66 | Chapter 12 explanation. This is also demonstrated by the finding that people who recently moved out of the parental home were more likely to choose a competitive club sport over a sport practised on an informal basis, as hypothesised. Perhaps when leaving home, old sports are traded for new ones, often in organised, competitive forms. So, it appears that when people move out of the parental home, they do not seem to balk at organisational forms that are demanding in terms of both time and social commitment. Starting to cohabit or getting married had no effect on the odds of starting a sport in general, but it was associated with reduced odds of starting a competitive club sport. This confirms our hypotheses, based on the idea that competitive sports are more difficult to combine with a couple’s joint time budget, and regardless, couples living together have little need for extra social contacts (Kraaykamp et al., 2009). A second major life event toward family formation, having a first child, was associated with smaller odds of starting a sport. The increased care demands imposed by a new baby seem to cut so deeply into individuals’ time budgets that a negative influence results, as hypothesised. This is in line with previous findings that parents with young children participate less in sport activities (Breedveld & Mulleneers, 2011; Tiessen-Raaphorst et al., 2010). Regarding the two major life events that take place later in the life course, we found that starting a sport becomes more likely when the demands of work and family ebbed. In line with our hypotheses, the odds of starting a sport increased both when one’s children left the parental home and upon retirement, and retirement also increased the odds of choosing a competitive club sport over a sport practised on an informal basis. In this respect, it is worth noting that in the past people often entered retirement at a relatively young age (between 55 and 65 in the Netherlands). Previous research has shown a resurgence of sport participation within the group of 55-65 year olds, but found a clear diminishment in frequency of sport participation after the 65th year of age (Tiessen-Raaphorst et al., 2010). For future retirees, then, it is questionable whether the increased likelihood of starting a sport upon retirement will continue. In closing, a number of shortcomings of our research should be pointed out. First, there is the issue of recall error, due to the retrospective questioning method employed. Respondents could have recalled the life events earlier or later than they actually occurred. Given the importance of the life events