Understanding variations in participation across sports - statistical modelling

Every sport has unique characteristics which will influence its level and patterns of participation.

Having first created a ‘selection’ model (to identify those who engage in at least some sport), Mindshare then created 11 separate sport models to better understand participation in each sport. The 11 sports modelled were: athletics, badminton, cricket, cycling, football, golf, rugby league, rugby union, squash, swimming, and tennis.

Whilst the modeling identified considerable variation in the characteristics of participation across the 11 sports, more general findings included:

• Gender gap: women are less likely to participate across most of the sports tested (athletics, badminton, cricket, cycling, football, golf, rugby league, rugby union and squash).
• Asian ethnicity: of the 11 sports, Asian people are significantly less likely to participate than other ethnicities in athletics and rugby union but more likely to participate in badminton and cricket
• Club sports: for the team based sports of rugby league, rugby union, cricket and football, there is a negative relationship between sports club membership and gym membership
• Family sports: in contrast to many sports, frequency of participation in cricket and swimming increases with the number of children in the household. Furthermore, for tennis and badminton, having older children (rather than younger children) increases the frequency of participation.

A fuller summary of the findings from the sports modelling work, including sport by sport summaries, work is available to download.

A fuller explanation of the modelling undertaken can be found in section 5 of the full technical report.

Alongside the quantitative findings, case studies for athletics and tennis were produced through the qualitative element the project and provide further interesting insights into these two sports.