Parental modeling, education and children's sports and TV time: the ENERGY-project.

OBJECTIVE: We assessed whether differences in children's sports participation and television time according to parental education were mediated by parental modeling. Moreover, we explored the differences between parental and child reports on parental sports participation and television time as potential mediators.

METHODS: 5729 children and 5183 parents participating in the EuropeaN Energy balance Research to prevent excessive weight Gain among Youth (ENERGY-project) during 2010 in seven European countries provided information on sports participation and television time using validated self-report questionnaires. Multilevel country-specific mediation models analyzed the potential mediation effect of parental self-reports and child-reports on parental sports participation and television time.

RESULTS: Significant mediation effect was found for parental self-reported television time in four countries (Greece, Hungary, The Netherlands and Slovenia), with the highest proportion for Slovenia (40%) and the lowest for Greece (21%). Child-reported parental television time showed mediation effect in Greece only. Parental self-reported sports participation showed significant mediation effect only in Greece. With child-reported parental sports participation, significant mediation was observed in Greece and Norway.

CONCLUSIONS: Parental behaviors appear to be important in explaining parental educational differences in children's sports participation and television time. However, child reports on parental behavior appear to be more relevant than parents' self-reports as correlates of children's own sports participation and television time.