The ‘social worlds’ concept: a useful tool for public health-oriented studies of drinking cultures

Abstract

Intervening in heavy drinking cultures within groups below the level of the population has been under-utilized as a means of reducing alcohol-associated harms. We argue that the concept of ‘social worlds’ is useful for the identification and investigation of heavy drinking cultures in collectivities at this level. The concept may also support investigations into other practices with health implications, such as other substance use or gambling. Social worlds are understood to be loosely bounded groups that change over time and any individual may be affiliated with many. Membership of a social world entails shared commitments, practices and norms generated and reiterated through interaction with other members, albeit not necessarily with all members participating together or at once. Social worlds of heavy alcohol consumption are also framed by the settings where people drink, products consumed and technologies used in doing so. As a tool to support public health efforts, we suggest that these social worlds should entail collective drinking, include sufficient members and involve a magnitude of harm to warrant public health investment and be accessible for research and intervention. Researchers can usefully consider how wider forces, including discourses about alcohol and the gendering of drinking practices, are enacted within particular social worlds. Although they may be explored through empirical research, social worlds of heavy drinking are analytic devices rather than perfect reflections of an objective reality. To see them as such allows us to define them strategically, looking for opportunities to modify cultures associated with harms.

Keywords:

• Alcohol
• cultures
• social world
• research

Introduction

While the importance of addressing alcohol availability remains clear (World Health Organization 2017), policy makers have called for renewed attention to ‘drinking cultures’, arguing that this offers additional opportunities to reduce alcohol-related harms (Her Majesty’s Government 2012; Commonwealth of Australia 2018; VicHealth 2019a). Efforts to change drinking cultures can focus on the societal level, for example through campaigns tackling drink driving across a population, or they may be targeted at subgroups where heavy drinking is particularly prevalent, for instance, men who drink to commiserate or celebrate their team’s success at particular sporting events. Understanding drinking in subgroups can support the identification of diverse targeted interventions. Examples of this include social media advertisements encouraging same-sex attracted women to question the place of heavy drinking at community events, or campaigns co-designed with disadvantaged young people to challenge risky drinking practices within this group (VicHealth 2019b). Exploring the social and cultural factors that shape drinking within particular sub-societal entities and implementing focused interventions has been under-utilized as a means of reducing alcohol-associated harms (Room 1975; Savic et al. 2016; Room et al. 2020).

Despite longstanding evidence that cultures influence drinking patterns and a substantial literature on ‘drinking cultures’ (MacAndrew and Edgerton 1969; Järvinen and Room 2007), the meaning of the term is often unarticulated and taken-for-granted, thereby remaining ambiguous (Savic et al. 2016). For the purposes of this article, we understand a drinking culture as ‘the way a group of people drink [alcoholic beverages], including their shared understanding of formal rules, social norms, practices, values and beliefs around what is and what is not socially acceptable when they get together’ (VicHealth 2019a, p. 1). Drinking cultures influence ‘when, where, why and how people drink, how much they drink, their expectations about the effects of different amounts of alcohol, and the behaviors they engage in before, during and after drinking’ (Savic et al. 2016, p. 280). Central to this conception is that norms and practices that frame drinking are shared, and inhere within units of social interaction, or social collectivities. Drinking cultures are not static but evolve through their reiteration in occasions when subsets of the collectivity get together (Hannerz 1992).

Importantly for this article, while this definition outlines the broad contours of the concept of a drinking culture, it leaves open the question of the nature and scale of collectivities in which drinking cultures, and the practices through which they are enacted, may be found. The literature does not provide a definitive answer here. Indeed, drinking culture/s are described at multiple levels: as a globally dispersed cultural orientation of young Anglophone people in the case of the ‘culture of intoxication’ (Measham and Brain 2005; McEwan et al. 2010); at the level of a nation (e.g. the ‘Australian drinking culture’; Sargent 1968); within religious or occupational collectivities (Iacuone 2005; Heath 2009); among groups composed on the basis of shared demographic features, for instance, ‘youth’ (e.g. Järvinen and Room 2007); and as characteristics of friendship networks (e.g. Pennay 2012).

In this article, we argue that the concept of ‘social worlds’ (Strauss 1978; Unruh 1980) has theoretical and analytic utility in alcohol research and in studies of health practices more broadly. The social worlds concept supports identification and investigation of groups below the level of society but above the level of friendship groups, where relatively coherent heavy drinking cultures may operate, and hence which may be targeted for intervention. It provides a means of conceptualizing, and drawing analytic boundaries around, the social collectivities whose practices are the objects of investigation. This is essential when the focus is on changing cultures that are shared.

We draw here on experiences from two research projects, both of which involved investigating several heavy drinking social worlds. The first project explored heavy drinking social worlds of middle-aged and older people. Social worlds investigated were middle-aged same-sex attracted women (MacLean et al. 2019), middle-aged male construction workers (VicHealth 2017a) and middle-aged and older sports bar attendees (VicHealth 2017b). The second considered men’s heavy drinking social worlds, including front line hospitality workers, professional corporate employees, and both sports spectators and (amateur) sports players in rural and metropolitan locales (Roberts et al. 2019). Our focus in this article is on the conceptual framing of studies rather than on research methods that may be employed to conduct them, and on offering suggestions for researchers working on studies of drinking or other cultures that are intended to inform the design of public health interventions.

We begin with an overview of drinking culture research, highlighting the strengths and limitations of this work with respect to a public health imperative of addressing alcohol-related harm. Next, we outline a set of parameters that we suggest are relevant considerations when seeking to identify subgroups of drinkers who merit public health attention. We then make a case for the adoption of the social worlds concept as a ‘sensitizing notion’ and heuristic device for study design. Next, we describe how the social world construct has been integrated within methodological approaches and is used in a practice tool, followed by a consideration of how intersections between social worlds and wider forces such as gender and class might be understood in operationalizing the notion of social worlds of heavy drinking. In concluding the article, we identify opportunities for broader application of this conceptual tool.

Background

The study of drinking cultures has a long and rich history in social science studies of alcohol. Historically, researchers in this field have often focused on the macro-societal level and compared the cultural position of alcohol across different societies and/or nations (Heath 2009; McEwan et al. 2010). Comparative societal-level typologies such as ‘wet’ and ‘dry’ drinking cultures have been central in this regard (e.g. Room and Mäkelä 2000; Room 2010). Yet, it is evident that norms, meanings and practices of drinking are not uniform in large complex societies (Room et al. 2016, 2019). Societal-level conceptions of a drinking culture obscure the diversity of drinking practices and patterns in subgroups below the level of a population as a whole.

Some studies have focused on subgroups within the overall population. Such studies broadly fall into two types. The first type comprises epidemiologically informed analyses of population surveys or of subgroups constructed by researchers on the basis of shared demographic or other common social characteristics (e.g. younger and older people, women and men, professional or trade occupations). The second type of studies are qualitative investigations, largely ethnographic or interview-based in design.

Epidemiologically informed drinking cultures studies have examined drinking patterns (consumption levels, frequency, locations) and identified substantial commonalities within, as well as differences between, groups defined by dimensions such as ethnicity (Agic et al. 2011), gender (Wilsnack et al. 2009), age or generation (Livingston et al. 2016), or other demographic categories. Such research can point toward groups of people who are at elevated risk of alcohol-related harm. However, it tends to yield little information on the cultural norms, meanings and practices underlying the observed commonalities and differences in drinking patterns. Moreover, approaches that aggregate (e.g. in measuring the ‘proportion’ who may drink more than is recommended) across an ascribed demographic or social category erase differences between the individuals within the category. The heavier drinking by some members of the ascribed social category may be an effect of their membership in other social collectivities, rather than being reflective of a shared heavy drinking culture in the ascribed social category. For instance, members of the social category ‘middle-aged women’ will belong to multiple subgroups, some of which have a heavy drinking culture and others that do not. Importantly, also, such research does not usually consider whether any particular member feels an affiliation with or socializes with others in the social category to which she or he is ascribed.

Detailed qualitative explorations of drinking cultures in a specific interacting social group have mostly been ethnographic in design. Researchers have studied people with connections to each other, such as the friends whose alcohol and drug use practices are described by Pennay (2012) and the ‘Skinheads’ in Moore’s (1994) ethnography. Such studies often produce rich and nuanced accounts. However, they tend to report on small groups of people, making it difficult to know how widespread the practices described are and how extensively interventions developed in response may be rolled out. Interview-based studies, and/or multimethod approaches, can offer researchers capacity to access broader groupings than focused ethnographic studies – for instance, researchers have explored drinking practices and meanings through interviews conducted with young Australian university students living on campus (Supski et al. 2017) or combined focus group, interview and observation methods to investigate drinking among men who work in the hospitality industry (Roberts et al. 2019).

As observed above, drinking cultures have been investigated at multiple levels. However, the limitations in research on drinking cultures we have identified indicate that, for public health-oriented research, different conceptual tools are needed to examine drinking cultures in their complexity.

Investigating drinking cultures from a public health perspective

We suggest that studies of drinking cultures within subgroups that are intended to support public health activities should be designed with various concerns in mind. Collective drinking is crucial for the production and reproduction of shared drinking norms and practices – that is, a drinking culture. Thus, groups should be selected such that collective drinking plays a part in a group’s repertoire of activities. Drinking does not need to be a defining activity of the group, and neither do members need to drink with, or indeed even know, all other members – indeed, as long as drinking occurs with some other members, groups may be geographically dispersed. Next, where the intention is to mitigate harm associated with heavy drinking cultures, the groups selected for study should manifest some evidence of drinking patterns and practices that place members at risk of harm (e.g. routinely consuming high quantities of alcohol, indications of alcohol-related problems such as fighting). Relatedly, groups to be investigated should entail sufficient numbers to warrant investments in application of targeted interventions. Benchmarks for assessing whether a social world is sufficient in size to justify expenditure on interventions will vary in relation to availability of resources, range and extent of harms experienced by social world members and likely cost of the interventions. Finally, groups should be accessible for both research and targeted strategies, both of which are likely to be facilitated by the collective nature of drinking in social worlds.

To identify a focus for their enquiries, public health researchers need to define and draw some boundaries around the groups to be investigated – a process that inevitably involves strategic decisions. This means that identified social worlds are heuristic devices, even though they reflect a clustering of cultures and practices that is evident in the world. It should be recognized that individuals in complex societies are simultaneously members of multiple sociocultural entities, that engagements in these groupings are fluid, that social collectivities themselves contain variation, and that forces from outside the social group such as alcohol pricing policies and broader drinking cultures also impact on drinking within a group. In what follows, we argue that the social worlds concept is a useful tool in dealing with these issues.

Conceptualizing the group to be studied: the utility of the ‘social worlds’ concept

The conceptual tools utilized to identify heavy drinking subgroups as part of public health endeavors should provide researchers with the capacity to manage both the complexity and specificity of subgroups where heavy drinking is prevalent. An array of terms has been used to define subgroups below the level of society as a whole, among them ‘subpopulations’, ‘subcultures’, ‘peer crowds’, ‘countercultures’, ‘social worlds’, ‘communities of practice’ and ‘(neo)tribes’ (Thornton 1995; Maffesoli 1996; Moore 2004; Sweetman 2004; Moran et al. 2017). The proliferation of terms reflects social scientists’ unease with sociological applications of the term ‘subculture’, which has often been understood either as marked by deviance and delinquency or celebrated as a form of working-class collective resistance to a dominant culture (Blackman 2014). Some researchers have argued that the subculture notion also implies a collectivity that is too bounded and too central to its members’ identities to explain most affiliations in the contemporary world (Bennett and Kahn-Harris 2004). ‘Peer crowds’ are defined as ‘broader, more distal, macrolevel subcultures’ with which ‘youths identify’, that are constituted by a ‘shared set of behaviors, values, norms and lifestyles’ (Sussman et al. 2007; Moran et al. 2017, p. 389). With its narrow focus on youth (predominantly those in the United States; Sussman et al. 2007), this contemporary reworking of the subcultures concept is likewise limited by the notion of an all-encompassing, fixed and bounded collectivity.

Post-subcultural terms are closer to what we need to meet the requirements for public health outlined above, with their insistence on the constitutive effects of place and the multiple and fluid affiliations that operate for any one person. Maffesoli’s term ‘neo-tribe’ describes groupings that come together through ‘a succession of ambiances, feelings and emotions’ (1996, p. 11). While acknowledging the untidy nature of sociality and that the dynamic and affective basis of affiliation is critical, this formulation does not capture the routinized and day-to-day nature of much social life. Moreover, formulations such as neo-tribe have been critiqued for failing to grapple with the influence of structural forces such as ethnicity, class or gender in affiliative groups, with attention to individual choices so focused that social constraints and inequities are written out altogether (Shildrick 2002; Blackman 2014). The notion of a ‘community of practice’ is appealing for the study of drinking cultures in its suggestion that the shared practice of drinking is what constitutes a social unit. However, defined primarily as a group which meets face-to-face to improve their skills at a particular activity (Wenger 1999), this is too restrictive for our purposes here. The term ‘subpopulation’ describes sub-societal groups understood to share one or more sociodemographic attributes – examples here include groups such as people living in rural locations or people with a disability. However, as discussed above, such ascribed categorical groupings do not necessarily share mutual drinking norms and practices.

A social world has been described in plain terms as:

…a group of people who get together around a common interest or activity. Members of social worlds may or may not necessarily know one another but they share social norms and practices, including expectations about how people behave when they meet (VicHealth 2019a, p. 2).

In the sociological literature ‘social worlds’ have been defined as ‘amorphous and diffuse constellations of actors, organizations, events, and practices which have coalesced into spheres of interest and involvement for participants’ (Unruh 1980, p. 277). The social worlds concept has been attributed to the work of Chicago School theorists, particularly symbolic interactionists, often with reference to Shibutani (1955). The work of Strauss (1978) and Unruh (1980) served to further articulate the concept, and it has thereafter been used (sporadically at best) across disciplines, including sociology, criminology, psychology, and related fields. While the term appears in the titles of many works, it is most often used to describe the broad realm of the social (differentiating it, for example, from the natural world), rather than subgroups as in Strauss (1978) and Unruh (1980). Contemporary exceptions to this descriptive use of ‘social worlds’ can be seen in the field of leisure studies, where researchers have employed the concept as articulated by Strauss and Unruh to the study of long-distance running, climbing and yoga (Robinson et al. 2014; Patterson et al. 2016; Kacperczyk 2019).

We argue the concept of social worlds is useful in drinking culture research in four ways. First, it signals the fuzzy boundaries of such groupings, incorporating the notion that these groups are dynamically constituted, and hence that they change over time. Unlike ‘subculture’, ‘social world’ denotes a grouping where membership may not necessarily be experienced as deliberate. For example, particular drinking cultures are embedded in drinking in sports bars (VicHealth 2017b), but while doing this as a part of one’s life may be enjoyable, and some kind of sense of shared purpose may exist, attending sports bars is unlikely to be a central component in anyone’s identity construction or an enduring commitment for any individual. Most of us are members of many social worlds, and a given social world is but one influence on a practice such as drinking.

Despite amorphous boundaries, a second sense in which the device of ‘social worlds’ supports research on drinking or other collective health practices is that it entails the existence of some kind of coherence within the group. Social worlds entail people with common interests who develop shared sets of ideas or ‘ideologies about how to go about their business’ (Clarke and Star 2008, p. 115). Thus, a social world is one that has coalesced ‘into a meaningful and interactionally important unit of social organisation’ (Unruh 1980, p. 272). In a social world of heavy drinking, consuming alcohol may be just one among many socially constitutive acts of a group, although in many heavy drinking social worlds it may well be an important one. Focusing on social worlds where drinking is a significant element of the group’s practices enables researchers to attend to groups where heavy drinking and potential harms are more apparent, while remaining attuned to the multiple and fluid connections people have in society.

A third form of utility of the social worlds concept is that it situates social interaction as critical in producing and reproducing cultural understandings and practices. Theorists of social worlds insist that the boundaries of social worlds are determined by ‘interaction and communication’ (Unruh 1980, p. 271) or set by the ‘limits of effective communication’ (Shibutani 1955, p. 566). Although their parameters are rarely sharp, social worlds are constituted through shared participation in activities, and often involve some form of induction for newcomers, where ‘experienced members articulate a version of the past, population, practices, and problems’ of the social world (Pollner and Stein 1996, p. 203). This does not mean that everyone in a social world interacts with each other member, only that cultures are reproduced in the context of social engagements. Although not envisaged by theorists who initially developed the concept, a social world can also emerge through virtual engagement between people on the internet (Quercia et al. 2012). In utilizing the concept of social world for our purposes, then, the heavy drinking practices should occur collectively between members and entail interactions between them, albeit not with all members necessarily participating at once or even knowing all other members.

Framing social worlds in this way allows public health researchers to investigate how group dynamics reinforce cultural norms that moderate heavy drinking practices, as well as exacerbating them. Various studies, some inspired by social identity theory, have shown how peers support others in recovery from drug dependency by censuring practices that contravene group norms or expectations, either subtly or explicitly (Best et al. 2016). Norms within networks of substance users (particularly the stigma around uncontrolled use) allow ‘active limit-setting’ and the promotion of more moderate forms of consumption (Duff 2004). Understanding how social worlds monitor and intervene in members’ actions provides opportunities to reinforce norms that involve avoiding heavy drinking or mitigating unsociable drunken behavior. The importance of interaction in constituting social worlds means that the construct cannot be applied in some contexts. For example, while drinking alone at home is influenced by social norms, it does not constitute drinking as part of a social world. Interventions to address solitary home drinking cannot be targeted at drinking cultures that are sustained and reproduced through interaction, as occurs in a social world of heavy drinking.

A fourth advantage of the social world concept for public health research is that relationality within this notion extends beyond the human participants. Social worlds are constituted also through material elements (Strauss 1991; Clarke and Star 2008) or connections between people and the contexts within which they live (Shibutani 1955; Strauss 1978; Unruh 1980). In doing so, the social world concept accommodates an ontological shift that is evident in poststructural research approaches including Actor-Network Theory (Latour 2005), where social life is understood to emerge through dynamic configurations of human and non-human forces, including settings, objects, discourses and norms. Hence, social worlds cohere in places where people meet (e.g. bars where men who work in the financial sector drink after work, or the ‘team bus’ when a sports team collectively travels for an end of season trip (Roberts et al. 2019)). Social worlds are produced and reproduced at least in part around technologies (e.g. the use of corporate credit cards by financial sector employees who ‘sign off’ one another’s buying of rounds for colleagues, or employing mobile phones to take photos to document drinking event sociality on social media, broadcast both in the moment and after the event (Roberts et al. 2019)). The alcohol products that members tend to consume (so that drink product choice can signify a social identity (Emslie et al. 2017; MacLean et al. 2019)) is a further material element of drinking in social worlds.

Using the social worlds construct in research paradigms and practice tools

The social worlds construct does not constitute a research methodology in itself, and it is beyond the scope of this paper to comment on how this concept might inform the design of research methods for data collection. However, we note that several methodological approaches have been used to frame research describing cultural practices within social worlds.

Social practice theory (SPT) (Shove et al. 2012) has been used to frame studies that investigate how drinking norms, meanings and practices impact on and are reproduced using qualitative (Supski et al. 2017; MacLean et al. 2019; Roberts et al. 2019) or quantitative (Ally et al. 2016) data. Although not explored in their analysis, Ally et al. (2016) conclude that SPT can support a fine-grained understanding of drinking in specific groups, and that it is useful in monitoring changes in alcohol norms and patterns within these entities over time. Yet SPT studies such as these tend to leave unexamined the problem, with which we are concerned here, of how to define and draw boundaries around the groups of people who are carriers of the practices being investigated, or how these groups themselves influence members’ practices (Giesbrecht 2018).

The notion of ‘social worlds’ has been adopted in a recent tool, the Alcohol Cultures Framework, developed by the Victorian Health Promotion Foundation (VicHealth) in Australia, to support the design, testing and implementation of public health action on alcohol (VicHealth 2019a). The Framework identifies drinking cultures that operate at the intermediary level of ‘social worlds’ (between the societal and individual levels) as appropriate targets for health promotion, while considering the interaction with forces at other levels. Here, a social world is defined as ‘a group of people who get together around a common interest or activity’. In line with SPT, it proposes that social worlds should be explored though questions about features including shared practices such as drinking at a particular setting (for example, a sports bar), meanings about alcohol use (using alcohol to establish solidarity with others) or skills or conventions (such as expectations of how to respond when others become very intoxicated; VicHealth 2019a). This is intended to support the design of interventions. For example, if research were to uncover that sports bars offered cheap drinks when particular games were televised, and if patrons shared an understanding that the fortunes of one’s team should be celebrated by heavy drinking, it might make sense to limit price reductions that coincide with major sporting events.

Intersections between social worlds and wider social drinking cultures

The challenge for researchers investigating social worlds of heavy drinking is not only to identify the shared meanings or practices circulating within a social world, but also to understand how configurations of wider social forces and discourses influence, are understood and enacted within particular groups.

Drinking cultures in social worlds are always shaped by wider affinities, discourses and social forces, including national or international cultural frames (Järvinen and Room 2007; Room 2010). As in most aspects of our lives, gender and socioeconomic position influence how we drink (Törrönen and Roumeliotis 2014; Lennox et al. 2018). Structural factors, including alcohol availability and other government policies that reflect to some degree a society’s views about how alcohol should be regulated (Babor et al. 2010), are also influences on drinking cultures, as are the activities of the alcohol industry (McEwan et al. 2010). In designing research into drinking cultures that utilizes the notion of social worlds, it is useful to consider how these wider forces might be acknowledged. Indeed, doing so will help researchers avoid over-emphasizing the influence of forces that act from within a social world, rather than those that are more distal.

Social worlds take up, accommodate and transform ideas about alcohol that operate more broadly to their own contexts, such as the widely shared cultural associations between drinking and socializing, celebrating or marking time to relax. In our studies of heavy drinking social worlds, the alcohol consumption practices our respondents referred to frequently reflected broader social discourses about alcohol, albeit with differences in how these were interpreted and enacted across the groups. To a greater or lesser extent, cultural norms about buying drinks for others as a medium of reciprocity (e.g. the Australian custom of ‘shouting’ or ‘round buying’; Sargent 1968) were articulated across the three groups in research on drinking cultures in middle-aged and older social worlds (same-sex attracted women, construction workers and sports bar drinkers), though least for the same-sex attracted women (VicHealth 2017a, 2017b; MacLean et al. 2019). Such commonalities illustrate that while particular social worlds provide much of the normative guidance on drinking and alcohol-related practices, these are inevitably in interaction with broader cultural discourses and practices.

Likewise, gender strongly influences drinking cultures within a social world. Men are likely to drink more than women, and are overrepresented in alcohol-related harm statistics (Berends et al. 2012). Gender does not simply constrain how we drink; it also enables performances of masculinity and femininity. Gendered drinking practices vary between cultural groups, and over time as they evolve, to distinguish masculinity from femininity in varying articulations, underscoring differences between groups of men and women (see, for example, Törrönen and Roumeliotis 2014; Ravn 2018). In the men’s heavy drinking social worlds we studied (Roberts et al. 2019), we found that hegemonic masculine ideals of autonomy and self-determination played an important role in how these men drink, and whether and how they provided care or intervened in a friend’s drinking. For instance, men reported being discreet in providing care to male friends who were drinking heavily, so that they were not seen as encroaching on another man’s autonomy. In research on social worlds of heavy drinking among middle-aged and older people, same-sex attracted women described fewer problems associated with collective drinking than did the members of the other groups within that study (who were exclusively or predominantly male), alongside a greater willingness to intervene gently and supportively when someone became intoxicated (VicHealth 2017a, 2017b; MacLean et al. 2019). This again highlights that gender played out differently across the social worlds we investigated. It is useful to consider how forces such as gender play out in cultures that frame drinking, including greater and lesser willingness among social world members to intervene when someone contravenes boundaries of expected behavior. Where this occurs, efforts to amplify the response are a potential lever for change.

Both macro- and micro-social influences act on any individual and on the drinking cultures that influence the individual’s consumption of alcohol. Hence, in identifying a social world of heavy drinking, it is not necessary, we believe, to find a group where drinking patterns are unanimously heavy. In a study of heavy drinking social worlds among middle-aged and older people, we sought to select social groups where relatively heavy drinking occurs collectively with other group members; however, in each of these groups a sizeable proportion of people consume alcohol moderately or even not at all, and this was reflected in the research samples accessed. But this did not necessarily undermine the relevance of the groups investigated. For example, in research with sports bar drinkers (VicHealth 2017b), we found a high level of acceptance of heavy drinking, and expectation of drinking as intrinsic to being at a sports bar, even among those who did not themselves consume alcohol at a heavy level. This was paralleled in the men’s risky drinking study (Roberts et al. 2019), where male sports players, sports spectators, corporate workers and, to a lesser extent, hospitality workers, all noted similar levels of acceptance of heavy drinking, and a culture that encouraged it, even where they felt comfortable to drink moderately, or sometimes not at all. Understanding moderate drinking practices within a social world – by including in the research frame those who drink moderately – may help identify opportunities to encourage heavier drinkers to adopt less risky drinking styles that still fit within the broader cultural context in which they operate (see, for example, MacLean et al. 2019).

Conclusion: framing heavy drinking social worlds to support the development of interventions

A key public health purpose of researching drinking cultures is to identify opportunities to disrupt problematic drinking norms and practices that operate when specific groups of people drink together. We have argued that the concept of social worlds can be utilized in research to identify and study groups, below the level of society as a whole, where there may be cultural coherence in drinking or other practices and norms. This approach allows researchers to attend to the effects of wider social forces such as gender, government policies and the material settings of life on drinking cultures at a localized level. Focusing at the level of the social world enables a broader reach for research and intervention than is possible when attending to problematic drinking in individuals or their immediate friends and family, and a more nuanced approach to changing cultures that promote harmful drinking than can be achieved at a societal level. Thus, targeting social worlds provides an additional strategy for health interventions, complementing population and individual-level interventions. The social world concept, where cultures are understood to vary between social groupings, also provides a useful approach to understanding the heterogeneous effects of population-level interventions.

The social world concept is thus emerging as a useful conceptual mechanism in alcohol research. It is likely to have merit also in understanding how other cultural practices of interest to public health are embedded in groups below the level of the population as a whole. However, the concept of social world seems to be only rarely used in research on drug use, gambling, physical inactivity or poor mental health, and these are areas where its future application might be considered. Likewise, the concept can also be applied to understanding and promoting collective practices that are associated with health such as exercise (Ditton et al. 1992; Robinson et al. 2014), lighter drinking or alcohol avoidance.

In closing, it is important to acknowledge that although they are explored and clarified through empirical research, social worlds or other conceptual tools used in studies such as ours are heuristic or analytic devices rather than perfect reflections of an objective reality that is independent of the research methods used. The categories that researchers choose can never fully reflect the complexity of people’s lives. This insight allows us to define social worlds strategically, identifying groupings where opportunities exist for public health strategies to modify heavy drinking cultures and reduce harms.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

A.P. is supported by an Australian Research Council Discovery Early Career Researcher Award [DE190101074]. The Centre for Alcohol Policy Research (CAPR) receives core funding from the Foundation for Alcohol Research and Education. The work of CAPR staff on this paper was supported by the Victorian Health Promotion Foundation (VicHealth).