3 | 81 When do young adults stop practising a sport? METHODS Data and research strategy To answer our research questions, we used individual life-course data from the SportersMonitor 2010 (Van den Dool, 2010). These data refer to respondents from the GfK ConsumerJury Panel, which adequately represents the noninstitutionalised Dutch population aged 15 years and older. Respondents (3540 individuals, net response rate 67.3%) participated through computer-assisted web interviews (CAWI). One module in the online survey registered important aspects of an individual’s life course, such as the exact occurrence of major life events and a complete history of sport participation using retrospective questioning. Questions referred to personal behaviours that were relatively easy to remember, for example, the age at which an event occurred, ages of starting and stopping, participation context and frequency (Engel & Nagel, 2011). So, various techniques structured this autobiographical recall process to minimise recall bias (Reimer & Matthes, 2007). To test our expectations, we applied discrete time models (Allison, 1984). This event history technique allowed us to accurately reflect the dynamic character of sport participation and the timing and occurrence of events, to better understand the effects of life events on stopping sport participation. We constructed a person-period file in which every record held information on a respondent for a particular year, like the occurrence of certain life events and sport behaviour. A person entered the risk set at the age of 18 years and was removed from the risk set after age 35. Respondents who were younger than 35 years at the time of the interviews were included in the risk set up to the age upon the interview. We studied stopping participation in a sport (2854 events) and ending a sport club membership (1447 events) and created two different risk sets for these events. Out of the 3540 respondents, 1268 did not practise a sport at all, and 2010 were not a member of a sport club between ages 18 and 35. Because such respondents were not at risk of stopping sport participation or ending sport club membership in young adulthood, they were not included in the corresponding risk set. The risk set for stopping participation in a sport therefore consisted of 24,947 person-years in which 2272 respondents were at risk of stopping sport participation, because they were practising at least one sport (regardless of the context). The risk set for ending a sport club
RkJQdWJsaXNoZXIy MjY0ODMw