Toward a Theory of Behaviour Change in Social Marketing

Abstract

Contemporary societies all over the world face issues that have been characterised as “wicked problems”. Wicked problems are difficult to define and are ingrained in complex systems, with numerous stakeholders involved. Although the causes of such problems cannot be easily delineated, they commonly can be associated with the practice of preventable behaviours at the individual level. An example is sedentary behaviours, which combined with an unhealthful diet can lead to problems like obesity, diabetes type 2 and more (World Health Organization, 2015c). Obesity is considered to be one of the leading causes of mortality globally. In Australia, for example, almost two thirds of the population are considered to be overweight or obese (Australian Institute of Health and Welfare, 2016). A few examples of other important societal issues are harmful alcohol consumption, environmental degradation, and smoking, among others (Centers for Disease Control and Prevention, 2016). There is an urgent need for effective interventions that can achieve positive behaviour change to decrease the incidence of these issues. Social marketing is a discipline that utilises the tools and techniques of commercial marketing to combat social issues delivering programs valued by citizens and communities (Gordon, 2011). The ultimate goal of social marketing is to change behaviours for the better of society, by influencing people to engage in the desired behaviour. Social marketing is considered to be an effective approach to tackle social issues, and it has been applied to a range of behaviours, the most common being health (Gordon, McDermott, Stead, & Angus, 2006). Social marketers developed a set of principles that can improve social marketing programs’ effectiveness, named the Social Marketing Benchmark Criteria (SMBC). There are eight elements, which combined, can generate behaviour change, namely: behaviour change, audience orientation, insight, segmentation, exchange, marketing mix, theory, and competition. This research focusses on theory in social marketing, and the core criterion behaviour change. Theory provides a structured framework that can serve as a guide to practitioners in the development and implementation of social marketing programs, as well as to researchers. Researchers argue that theory use is fundamental to achieve successful programs (Eagle et al., 2013; Lefebvre, 2001). However, evidence in the literature shows that theory use is scarce (Luca & Suggs, 2013; Truong, 2014; Truong & Dang, 2017) and the majority of theories applied in social marketing are borrowed from other fields. Furthermore, even when theory is reportedly used, it is unclear how theory is applied to the research (Pang, Kubacki, & Rundle-Thiele, 2017). An examination of the most commonly used theories employed in the social marketing discipline show that they approach behaviour as a static phenomenon limiting research attention to behaviour and not behavioural change. Since social marketing’s core is behavioural change and theory should provide guidance on to how to achieve change, theories used in social marketing must be behaviour change, and not behaviour, theories. The overarching aim of this thesis is to take a first step toward developing a theory of behaviour change in social marketing. This thesis starts by understanding the process of behaviour change, and empirically examines how behaviour change differs from behaviour. Additionally, this thesis empirically examines determinants of behavioural change, and applies different methodologies to assess change. More specifically, the purpose of this research is to; first, understand whether behaviour and behavioural change are conceptually and empirically distinct; second, explore the multifaceted characteristics of behaviour change, and empirically test what are the determinants of behaviour change; and third, test the potential of a dynamic methodology to empirically examine change in social marketing. To achieve the aims of this project, three studies were conducted. Study 1 involved a conceptual and operational distinction of the concepts of behaviour and behavioural change. This study aimed to illustrate with empirical evidence from one case study, that determinants of behaviour may not the same as determinants of behavioural change. The method chosen for statistical analysis in Study 1 was multiple linear regression using time point 2 for analysis of behavioural determinants, and change scores for dependent (DV) and independent variables (IV) to examine behavioural change and its determinants. Findings from this study demonstrate that the two concepts are distinct. Empirical evidence shows that when a static model was examined in the context of walking to and from school behaviour, determinants of the behaviour at time point 2 were intentions and barriers to walk to and from school. The dynamic model, analysing only changers and using change scores as IVs and DV, found that a change in injunctive norms was statistically significant in explaining change in walking to and from school. Study 2 was an empirical exploratory study that aimed at examining which determinants were associated with behaviour change, by looking at the different types of behaviour change. Study 2 also applied multiple linear regression for statistical analysis. A series of tests were performed to examine the complexities and different forms of looking at behavioural change. To assess behavioural determinants in a physical activity context, a static model was first tested. Next, a series of tests to explore behaviour change were undertaken. First, a model testing explanation of change using static variables at time point 1 was conducted. The second model involved using change scores for dependent and independent variables to examine a completely dynamic behavioural change model. Lastly, due to change being dynamic and having different directions, two models testing the determinants of different types of behaviour change, namely undesired (negative changers only) and desired (positive changes only) were conducted. Findings indicate that behaviour change should not be treated as one thing. The key outcome of this study is that care should be taken when assessing change, since determinants differ for the different types of change. Study 3 aimed to investigate the potential of a dynamic methodology, the Hidden Markov Model, to simultaneously assess determinants of both behaviour and behavioural change for social marketing. Due to the complex nature of behaviour change, a methodology that can capture the dynamics of change, such as different behavioural states, change rates and directions of change is needed. Study 3 involved an empirical examination to explore the potential of using the Hidden Markov Model to assess behaviour change in social marketing. This research contributes to the literature by advancing social marketing theoretical enquiry beyond static behavioural explanation and prediction, representing a move beyond dominant cross-sectional research designs evident in downstream social marketing studies. Thus, this research takes a first step toward the development of a Theory of Behaviour Change. A future program of work that aims to expand research focus towards the factors that can explain behaviour change aligns theory more closely to social marketing’s end game, namely behavioural change.