Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Systematic literature reviews are arguably one of the most popular methods in social research (Webb and Roe 2007; Littell et al. 2008). Petticrew and Roberts (2006) define systematic literature reviews as a method of making sense of a large body of information, and a means of contributing to the answer to questions about what works and what does not […]. They are a method of mapping out areas of uncertainty, and identifying where little or no relevant research has been done, but where new studies are needed (p. 2). Systematic literature reviews aim to identify and evaluate all available research evidence relevant to a question. They may be highly formal, quantitative syntheses or qualitative summaries of observational data (Glasziou et al. 2003). Whether they cover a research topic or a discipline, systematic literature reviews often involve a considerable amount of data and are thus time consuming. However, they are important to benchmarking the progress of research on a topic or within a discipline while informing directions for future research with respect to topical, theoretical, and methodological trends (Truong 2014).

Refereed journal articles are perhaps the most important source of evidence for systematic literature reviews (Luca and Suggs 2013; Carins and Rundle-Thiele 2014; Truong 2014). Yet, other types of publications, such as project reports, are also valuable in that they provide practical information as well as real-world experiences. For example, in reviewing the influence of implementation on program outcomes and the factors affecting implementation Durlak and DuPre (2008) considered not only published academic literature but also additional project reports. In the field of social marketing, Briggs et al. (2012) reviewed both published, peer reviewed studies and publicly available data and trend reports to identify the key influencers of youth in high risk urban communities with respect to teen dating and dating violence. Therefore, systematic literature reviews are an observational research method that can be used to evaluate and synthesise the content of various forms of literature (Glasziou et al. 2003; Petticrew and Roberts 2006; Littell et al. 2008).

Systematic literature reviews can clearly be used as a research method in their own right. However, they can also be combined with other research methods to enhance the validity of the research results by minimising biases. In some cases, these methods are explicitly stated, such as Truong’s (2014) utilisation of the content analysis method in his review of the social marketing literature published from 1998 to 2012. In other reviews, such methods are not stated but can be inferred from the analysis and synthesis of the results (Carins and Rundle-Thiele 2014; Kubacki et al. 2015a, b). Nevertheless, systematic literature reviews share the commonality of using a pre-defined protocol that comprises a clearly stated set of objectives with predetermined eligibility criteria for studies; a systematic search that aims to identify all relevant studies; an assessment of the validity of the results of the included studies; and a systematic presentation, and synthesis, of the characteristics and findings of the included studies (Petticrew and Roberts 2006; Webb and Roe 2007; Green et al. 2008; Littell et al. 2008).

While the advantages of systematic literature reviews are numerous, the main drawback is the potential influence of the reviewer. Reviewer bias potentially constrains decisions on data collection, analysis, and interpretation in favour of the research question or hypothesis (Petticrew and Roberts 2006; Littell et al. 2008). Clarke (2007) suggests an objective systematic literature review is one in which all relevant studies have been identified before their results could influence decisions about their inclusion. This would help overcome the problem of publication and other biases where prior knowledge of the results of a study might influence the reviewer’s decision of whether it should be included in his/her review (Clarke 2007; Littell et al. 2008). However, it needs to be noted that systematic literature reviews are not different from other social science research methods in that the value of the application of the method much depends on the skills and experiences of the researcher and appropriate reflection on the research process.

This chapter will provide an overview of the application of systematic literature reviews in the social marketing field. It will then return to a case study of its application to the examination of formative research in social marketing health interventions.

Systematic Literature Reviews in Social Marketing

Social marketing is a fast growing field of study (Truong et al. 2015). Therefore, there has been a perceived need among researchers and practitioners for conducting systematic literature reviews to demonstrate the effectiveness of social marketing interventions, examine current practices and strategies, and assess the academic landscape of the discipline. As will be shown below, these reviews are common in that they primarily draw upon the academic literature and use Andreasen’s (2002) six social marketing benchmark criteria (behaviour change, consumer research, segmentation and targeting, marketing mix, exchange, and competition) to identify eligible studies.

Gordon et al. (2006) conducted three literature reviews to examine the efficacy of social marketing in improving diet, increasing physical exercise, and tackling substance misuse. Different databases were used to search for potential studies on each topic. Gordon et al. (2006) suggested that social marketing can be a potential approach to tackling public health issues at the individual, organisational, and societal levels. Similar findings were obtained by Stead et al. (2007) who reviewed the effectiveness of social marketing in promoting individual behaviour and generating environmental and regulatory changes in relation to alcohol, tobacco, drug abuse, and physical activity, and Truong and Hall (2013, 2016) who reviewed evidence of social marketing in tourism settings (see also Truong et al. 2014b).

Other systematic reviews seek to examine the ingredients for success and the factors affecting the effectiveness of social marketing interventions. Carins and Rundle-Thiele (2014) identified 34 social marketing studies promoting healthy eating behaviour published between 2000 and 2012. Only six studies were found to meet Andreasen’s (2002) six benchmark criteria and 16 studies reported positive changes in healthy eating behaviour. Carins and Rundle-Thiele (2014) suggest that social marketing interventions can potentially improve their behaviour change outcomes by adopting all the six benchmark criteria. Similar results are also reported by Fujihira et al. (2015) with respect to social marketing physical activity interventions among the elderly; Kubacki et al. (2015a) in the case of alcohol prevention interventions; and Kubacki et al. (2015b) with respect to social marketing studies targeting children under 12 years old.

Systematic literature reviews are also undertaken to investigate the strategies and practices employed in social marketing programs and their potential influence on behaviour change outcomes. For example, Luca and Suggs (2010) indicate that social marketing public health interventions tend to report scant usage of the full marketing mix (product, price, place, promotion, policy, and partnerships). In contrast, there is widespread use of the marketing mix among social marketing programs targeting water and sanitation products (Evans et al. 2014). In particular, the design of the social marketing product is often not grounded in theory (Alcalay and Bell 2001; Luca and Suggs 2013) and involves limited stakeholder input (Buyucek et al. 2015). Most programs tend to focus more on communication channels and activities rather than the location where the target audience access the product and adopt the proposed behaviour (Alcalay and Bell 2001; Edgar et al. 2015). This is because programs often associate the location where the audience receive information with the place where they do the behaviour or access the product (Edgar et al. 2015). Such practice potentially undermines program effectiveness because it deflects social marketers away from environmental and infrastructure change that is an important component of an effective place strategy (Edgar et al. 2015). Furthermore, evaluations of intervention outcomes are often based on self-reported knowledge and behavioural effects, while experimental designs and randomised trials are rarely used (Alcalay and Bell 2001; Luca and Suggs 2010; Evans et al. 2014).

In addition, systematic literature reviews are conducted to assess the academic landscape of the social marketing discipline. Truong (2014) suggests that research on social marketing is dominated by the English-speaking world, with US- and UK-based researchers and institutions having contributed significantly to shaping knowledge in the field. It is argued that if this trend continues, the social marketing field may remain limited not only in terms of topical coverage (i.e. public health), research perspectives (i.e. downstream influence), but also geographically. Therefore, the potential of social marketing for social issues other than health confronting less developed countries may not be realised (Truong 2014). Furthermore, while many scholars claim that social marketing is an academic discipline in its own right (Buyucek et al. 2015), others argue that this is a debatable point. For example, Truong et al. (2014a, b) identified virtually no clear ‘academic home’ for doctoral students to undertake social marketing research, and argued that the legitimacy of social marketing as an academic field is questionable if undergraduate course offerings and postgraduate dissertation completions are considered important criteria (see also Andreasen 2002; Kelly 2013).

On the whole, there are few systematic reviews in the extant literature of how formative research has been undertaken to inform social marketing programs, one notable exception being Carins and Rundle-Thiele (2014), who examined formative research in their review of social marketing interventions promoting healthy eating behaviour. The next section, therefore, seeks to extend this knowledge by systematically reviewing the use of formative research in self-identified social marketing health interventions.

A Systematic Review of Formative Research in Social Marketing Health Interventions

Social scientists recognise that developing effective interventions plays an important role in improving the health and wellbeing of populations. However, delivering effective programs in real-world settings, and maintaining them, requires insightful understandings of the target audience and the many complex contextual factors affecting their lives. It is important that the potential value of new interventions is adequately tested, and this is impossible without attending carefully to the beliefs, attitudes, and behaviours of the target audience, a process that is referred to as formative research in the social, behavioural, and health science literature (Gittelsohn et al. 2006; Donovan and Henley 2010). Within respect to social marketing, formative research provides an opportunity for the social marketer to learn about the target audience and tailor an intervention to their needs and preferences (Donovan and Henley 2010), and it constitutes part of Andreasen’s (2002) six social marketing benchmark criteria as noted earlier. This process often involves the participation of the people for whom interventions will be designed as well as other relevant stakeholders, and uses specific theoretical frameworks to develop formative research approaches and questions (Gittelsohn et al. 2006). However, little is known about the way formative research is undertaken in the social marketing literature (Carins and Rundle-Thiele 2014; Kubacki et al. 2015a, b). To fill this knowledge gap, this section reviews the use of formative research to inform social marketing health interventions. It seeks to answer the following questions:

  • Q1: Is formative research conducted to inform social marketing health interventions?

  • Q2: What are the objectives of formative research as reported by those interventions?

  • Q3: What theories are used in the formative research process?

  • Q4: What methods are employed in the formative research process?

  • Q5: Who are the stakeholders that are involved in the formative research process?

Potential studies are required to meet the following eligibility criteria: report a social marketing intervention, seek to promote behaviour change in any area of the public health field, and are published in peer-reviewed English-language journals between 2000 and 2015.

Search Strategy

To identify potential studies, a systematic search strategy was implemented. First, comprehensive searches were conducted of online databases, including PsycInfo, Medline, Embase, PubMed, EconLit, Social Policy and Practice, Cochrane Database of Systematic Reviews, Health Technology Assessment Database, Database of Abstracts of Reviews of Effects, and Business Source Complete Database. Other online databases were also mined, such as Scopus, JSTOR, and Web of Science (Luca and Suggs 2010, 2013; Carins and Rundle-Thiele 2014; Truong 2014; Kubacki et al. 2015a, b; Rundle-Thiele et al. 2015). A variety of keywords were used in combination in the search, including ‘social marketing’, ‘HIV’, ‘AIDS’, ‘condom use’, ‘safe sex’, ‘alcohol’, ‘substance misuse’, ‘smoking’, ‘testing’, ‘mass media’, ‘campaign’, ‘intervention’, and ‘behaviour change’.

Second, the reference lists of prior systematic reviews were examined for potential studies (e.g. Luca and Suggs 2010, 2013; Carins and Rundle-Thiele 2014; Truong 2014; Truong et al. 2014a; Buyucek et al. 2015; Fujihira et al. 2015; Kubacki et al. 2015a, b). In addition, all issues of Social Marketing Quarterly (from 2000 to 2015) and Journal of Social Marketing (from 2011 to 2015) were searched for relevant articles. Although social marketing studies are published in a wide range of journals, these are recognised as two main outlets that exclusively focus on social marketing research and practice (Truong et al. 2015).

The literature search was undertaken over a 3-month period, from October to December 2015, which resulted in 1691 papers. Given that the online databases that were mined tend to include similar journals, duplicate articles were removed. Next, conference papers, newspaper articles, articles published in non-English languages, and articles published before 2000 were excluded, reducing the number of articles to 1084. The titles and abstracts of these 1084 articles were then examined. Editorials, review articles, conceptual articles, articles without a significant social marketing focus, and articles not reporting social marketing health interventions were removed. A final set of 242 articles reporting 166 social marketing health interventions were included in this review.

Article Coding

The identified articles were coded on seven dimensions of interest: health topic, target audience, use of formative research, formative research objectives, use of theory and/or model, research methods, and stakeholder involvement. Two independent coders participated in the coding process. After each article was coded, the two coders met and discussed any discrepancies that were present. In cases of discrepancy, the two coders re-examined the articles where discrepancies arose, and discussed until mutual agreement was reached.

Results

Up to the end of 2015, 166 self-identified social marketing health interventions were found, which were reported in 242 refereed journal articles as noted. The full list of 242 articles for each intervention is provided in “Appendix”. In terms of location, a large number of these interventions were implemented in developed countries. With respect to topical areas, nutrition was the most popular, which was reported in 31 interventions (18.67 %), followed by alcohol prevention/cessation (24 interventions, 14.46 %), HIV/AIDS prevention (22 interventions, 13.25 %), and physical activity (20 interventions, 12.05 %). Obesity prevention was the focus of 12 interventions, accounting for 7.23 %. Other topics, such as sanitation, cancer prevention, and violence prevention, were reported in a smaller number of interventions. Given the prevalence of these topics, it is understandable that a large majority of the identified interventions targeted schoolchildren, adolescents, college students, and young adults. Adults aged between 30 and 60 were also the target audience of a considerable number of interventions, which could be explained by the fact that many of the above health problems (e.g. HIV/AIDS and cancer) affect increasing numbers of adults on a global scale. Overall, the identified interventions are similar in that they primarily focused on promoting behaviour change at the individual and family levels. This could be because it is often easier to obtain the desired behavioural change outcomes at these levels. Those encouraging behaviour change on a community scale were less popular, such as the 5+ a day campaign (Ashfield-Watt 2006), the Sumter country program (Burroughs et al. 2006), and the PATH initiative (Wilson et al. 2010, 2013).

Use of Formative Research

A large majority (143/166) of the identified social marketing health interventions reported conducting some type of formative research activities (Table 1). Formative research was not reported in the remaining 23 interventions. As can be seen in Table 1, all social marketing efforts targeting obesity prevention, diabetes prevention, and sanitation reported undertaking formative research. In the areas of nutrition, physical activity, alcohol prevention/cessation, and HIV/AIDS, the number of interventions reporting formative research far outweighs those that did not report any form of formative research activities.

Table 1 Analysis of formative research in social marketing health interventions (2000–2015)
Full size table

While some studies provided very detailed information about how formative research was conducted to inform the development of social marketing interventions (e.g. Cortes et al. 2001; Eitel and Delaney 2004; Maddock et al. 2008; Wayman et al. 2008; Hull et al. 2013), others were less clear about the formative research phase of the reported interventions (e.g. Acharya et al. 2006). However, it needs to be noted that some of the identified interventions may have conducted formative research to gain insights into their target audience but they did not report such activities because they only sought to publish intervention outcomes or because limited spaces were allowed by journals.

Objectives of Conducting Formative Research

The 143 interventions that reported conducting some form of formative research activities were investigated with respect to objectives. According to Atkin and Freimuth (2001), formative research can be divided into two categories: preproduction research and production testing (or pretesting). Preproduction research seeks to gather information about audience characteristics, the behaviour of interest, and/or the potential message channels that can be used. Pretesting involves testing initial messages with members of the target audience, where feedback on appropriateness, persuasiveness, comprehension as well as recall is obtained (Atkin and Freimuth 2001). In the current review, 115 interventions (80.42 %) conducted formative research to gain insights into audience knowledge, attitudes, behaviours (Cates et al. 2014; Dickey et al. 2015), the barriers and enablers to adopting the proposed behaviour (Quinn et al. 2007; Pérez-Escamilla 2012), and preferred message channels (Giordano et al. 2013). This was either explicitly stated or implied in these interventions. Those that reported using formative research to pretest messages among members of the target audience account for only 1.4 %, or two interventions (Glik et al. 2001; Russell et al. 2005). In the remaining 26 interventions (18.18 %), formative research was conducted to both obtain information about audience characteristics and pretest messages with members of the target audience.

Theories for Conducting Formative Research

Social marketing draws upon the appropriate use of behavioural theory to provide frameworks for developing intervention strategies by specifying the determinants of health behaviour. However, recent research suggests that social marketing campaigns tend to report scant usage of theories and models (Luca and Suggs 2013). In the current review, only 33 (23.08 %) of the 143 interventions that reported conducting some form of formative research activities used theories or models in the formative research phase. Theory and model use was not explicitly stated in the remaining 110 interventions (76.92 %). Of those reporting using a theory/model, the Theory of Planned Behaviour/Reasoned Action was used most often (ten interventions), followed by the Stages of Change Theory (nine interventions), the Health Belief Model (seven interventions), and the Social-Cognitive Theory (seven interventions). Other theories/models were used less often, including Diffusion of Innovations Theory, Social-Ecological Model, and Social Learning Theory, among others. This finding is relatively consistent with prior studies (Truong et al. 2014a, b).

Some interventions used specific theories or models to develop formative research approaches and questions. In the Be Under Your Own Influence campaign, the Theory of Reasoned Action (TRA) was used to explore adolescent girls’ perceptions of the costs and benefits of living a drug-free lifestyle (Kelly et al. 2006). The Theory of Planned Behaviour (TPB), an extension from the TRA, was utilised to guide the development of the Wheeling Walks campaign’s messages, where the core beliefs in attitudes, norm, and perceived behavioural control factors that supported walking for exercise for the target audience were identified (Reger et al. 2002). Other theories were also used to develop campaign messages, such as the Health Belief Model and the Stages of Change Theory in the Control Your Diabetes. For Life project (Gallivan et al. 2007), and the TPB and the Social-Cognitive Theory in the VERB project (Huhman et al. 2007). Overall, most of the identified interventions used theories to gain insights into the personal factors affecting the behavioural choices of the target audience. Only a limited number of interventions sought to examine the contextual and environmental factors that influenced the behaviour of the target audience, and these were primarily based on the Ecological Model or the Social-Cognitive Theory. Examples include the Get Firefighters Moving campaign (Staley 2009) and the Power Play! campaign (Keihner et al. 2011). Meanwhile, theory and model use was mentioned in some interventions but how they contributed to the formative research phase was not clearly stated (e.g., Long et al. 2008; McCausland et al. 2009).

Research Methods for Conducting Formative Research

Prior research suggests that qualitative research methods are predominantly used in social marketing studies (Truong 2014). Relatively similar findings are found in the current review, where 62 (43.36 %) of the 143 identified interventions employed qualitative methods in their formative research phase. Quantitative and mixed methods approaches were adopted in 39 (27.27 %) and 42 (29.37 %) of the identified interventions, respectively.

Among those interventions that used qualitative methods in their formative research phase, eight combined three methods (literature review, focus groups, and interviews) (e.g. Quinn et al. 2006, 2009; Fraze et al. 2007) and 18 combined two methods (literature review and focus groups, literature review and interviews, focus groups and interviews, or focus groups and observations) (e.g. Friedman et al. 2014; Lynch et al. 2014). In both cases, a review of literature was often conducted to learn from prior interventions and/or regulatory documents, which was followed by focus group discussions or interviews with members of the target audience. The remaining 36 interventions based their formative research on a single qualitative method, where focus group discussions were the most frequently used (e.g. Hull et al. 2013; Thomson et al. 2013; Plant et al. 2014). By contrast, no combinations of quantitative research methods were found. Indeed, all of the 39 interventions that reported using quantitative methods based their formative research on questionnaire surveys (e.g. Kremer et al. 2011; Croker et al. 2012).

Stakeholder Involvement in Formative Research

Social marketing seeks to promote behavioural change for the benefit of individuals, communities, and society. This is a challenge not only because it is generally difficult to define a behaviour and communicate its (often intangible) benefits to the target group, but also because behaviour change takes place in complicated, or at least contested contexts (Lefebvre 2013). Therefore, it is widely recognised that successful social marketing interventions require the involvement of different stakeholders, such as interest groups, the media, community organisations, private and public institutions, and policy makers (Kotler and Lee 2009; Donovan and Henley 2010), who may participate in the design, implementation, and/or evaluation of social marketing initiatives. The involvement of such stakeholders in social marketing interventions has been described in the forms of partnerships (Abercrombie et al. 2012), coalitions (Kennedy et al. 2000), alliances (Andreasen and Drumwright 2001), and community based social marketing (Flocks et al. 2001). Therefore, identifying which stakeholders are involved in formative research is important not only because it represents the first step towards understanding stakeholder involvement in social marketing initiatives, but also because of the extent to which it should help underlie any behavioural change strategy.

In the current review, the involvement of stakeholders in the formative research phase of the identified interventions varies significantly by context. The number of stakeholders tended to be very limited among interventions that were undertaken in college settings, where students were the target audience and, at the same time, the participants in focus group discussions, interviews, and/or surveys (e.g. Gomberg et al. 2001; Mattern and Neighbors 2004). Stakeholder involvement was greater among interventions that targeted young and school-aged children, where parents, teachers, and school staff were engaged in formative research activities. Examples include Project FIT (Eisenmann et al. 2011; Paek et al. 2015), Ma’alahi Youth project (Fotu et al. 2011), and Food n Fun (Cork 2008). The largest number of stakeholders was reported in interventions that sought to promote community-wide change. For example, the Thunder and Lightning and Rain campaign engaged people with diabetes, the public, health professionals, payers and purchasers of healthcare services, and stakeholder groups in its formative research activities (Almendarez et al. 2004). Likewise, the Let’s Go Local campaign attracted not only community residents but also advocacy groups, NGOs, and policy makers in the formative research phase that included interviews and community consultation (Englberger et al. 2010, 2011; Kaufer et al. 2010). Overall, a large majority of the identified interventions conducted formative research within their target audience groups. Relatively limited involvement was found, particularly of policy makers, with notable exceptions being the Alcohol and Pregnancy project (Elliott et al. 2006; France et al. 2010; Payne et al. 2011a, b, c) and the Let’s Go Local campaign (Kaufer et al. 2010; Englberger et al. 2010, 2011), among others.

Discussion and Conclusion

Social marketing has developed relatively quickly over the last 45 years, leading to its potential being explored in many new areas other than health, such as environmental protection (Takahashi 2009), poverty alleviation (Kotler and Lee 2009), tourism (Truong and Hall 2013, 2015, 2016), and animal conservation (Drury 2011; Truong et al. 2016). Therefore, systematically reviewing the social marketing literature has become vital to informing future research and practice. This chapter has explored some of the potential applications of the systematic literature review method to social marketing. The most significant aspect of systematic literature reviews is a clear understanding of the process and the systematic selection and synthesis of appropriate studies. The case study that has been presented in this chapter illustrates both the process of undertaking a systematic literature review and its application to an under-researched issue in contemporary social marketing debates. It has suggested that a large majority of social marketing health interventions reported conducting some form of formative research activities to gain insights into the target audience’s beliefs, attitudes, behaviours, and preferred communication channels. Pretesting of messages with members of the target audience was less popular. This finding suggests that social marketing researches and practitioners appear to understand the true nature of formative research of trying to understand the target audience before developing or implementing a behaviour change strategy. While in some interventions detailed information about the formative research phase was provided, in others this was less clear or was only implied. It is recommended that social marketing health campaigns clearly report their formative research activities, which would help establish a clear linkage between formative research and intervention outcomes. In addition, reporting such activities may provide valuable lessons for the design, implementation, and evaluation of future campaigns. To realise this, academic journal editors need to recognise that reporting the formative research phase of any social marketing intervention is as important as reporting its outcomes.

If social marketing is concerned with changing behaviours, then it is often assumed that this is undertaken upon the use of appropriate theories and models to understand what actually causes people to change their behaviours (Lefebvre 2013). Lefebvre (2013) suggests that theories serve to explain how and why things are related, assist in identifying what should be focused upon, suggest what questions should be asked, help formulate assumptions about what should be done about a social problem, suggest the type of outcomes that should be set, and determine how success should be measured. However, this chapter has found that most of the identified social marketing health interventions did not base their formative research on any theory and model. This is not surprising given that prior research has suggested that many social marketing campaigns tend to be undertaken on the basis of lay knowledge and assumptions, often without reference to theory or evidence-based methods of changing behaviours (Luca and Suggs 2013; Truong 2014). If social marketing interventions are developed on the basis of intuition or personal beliefs with respect to individual behaviour and responsibility, then it is difficult to identify common factors in effective interventions.

Most interventions reported using theories/models to identify the personal factors affecting the behaviour of the target audience. Only a few interventions used the Social-Cognitive Theory or the Ecological Model to understand the broad contextual and environmental influencers of the behavioural choices of the target audience. Again, this comes as no surprise since most social marketing programs tend to focus on individual behaviour change (Lefebvre 2013; Truong 2014). Lefebvre (2013, p. 79) argues that it is only when social marketers switch to social cognitive and diffusion theories that [they] begin to consider aspects of a person’s “outside world” or environment in solving the puzzles that are presented [to them]. Some recent research has explored the potential of social marketing in influencing environmental change rather than individual change, and has offered further implications for the use of the Social-Cognitive and Ecological Models (Truong 2016).

In terms of methods, nearly half (62) of the identified interventions employed qualitative methods to conduct formative research, where combinations of two and three methods (literature review, focus groups, interviews) were found in 26 interventions. The remaining 36 interventions used a single qualitative method in their formative research phase, where focus group discussions were the most frequently reported. Questionnaire surveys were the only quantitative method reported. On the whole, ethnographic research methods were rarely used. While qualitative methods, such as focus groups and interviews, are useful in that they allow the social marketer to understand the social problems of interest from the perspective of the target audience, they are not always reliable, particularly if they are not complemented by other methods. As Western (2007, p. 329) has critiqued with respect to the use of focus groups, If you ask people conscious questions about unconscious processes, they will be happy to offer you their theories. But most of the time, these theories are wrong. This has been referred to as the ‘say-mean’ gap, which typically arises when the target group makes non-routine decisions (Zaltman and Zaltman 2008, p. 9). With respect to social marketing, the gap between saying and meaning may arise when a culturally sensitive behaviour is promoted. For example, Asian men who have sex with men may not be comfortable with disclosing their HIV status and sexual behaviour because they want to protect themselves and their family from shame (Yoshioka and Schusmack 2001).

Likewise, the use of surveys to examine the target audience’s beliefs, attitudes, and behaviours may also not be sufficient. As Zaltman and Zaltman (2008) argue, when the audience group is asked to attach ratings of agreement to statements, the response reveals only thoughts about what the social marketer deems important but which might not actually be the most relevant drivers of individual behaviour. Put another way, the audience is responding to ideas imposed on them, not generated by them. To overcome this barrier, it is necessary that ethnographic research methods, such as observations, should be employed to provide additional evidence regarding the beliefs, attitudes, and behaviours of the target group. In addition, these methods should explore not just individuals’ beliefs, attitudes, and behaviours, but also the socio-cultural context and the structural barriers and enablers to adopting the proposed behaviour. To this end, all relevant stakeholders should be involved in the formative research phase and their views integrated in the design and implementation of social marketing interventions (Donovan and Henley 2010; Lefebvre 2013). This study has found limited participation of stakeholders, particularly policy makers in formative research activities. Although it is generally difficult to access policy makers and senior government officials, these people are often gatekeepers of information and potential sources of rich data for researchers. Because they have access to special data, knowledge, and power, they are providers of valuable policy information that would be very helpful for social marketers to develop upstream strategies. The combination of different research methods and stakeholder input in the formative research process would contribute to enhancing the quality of baseline survey questions and the outcome evaluation study.

Finally, some limitations in this chapter should be noted. First, this chapter has explored some potential applications of the systematic literature review method and illustrated it through the examination of formative research in social marketing health interventions. Only interventions labelling themselves as social marketing were analysed. Future research may, therefore, examine the use of formative research by interventions that do not label themselves in social marketing terms, as well as those that were undertaken in other topic areas. Second, several relevant interventions might have been missed notwithstanding multiple databases were mined. Third, interventions published in non-English languages were not included and thus further research is needed, particularly by scholars outside of the English-speaking world. A greater evidence base would then be established of the use of formative research in social marketing interventions.