PurposeFirst, the income elasticities are calculated for different levels of income, for both the decision to spend money on sports and the amount of money that is spent. Second, the study researches whether different operationalisations of income (i.e. family versus personal) result in different elasticity values. Third, the effect of sports-specific and non-sports leisure variables on sports participation is investigated.Design/methodology/approachA representative dataset of 3,775 adults is used containing a wide variety of leisure characteristics, gathered by means of a face-to-face survey. By means of a Tobit regression model both the determining factors of sports expenditure and the income elasticities are analysed.FindingsFor lower income individuals, a rise in income has a relatively bigger influence on the probability to spend money on sports participation, than is the case for higher income individuals. A positive relationship is found with sex (male), education, number of minutes and disciplines of sports and membership of a socio-cultural organisation, while age, watching TV and attending cultural events have a negative effect.Social implicationsThe study provides evidence that income-based segmentation of sports participants could turn out to be an efficient policy tool. By lowering the monetary-burden for lower incomes, it can be expected that participation rates can be raised efficiently.Originality/valueFor the first time the relationship between income and expenditure is explored for different levels of income and for two operationalisations of income. Moreover, the inclusion of non-sports leisure variables allows investigating relationships between sports consumption and other leisure activities.