Using two-step cluster analysis to identify homogeneous physical activity groups

Purpose – The purpose of this paper is to: first, illustrate how market segmentation using two-step cluster analysis can be used to identify segments in the context of physical activity; second, identified segments are used to offer practical implications for social marketers working in the area of physical activity. Design/methodology/approach – A total of 1,459 respondents residing within 20 kilometres of the Melbourne Central Business District participated in an online survey. The questions in the survey included items relating to respondents’ health perceptions, health knowledge, attitudes, intentions to start a new physical activity, demographics, place of residence and self-reported physical activity. Two-step cluster analysis using the log-likelihood measure was used to reveal natural groupings in the data set. Findings – This research has identified four distinctive segments in the context of physical activity, namely: Young Disinteresteds, Successful Enthusiasts, Vulnerables and Happy Retirees. Research limitations/implications – The study was conducted in March and some sports were not in season at the time of the study, therefore future research should extend the current sample to take seasonality and geography into account and to ensure the clusters are fully representative of the Australian population. Originality/value – This paper contributes to the literature by outlining a two-step cluster analytic approach to segmentation that can be used by social marketers to identify valuable segments when developing social marketing programmes.