Introduction

To become a smart, sustainable and inclusive economy by 2020, Europe has set 5 ambitious targets to meet the Europe 2020 goals. One of these 5 headline targets is to reduce the dropout rates to less than 10%. Reducing dropout is important since this dropout is not only detrimental for the individual’s further life and employment, but this dropout is also an important constraint to smart and inclusive growth in Europe (European Commission, 2010). Starting from 2000, there has been a steady decline in the number of dropouts at the European level, to the current level of 14.1% students between 18-24 with at most lower secondary education and not in further education or training. However, the situation varies in the different EU-countries. In this study, we focus on the situation in Flanders, Belgium. More specifically, we focus on the effect of different growth trajectories of engagement on dropout in secondary education. We use a discrete-time survival mixture model to examine whether distinct trajectory patterns of engagement exist and whether these trajectories are associated to the chance to dropout.

Theoretical framework

Traditionally, research focusing on dropout only focused on a fixed set of variables (e.g. achievement, gender, SES) which were measured only once in a student’s career. This limited view on dropout is problematic for two reasons.

First, dropping out of secondary education is not only the result of low achievement for example, but can also be caused by a decline in school engagement. If a student feels more engaged to school, he is more likely to perform better and is less at risk for dropout. In the literature, different conceptualisations of school engagement exist but these conceptualisations are not always (correctly) integrated in empirical research on dropout. In this study, we use the conceptualisation  as proposed by Willms (2003), with a psychological and behavioural component of engagement.

Second – and more important – dropout can be seen as a gradual process of disengagement from school. However, when engagement was previously implemented in dropout research, it was measured only once, thereby neglecting the gradualness of disengagement. Only few studies took the longitudinal nature of disengagement into account. Janosz, Archambault, Morizot, and Pagani (2008) constructed school engagement trajectories for every student, based on five measurements occasions. They identified seven different trajectories, related to different odds for dropout. To our knowledge, the study of Janosz et al. is the only one explicitly modelling the growth of engagement in relation to student dropout.

Research questions

We consider three research questions:

1. Can we define subgroups of students based on different trajectories of school engagement during secondary education?
2. Do different trajectories show a different chance to dropout, i.e. a difference in survival and hazard function?
3. Which factors have an effect on the predicted class membership?

European Commission. (2010). Reducing Early School Leaving. Working paper. Brussels: European Commission. Janosz, M., Archambault, I., Morizot, J., & Pagani, L. S. (2008). School engagement trajectories and their differential predictive relations to dropout. Journal of Social Issues, 64(1), 21-40. Muthén, B., Masyn, K. (2005). Discrete-Time Survival Mixture Analysis. Journal of Educational and Behavioral Statistics, 30(1), 27-58. Van Damme, J., De Fraine, B., Van Landeghem, G., Opdenakker, M. C., & Onghena, P. (2002). A new study on educational effectiveness in secondary schools in Flanders: An introduction. School Effectiveness and School Improvement, 13(4), 383-397. Willms, J.D. (2003). Student Engagement at School: A Sense of Belonging and Participation. Results from PISA 2000, OECD.