1 | 35 Synthesis LIMITATIONS AND FUTURE AVENUES FOR RESEARCH While this dissertation makes important and relevant contributions to the existing body of knowledge on sport participation over the life course, some limitations bear mentioning, as well as opportunities for future research. To start, future research could invest in empirically analysing changes in resources as the underlying mechanism explaining the impact of major life events on sport participation. Throughout this dissertation, I employed a resource perspective and changes in resources were derived via theoretical expectations formulated in connection with the major life events under investigation. However, because of the lack of dynamic data on respondents’ resources, my quantitative studies come short on empirical rigour concerning changes in the specific resources associated with the occurrence of major life events and the impact of these changes on sport participation. In my last qualitative, narrative study, I was able to examine this, finding that alterations in resources associated with major life events indeed play a decisive role in why and how changes in sport participation occur after experiencing those events. So, for future quantitative research it would be an improvement to also have direct and dynamic measurements of resources (e.g., leisure time, social contacts/network, health condition, financial means) and relate changes in these resources to the occurrence of major life events and changes in sport participation. From the data used in this dissertation, three other limitations arise that I would like to point out. First, in the SportersMonitor 2010, used in my studies regarding starting and stopping a sport, no life course data were gathered on respondents who reported to never have been active in a sport during their lifetime. This was not problematic for my analyses of stopping a sport. Because such respondents were not at risk of stopping sport participation or ending sport club membership, they would not be included in the risk sets anyway. However, it limits my analyses of starting a sport, since I had to use data from a selective sample consisting only those who had participated in at least one sport during their lifetime. The absence of non-sport participants could have affected the results, as they would have reduced the overall likelihood of starting a sport. Future research could benefit from more representative samples consisting of both sport participants and non-sport participants, preventing sample selection bias and improving the generalisation of results. Second, there is the issue of recall error, as I use retrospective life course data
RkJQdWJsaXNoZXIy MjY0ODMw