Social Marketing Theory Development Goals

Social Marketing Theory Development Goals: An Agenda to Drive Change

Abstract

Environmental, health, and social change is complex and often seen as a problem to be solved.
Theory, as an organizing framework, is under-utilized in behavioral and social sciences.
Individual psychological lens dominates, focusing on individual behavior rather than change.
Advocates for frameworks that extend beyond individuals to all citizens.
Proposes new evaluation approaches to assess individual and structural changes.
Outlines Ten Social Marketing Theory Development Goals (goals) categorized into:
- Research design
- Building social marketing theory
- Methodological innovation
Aims to develop new ways of thinking that will deliver the theory and evidence base for practitioners and policymakers to effect change.

Keywords

Behaviour change
Social marketing
Theory
Theory application

Background

Marketing evolved from economics, psychology, communications, sociology, anthropology, and management science.
In the 1970s, marketing expanded beyond business transactions to social exchange.
Social marketing emerged as a sub-discipline over 40 years ago (Kotler & Zaltman, 1971).
Theories used have been borrowed from other disciplines (Luca & Suggs, 2013; Truong, 2014; Truong & Dang, 2017).
Development of unique social marketing theories has been lacking.
Social marketers need to:
- Encourage people to stop ingrained behaviors (e.g., smoking, phone use while driving).
- Promote adoption of new behaviors (e.g., exercise, recycling).
- Persuade people to modify existing behaviors (e.g., healthier foods, reusable plastics).
Traditional marketing aims to sell products offering unique value.
Social marketing objectives differ, likened to selling the worst product in the market.
Field expansion requires theories acknowledging these differences.
Marketing understands the assumption of utility maximization is flawed, especially with increased product choice and information (Lye, Shao, Rundle-Thiele, & Fausnaugh, 2005).
Marketing is a strategic process centering on heterogeneous markets to drive profit.
Organizations need to work with stakeholders to deliver mutual value.
Economic principles drive marketing to sell more products to more people, more often.
Specializations like advertising and market research are individual-focused.
Parallel sub-disciplines focus on strategy, management, market complexity, and broader systems (Lehmann, 2005).
Societal awareness increases marketing's accountability for negative externalities.
Marketing stimulates demand, contributing to economic growth.
Marketing is blamed for some of society's problems (Gordon, Russell-Bennett, & Lefebvre, 2016).
Business practices such as consumer deception, intrusion, community co-optation and commercialisation, and societal seduction and degeneration are widely documented (Stoeckl & Luedicke, 2015).
Social marketers create value between actors within a socio-material configuration (Kimbell, 2011; Meroni & Sangiorgi, 2011).
Programme planning considers citizen's needs and the wider environment.
Citizen requirements are balanced against stakeholder resources to facilitate sustained change.
Healthy Together Victoria used social marketing to change hearts and minds, altering structures for healthier choices (Venturini, 2016).
Duane, Domegan, McHugh, and Devaney (2016) demonstrated why change failed via stakeholder identification of structural barriers.

The Role of Theory

Clear direction is needed for social marketing practitioners to understand what to do, when, and where.
Theory provides an organizing framework.
Theory represents 'the accumulated knowledge of the mechanisms of action (mediators) and moderators of change as well as the a priori assumptions that can deliver this understanding' (Davis, Campbell, Hildon, Hobbs, & Michie, 2015, p. 324).
Despite acknowledging the need for theory (Gordon et al., 2016), knowledge of levers for change is lacking.
Pang, Kubacki, and Rundle-Thiele (2017) found that less than one-third of school-based interventions applied or tested theories.
Prestwich et al. (2014, 2015) found that 36-89% of health interventions are not explicitly based on theory.
When theory is used, social marketing research defaults to an individual psychological lens (Truong & Dang, 2017).
This paper outlines 10 Social Marketing Theory Development Goals to stimulate the development of social marketing theories to inform policy and practice.
The paper offers guidance on directing research towards understanding levers of change and overcoming theory inertia.
Social marketing researchers will be able to offer frameworks outlining how social marketing works in partnership to create mutual value across constellations within complex stakeholder systems.

Literature Review

Growing recognition exists within social marketing and behavioral science that theory application in the design, implementation, and evaluation of interventions enhances outcomes and facilitates future refinement and improvement (Davis et al., 2015; Glanz & Bishop, 2010; Truong & Dang, 2017).
Theory enables appropriate identification of antecedents of past and current behaviors, intentions to change behavior, and the causal determinants of behavior change (Hardeman et al., 2005; Michie & Abraham, 2004; Michie, Johnston, Francis, Hardeman, & Eccles, 2008).
This understanding should inform the selection and tailoring of component behavior change techniques (Michie, Johnston, et al., 2008; Michie & Prestwich, 2010; Rothman, 2004).
For example, Schuster, Kubacki, and Rundle-Thiele (2015) drew on the Theory of Planned Behavior (TPB) to identify segments of caregivers based on their behavioral, normative, and control beliefs, thereby delivering actionable insights to inform programme planning. This study was able to identify that intentions were low for families living more than five kilometers from their school.
Identification of theoretically derived mechanisms of action (i.e., mediators) enables researchers to determine why unsuccessful interventions have failed.
For example, researchers can examine if the intervention has had no effect upon the hypothesised mediator, or the hypothesised (and successfully influenced) mediator has had no effect upon behaviour.
Interventions based upon a well-articulated set of theoretical principles provide an invaluable opportunity to test the strength and accuracy of theory (Rothman, 2009).
This then aids in the development of potentially more useful theories which support the optimisation of future interventions (Michie, Hardeman, et al., 2008; Rothman, 2004).
The literature indicates that theory summarises the cumulative knowledge of how to change behaviour across different populations, behaviours and contexts, thereby delivering structured guidance (Michie & Prestwich, 2010).

Levels of Theory Use

The question of whether interventions explicitly based on theory are more effective is complex (Davis et al., 2015).
Without rigorous application and testing, understanding the effectiveness of applying theory is limited.
Research indicates the use of theories and models is scarce (Davis et al., 2015; Luca & Suggs, 2013; Michie & Prestwich, 2010; Painter, Borba, Hynes, Mays, & Glanz, 2008; Prestwich et al., 2014; Truong, 2014; Truong & Dang, 2017).
When theory use is reported, the detail surrounding theory application in research and practice is neither clearly nor consistently reported (David & Rundle-Thiele, 2018).
Consequently, evidence on the role of theory in intervention effectiveness remains mixed (Kadir & Rundle-Thiele, 2018).
Given the lack of clear guidance from scholars, practitioners report finding theory confusing and intimidating.
Choice of theory is acknowledged as challenging (Davis et al., 2015), but practitioners want theory to help design and implement interventions with the greatest possible impact (Davidoff, Dixon-Woods, Leviton, & Michie, 2015, p. 1).
While there are many theories to choose from (Chatterton & Wilson, 2014; Gainforth, West, & Michie, 2015), there is little guidance on which theory to use, for what purpose, and how to apply theory to deliver behavioral change.

Extending the Theoretical Lens

The number of different models and theories relating to understanding and predicting behaviour have been estimated at between 60 and 83 (Chatterton & Wilson, 2014; Davis et al., 2015; Gainforth et al., 2015), yet few are behavior change focussed.
Theories that focus attention on how people think, feel, and act such as the TPB (Ajzen, 1991), Social Cognitive Theory (SCT; Bandura, 1986), The Transtheoretical Model or Stages of Change Model (Whitelaw, Baldwin, Bunton, & Flynn, 2000), the Health Belief Model (HBM; Sundstrom et al., 2015), and Diffusion of Innovation Theory (Rogers, 2010), dominate social marketing research practice.
Many of the theories listed fail to account for complexity arising when a larger systems’ viewpoint is taken.
Other theoretical approaches such as socio-ecological or systems approaches acknowledge that individuals are influenced by their surrounding social and built environments (see Brennan, Previte, & Fry, 2016).
A preponderance of a small number of theories being repeatedly used to inform intervention design inadvertently delivers a dominant lens through which investigators come to understand the domains in which they work (Rothman, 2009).
Although these theories may serve their purpose (i.e. provide an overview of factors related to predicting behaviour), investigators may ‘habituate to the perspective afforded by a theory in the way that one can forget that one is wearing glasses’ (Rothman, 2009, p. 150s).
In addition, it is noted that while many (if not all) theories offer relevance, all require rigorous testing and replication (David & Rundle-Thiele, 2018) to determine their analytical and predictive powers to deliver change across diverse environmental, health and social contexts; given that levels of theory application are weak at best (Pang, Kubacki, & Rundle-Thiele, 2017).
A review of the literature indicates that strong support for theory use exists. Theory-driven approaches to intervention development have been found to enhance outcomes across socio-economic groups (Schneider, 2006).
Moreover, there is substantial evidence that when theories are used to inform the development of interventions in practice, the interventions have proven to be more successful than interventions based on ‘practical experience’ alone (Fishbein, Hall-Jamieson, Zimmer, Von Haeften, & Nabi, 2002; Fishbein et al., 2001).
Poor theoretical underpinning makes it difficult to understand and explain how and why interventions succeed or fail in achieving objectives set (Nilsen, 2015); and may also provide an explanation for the mixed evidence on the effectiveness of theory-informed interventions when compared to those not informed by theory at all (Kadir & Rundle-Thiele, 2018).

Social Marketing’s Theory Focus

The degree to which a theory successfully provides a framework for generating testable hypotheses, integrating empirical evidence and ultimately, providing a road map to guide the design and implementation of interventions depends largely on the quality of the theory (or theories) chosen and the quality of their subsequent application.
Criteria exist to guide theorists (for example see Davis et al., 2015, Wacker, 1998) and facilitate a consensus understanding of what constitutes a ‘good’ theory.
Two levels of theory exist: explanatory and predictive.
Explanatory theories deliver an understanding of relationships between all constructs measured at one, single point in time (for example see Table 1 in Crespo Casado & Rundle-Thiele, 2015).
In the absence of a causal research design, the direction of the relationships observed cannot be inferred (Mitchell & James, 2001).
Predictive theories require researchers to analyse constructs and measures at more than two points in time, focussing on the relationship direction.
A review of the social marketing literature indicates that theoretical focus has centred on explaining and predicting individual behaviour with cross sectional research designs and group comparisons for intervention evaluation dominating evaluation practices (see Table 1).
A range of explanatory and predictive theories have been widely tested and applied to understand behaviour and/or intentions to perform behaviour in the future.
Frequently used theories in social marketing include the HBM (Sundstrom et al., 2015), the Theory of Reasoned Action (TRA)/TPB (Ajzen, 1991), the Integrative Model of Behavioural Prediction and Change (IM) (Fishbein & Cappella, 2006) and more (see Brennan, Binney, Parker, Aleti, & Nguyen, 2014 for additional theories).
Available theories in social marketing do not seek to explain why some individuals remain unchanged, which represents a vast gap in current understanding.
Moreover, the prevailing individual and largely psychological theoretical focus restricts understanding to individuals whose behavior needs to change, and in so doing, ignores social marketing best practice which works in partnership to create mutual value across constellations within complex stakeholder systems (see McHugh, Domegan, & Duane, 2018).
Group level assessments (e.g. Carins et al., 2017) aggregate all individuals seeking to make a determination on change, which oversimplifies behavioural change assessments given that some individuals may change positively and others may change in an undesired direction.
Within each individual, predicting behavioural change requires understanding 1) whether the individual transitions from one behavioural state to another (e.g. from smoker to non-smoker, increased rates of physical activity, or a decrease in littering/waste beha- viour), and 2) then determining the factors that predict change (i.e. the transition from one state to another).
Given that behavioural change is the end goal of social marketing, the absence of research attention directed towards behavioural change in theories and methodological approaches in social marketing is concerning.

A Dynamic Approach

Dynamic theories centre measurement on the transitions or changes that have occurred over time across all theoretical constructs.
Behaviour and behavioural change are conceptually and operationally different.
Behaviour is a unit of observation at one point in time.
Behaviour change includes a temporal component and units of observation for one or more behaviours, and other constructs must be measured repeatedly over time (Ployhart & Vandenberg, 2010).
A review of the literature indicates that cross-sectional research designs dominate (see Ployhart & Kim, 2013) social marketing practice, and as a result, our under- standing is largely restricted to static behaviour and its determinants (Ployhart & Vandenberg, 2010).
Relationships between constructs are much more complex, since change can occur in different ways (e.g. incremental, discontinuous, positive and negative).
Ten goals, categorised into three key areas of development (i) research design, (ii) building social marketing theory and (iii) methodological innovation, are proposed as a series of actions to guide the research community to build an evidence base to guide social marketing research, practice and policy.
The 10 goals are outlined in Table 2 and then each goal is discussed in more detail in the following section.
In considering the complexity associated with behavioural change the key is building an evidence base to understand: ‘What works, for whom, in what circumstances, and for how long?’ (Marteau, Ogilvie, Roland, Suhrcke, & Kelly, 2011, p. 264).
In response to the identified gaps in the literature, the 10 goals are presented as a series of actions to guide research under three main themes: research design (near goals), building social market- ing theories (far goals) and innovating research methodologies (far goals).

Research Design

Goal 1: Use Available Theory

In the near-term, researchers must use the available theory base to understand which constructs can explain and predict behaviour and behaviour change.
The most urgent action is to increase the utilisation of available theories.
For example, Schuster et al. (2015) applied the TPB within segmentation analysis, identifying that intentions for a child to walk to school were lowest in parents living more than five kilometers from the school.
This, and other theoretically guided longitudinal research for the project, delivered practical implications suggesting that the walk to school program in its current form reached parents living within a two kilometer radius of schools. Other programs would be needed for the over 2 km group.
Theory utilisation was able to identify the drivers of current behaviours and changes in the way people think to better understand what was working and why the intervention was (and was not) working (Schuster et al., 2015, 2016).

Goal 2: Use Longitudinal Research Designs for Program Evaluation

Given the conceptual differences between behaviour and behaviour change, we expect the determinants of behaviour to be different from that of behaviour change (Sundel & Sundel, 2004).
A review of the literature suggests that the high prevalence of cross- sectional research designs is limiting our understanding of the determinants of change.
Quasi-Experimental (Q-E) designs, or Randomised Controlled Trials (RCT), need to be used (for examples see Aghdas, Talat & Speideh, 2014; Jago et al., 2015).
Such study designs are increasingly emerging in marketing (for example see, Edineger-Schons, Lengler-Graiff, Scheidler & Wieseke, 2018; Rundle-Thiele et al., 2015; Shin, Casidy, Yoon, & Yoon, 2016; Zavattoro & Fay, 2019).

Goal 3: Apply Theory Identification Processes

Goal 3 focusses on the adoption of a theory selection process, which encourages researchers to systematically compare and contrast theories to advance understand- ing.
By understanding which theory offers the greatest predictive capability research- ers can deliver guides for practice while simultaneously building the evidence base to advance knowledge.
Manikam and Russell-Bennett (2016) offer a decision-making framework that researchers can apply to select a theory.
Direct empirical comparisons within the same context are needed to identify theories offering superior predictive capabilities.
This practice is evident in psychology (for examples see Cha & Hu, 2002; Schaefer, Egloff, & Witthöft, 2012; Taylor, Bagozzi, & Gaither, 2005), but very rare in marketing.
For example, Weinstein (1993) tested four competing theories, attempting to provide a comprehensive overview of the strengths and weaknesses of the models assessed.
$R^2$ (R-squared) and Adjusted $R^2$ can be used in traditional multiple linear regression or structural equation modelling (SEM) techniques to compare theories.
The higher the R-squared value, the better the model is in terms of explaining the behaviour(s) of interest.
Alternative techniques for comparing theories exist.
For example, Imai and Tingley (2012) offer a finite mixture model for theory testing where each observation is assumed to be generated either ‘from a statistical model implied by one of the competing theories’ (p. 218) or from a ‘weighted combination of multiple statistical models under consideration’ (p. 218).
In doing so, individuals are probabilistically matched to a theory and understanding of conditions where a theory is more (or less) applicable is gained.

Goal 4: Deliver Strong Levels of Theory Use

Where theory is reported, the quality is low (for an example rating level of theory use see Pang, Kubacki, & Rundle-Thiele, 2017).
Clear reporting is needed to document how theory was applied (Michie & Prestwich, 2010), which requires explicit reporting of links outlining how the intervention and theory components are connected.
To extend on the current evidence base, we need a comprehensive understanding of how inter- ventions are informed by, and tested using theory. Without clear reporting, this under- standing is yet to be realised.
‘Theory is rarely applied or tested with sufficient precision to allow theoretical conclusions to be drawn to inform future intervention development, even in well-designed interventions’ and ‘evaluations should measure change along the hypothesised causal pathway from theoretical determinants of behaviour change, allow- ing mediation analysis’ (Michie, Johnston, et al., 2008, p. 26).
Rigorous intervention analysis will improve identification of potentially effective versus ineffective intervention strategies.
In the absence of clear theory use, the role of theory remains contested.
For example, evidence suggests that theoretically informed interventions are no more effective than non-theory informed counterparts (Kadir & Rundle-Thiele, 2018).
This work is premature given that clear reporting of theory use is not available.
Until empirical evidence suggests otherwise, we hold the view that the application of theory will improve the chances of creating successful interventions (Davis et al., 2015; Glanz & Bishop, 2010; Trifiletti, Gielen, Sleet, & Hopkins, 2005).

Building Social Marketing Theories

Goal 5: Widen Theory Focus Beyond Current Dominant Theories Used

Theoretical and analytical approaches within behavioural change have not focussed on the transition between behavioural states (e.g. smoker to non-smoker).
Owing to the fundamental differences between behaviour and behaviour change, and the end goal of social marketing being behaviour change, the utilisation of theories focussing on behaviour change as a dynamic process, rather than a static behaviour, are of the utmost importance.
Therefore, research employing longitudinal designs, focussed on identifying the transition between behavioural states and the factors that have changed, is urgently needed in social marketing.
By changing the conceptual focus, insight into the determinants of change can be attained by social marketers, and more informed interventions can be developed to clearly identify the levers of change for social market- ing practitioners and policy makers.

Goal 6: Build Social Marketing Theory

Social marketing relies on ‘multiple scientific disciplines to create programmes designed to influence human behaviour on a large scale’ (Smith, 2006, p. i38).
As a result of borrowing theories from various disciplines (e.g., psychology, behavioural science, etc.), social marketing has heavily focused on individual psychological predispositions such as attitudes, perceptions, and behavioural intentions (for example see Schuster et al., 2015).
Consequently, individual-based theories such as the TPB, and HBM have been widely used in social marketing literature (for reviews see Luca & Suggs, 2013; Truong, 2014; Truong & Dang, 2017) directing attention to the people at the heart of a problem whilst overlooking the role of external influences.
Approaches that focus attention on indivi- duals can lead to blame, given a lack of research focus on other influential factors.
Widening the theory lens (e.g. in the context of active school travel – from the TPB to the ECAC model) is called for to advance our understanding of how change can be enacted within and across complex systems.

Goal 7: Build Theories Explaining Why People Don’t Change

As stated previously, social marketing theory focus needs to extend beyond under- standing individual’s psychological predispositions, and as outlined for Goal 3, alterna- tive theories need to be examined to identify alternatives offering higher explanatory and predictive capability.
Murtagh, Rowe, Elliott, McMinn, and Nelson (2012) found that the TPB accounted for 10% of the variance in actual active school travel behavior, while Pang, Rundle-Thiele, et al. (2017) adopted the ECAC model which extended research focus beyond individual psychographic factors to incorporating environmental factors (a walkability index) as well (Sirard & Slater, 2008).
The ECAC model explained 53.7% of the variance, which is five times the levels explained in the Murtagh et al. (2012) study.
Considering the ECAC model was found to provide more explanatory potential than the TPB in the context of active school travel, future behavioural change interventions should aim to utilise and evaluate theories such as the ECAC model to better understand how more change can be achieved within and across a complex system.

Methodological Innovation

Goal 8: Apply Dynamic Research Methodologies

For behavioural change to be assessed, variables have to be consistently measured at different time points, in exactly the same way, to ensure reliability (David & Rundle- Thiele, 2018; Kher & Serva, 2014).
For one to be able to identify and explain the causes of change (determinants), it is necessary to employ a longitudinal design (Ployhart & Vandenberg, 2010) and focus measurement on capturing behavioural change – in other words – capture the transition (or not) between behavioural states.
In addition, evaluation methodologies need to be dynamic to capture change and this requires a temporal aspect (Ployhart & Vandenberg, 2010).
Switching focus to analytical methods such as the Hidden Markov Model (Vermunt, 2004) and change scores (Allison, 1990) delivers opportunities to shift the research focus from behaviour to behavioural change.
Investigating what drives change, and how to most appro- priately assess behavioural change, has been a topic of discussion amongst research- ers for a long time (Singer & Willett, 2003).
As behaviour change is dynamic in nature, evaluative methods, such as the Hidden Markov Model (Vermunt, 2004) which assess change and its determinants dynamically, are necessary to accurately explain change.

Goal 9: Extend Theory to Understanding Desired and Undesired Change

Behavioural change is complex, and social marketing research fails to account for the different types of behaviour change.
Behaviour change can be desired or undesired, as well as linear, nonlinear, or discontinuous (Mitchell & James, 2001; Ployhart & Vandenberg, 2010; Stritch, 2017).
Methods that can recognise behaviour change as a process (such as the Hidden Markov model) and simultaneously examine drivers of no change, desired change and undesired change need to be applied broadly in social marketing.
Moving forward, social marketing researchers need to take the complexities and multi-faceted nature of behavioural change into account by applying new methodologies.

Goal 10: Increase Replication

The scarcity of replication studies in the social sciences, particularly in marketing journals, has long been a cause for concern among many scholars (Evanschitzky, Baumgarth, Hubbard, & Armstrong, 2007).
Replication is defined as a duplication of previously published empirical research and focuses research attention on assessing whether similar findings can be obtained across repeated studies (Hubbard & Armstrong, 2008).
Without replication, the validity, reliability and generalisability of study findings remains unknown, thereby inhibit- ing the advancement of knowledge.
As stated by Hubbard and Vetter (1996, p. 145), ‘whereas uncorroborated research outcomes must be considered tentative, their successful replication promotes confidence in the veracity of a discipline’s cumulative knowledge base’.
Other things being equal, replication protects against the uncritical assimilation of spurious empirical results into the literature.
The importance of replication is evidenced in many publications which have emerged in various marketing and management journals, emphasising the fundamental role of replication in research (e.g. Evanschitzky et al., 2007; Hubbard & Vetter, 1996).
Theories in the social sciences need to be tested in a natural setting with people living their normal lives.
Therefore, one of the most important techniques for verification of scientific knowledge is the replication procedure, however, we note that there is limited evidence of replication studies in the social sciences (Schmidt, 2009).
Without replication indicating consistent and expected theoretical relationships across a diverse range of target populations and target behaviours, the behavioural change research and practice community cannot confidently apply theories.
Our review of the literature indicates that in the absence of repeated and consistent application (i.e. using the same constructs and measures; see David & Rundle-Thiele, 2018) of a single theory applied over contexts’ the robustness of a theory remains to be seen.

Implications for Research, Policy and Practice

Social marketing is often funded directly or indirectly by governments who rely on taxpayers to raise revenue.
The social marketing research community has a responsibility to redirect research focus to effective behaviour change, ensuring that a reliable evidence base is built.
This resource can then be used with confidence by practitioners to design and implement activity aimed at achieving the social, health and environmental outcomes sought.
Every time outcomes are achieved, and a clear pathway of the processes involved is provided, the reputation of social marketing to effect behaviour change is enhanced.
Advancing social marketing theory to deliver a more refined understanding of the levers that can be reliably used to deliver behavioural change outcomes ensures a return on investment.
Communication of positive return on funds invested provides justification for additional investment over time in social marketing research and practice.
Our research focus must extend to ensuring that the work we do is focussed on maximising return on investment.
As marketing practitioners know, delivering reliable solutions builds trust and commitment among clients/partners.
An evidence base will provide confidence that social marketing (in tandem with other behaviour change approaches) should be commissioned.
It would also guide the practitioner community to better understand how they can deliver effective sustained behavioural change.
The issues detailed for social marketing extend more broadly to marketing, and indeed, other behavioural and social science disciplines.
Marketing scholars and social scientists could benefit considerably from adopting the theory development goals out- lined in this paper such as moving theory beyond mere explanation to the prediction of behaviour and behaviour change.
Delivery of theories that can be reliably acted upon to achieve desired outcomes improves practice and in turn can lead to a continued stream of funds to support research that achieves mutual benefit.
Moving forward, a proven evidence base enables policy approaches to be reconsidered.
Regulations are frequently implemented by governments and industry bodies to achieve outcomes such as environmental health, individual health or social welfare; and many of these policies seek to force behaviour change through compliance.
Given that governments who put compliance measures in place consistently experience resistance and reactance, these approaches should always be a last resort.
Delivering an evidence base that can be drawn upon to guide strategies aimed at inducing voluntary behavioural change offers considerable promise for governments given that voluntary action will always be more favourably received than forced compliance through legislation and regulation.
As Rothschild (1999) stated more than 20 years ago ‘carrots’ and ‘promises’ are far more favourably received when compared to a ‘big stick’.
Knowledge of the determinants of behavioural change allows governments to avoid potentially inefficient regulations and criticisms of ‘nanny states,’ and refocus on delivering programmes in partnership with on ground project stakeholders creating value for all within complex systems.
As noted earlier, many theories that are currently used in social marketing focus on individuals and behaviour.
Theory focus needs to broaden to capture social and environmental factors that may be barriers to, or potential enablers, of behaviour change.
Methodological focus needs to extend beyond explaining and predicting behaviour-to- behaviour change.
This can only occur when dynamic research methodologies, drawing from longitudinal research designs, are employed.
At present the social marketing (other behaviour change and the broader marketing) community are operating on imperfect knowledge and ‘expert’ advice drawn from poor theory foundations may therefore be leading interventions to poor behaviour change outcomes.
From a practical viewpoint an understanding of the barriers to and drivers of change is needed so that social marketing practitioners can confidently focus their attention on the factors that are known to lead to behavioural change. Without this, our practice may be seen by some potential stakeholders as akin to magic and the ‘dark arts’.

Conclusion

Acknowledging our current theory inertia, this paper outlined an agenda focussed on delivering an understanding of the levers of change (and the lack of change).
Ten theory development goals were provided along with a series of actions that researchers can apply to generate social marketing’s own theories.
Our research focus needs to move beyond theories of behaviour to theories that explain the changes achieved.
The pressure to communicate the value of our research is mounting and researchers who can deliver theories identifying the drivers of change stand to benefit enormously.
The 10 goals outlined in this paper provide guidance that if implemented, will identify drivers of change.
In sum, replication will build theories over time, and repeated application of longitudinal repeated measure designs applied before-and-after a theoretically guided intervention offers direct evidence of the change achieved during implementation.
Over time establishing an evidence base of the drivers of behavioural change provides individual researchers with impact narratives and this can enable communication of changes achieved from programmes (return on investment).