematics, transition entails the change to a more formal and abstract curriculum, and it seems that student disaffection occurs more frequently in mathematics than in other subjects. Very few longitudinal studies have examined transition to secondary school in mathematics; none in the last few years when there have been enormous changes to the school system. The lens that was chosen through which to view transition was that of perceived control. The perceived control model examines the amount of control students believe they have over the learning process, and in this study the effect of a number of other variables such as beliefs about self, coping strategies and self-regulation were also examined. The outcome measured was engagement in learning; cognitive, emotional and behavioural. The study followed several cohorts of students from late grade 6 to the middle of year 7, using a survey methodology. At each stage of the study students completed a survey measuring aspects of perceived control, beliefs about mathematics, coping, self-regulation and perceptions of the classroom environment. Students were also asked to comment on aspects of primary school and the transition to secondary school, and asked to reflect on their own ability in mathematics and others' perceptions of that ability. Traditional forms of analysis of longitudinal data revealed few changes to beliefs either during the grade 6 year or over the transition to secondary school. However did this reflect a situation of no change, or were there changes masked by the method of data analysis? In order to address this question, cluster analysis was employed to investigate trajectories of change in the perceived control construct over transition. This analysis was successful, finding four distinct patterns of transition. A highly successful group of students took transition in its stride, while a poorly engaged and unmotivated group continued to struggle. A third group, whose members were unsure and not highly engaged in primary school, seemed to find their feet in secondary school and show higher levels of engagement. The fourth and largest group of students would not have been identified by their primary teachers as potentially having problems at secondary school, and yet they suffer a decline in perceived control in combination with a decline in engagement, coping skills, self-regulation and perception of self. These children are particularly at risk of disaffection in mathematics, and a number of recommendations are made to help identify these students and to address some of their common problems.