Divergent Drinking Patterns of Restaurant Workers: The Influence of Social Networks and Job Position

Abstract

Restaurant workers have higher rates of problem drinking than most occupational groups. However, little is known about the environmental risks and work characteristics that may lead to these behaviors. An exploration of restaurant workers’ drinking networks may provide important insights into their alcohol consumption patterns, thus guiding workplace prevention efforts. Drawing from social capital theory, this paper examines the unique characteristics of drinking networks within and between various job categories. Our research suggests that these multiple, complex networks have unique risk characteristics, and that self-selection is based on factors such as job position and college attendance, among other factors.

INTRODUCTION

Recent national surveys have found that foodservice workers have among the highest rates of heavy alcohol use of any US occupation (Frone, 2006; Substance Abuse and Mental Health Services Administration, 2007). The National Survey on Drug Use and Health data for example, show that 12.0% of food service workers engage in heavy drinking (i.e., drinking five or more drinks per typical drinking occasion at least once a week), as compared to 8.8% of U.S. workers overall (Substance Abuse and Mental Health Services Administration, 2007). Because over 12 million workers are employed in the food service industry nationwide, heavy drinking among this population represents significant labor and public health concerns. Studies have shown or suggested relationships between employee drinking patterns and accidents and injuries, both on and off the job (Frone, 2008), as well as diminished productivity (Blum, Roman, & Martin, 1993; Cherpitel, 2007). An analysis of a survey of factory workers, for example, reported that heavy drinkers were significantly more likely than other workers to report arguments with supervisor and problems doing their job. Additionally, on average, heavy drinkers reported a greater number of work-related problems than those who were not heavy drinkers, including sickness, accidents, multiple grievance filings, arguments or fights, disciplinary layoffs, sleeping on the job and lost benefits (Ames, Grube, & Moore, 1997). Finally, sustained heavy drinking can result in a variety of health related problems, including liver and heart disease.

Research has identified several explanatory factors for the over-representation of problem drinkers in restaurant jobs. First, individuals at risk for alcohol dependency may self-select into the restaurant industry (Frone, 2003: Kjaerheim, Mykletun, & Haldorsen, 1996; Macdonald, Wells, & Wild, 1999). Second, young adults—whose alcohol use tends to be higher than those in the general population—constitute a sizable percentage of restaurant workers (Olkinuora, 1984). A third explanation is that alcohol availability in establishments where alcohol is served likewise may facilitate problematic rates of consumption among employees, and contribute to the self-selection of heavy drinkers into the industry (Frone, 2009). However, even workers who are employed in restaurants where no alcohol is served (e.g., fast food restaurants) or in establishments with strict policies against drinking on the job (e.g., corporate–owned restaurants and franchises) report high rates of alcohol use (Michailids & E-ali Elwkai, 2003; Moore, Cunradi, Duke, & Ames, 2009; Moore, Ames, Cunradi, & Duke, 2012). This suggests that findings on how social norms encourage heavy drinking in other occupational categories (Ames, Grube, & Moore, 2000) may also explain why rates of heavy drinking are higher among food service workers than among laborers in other occupations.

Given the preponderance of heavy drinking in the restaurant industry, workplace interventions may greatly benefit these laborers. Because these workplaces are so diverse, however, one-size-fits-all approaches are unlikely to prove effective. For example, as mentioned above, restaurants vary greatly both in terms of availability (i.e., whether or not they serve alcohol), propensity to hire young adults, and the extent to which they have a stated and enforceable alcohol policy. Moreover, because a large portion of restaurant workers lack employer-subsidized health insurance (Employee Benefit Research Institute, 2012), their ability to access substance abuse treatment programs may be limited.

Even within the same restaurant, different job responsibilities and work shifts may be associated with different drinking styles. Workers are sorted into particular positions and shifts based on factors such as circumstances and aspirations outside of work (e.g., supporting dependent children or attending college), and ethnicity (e.g., limited facility with standard English) among other considerations. These differences may result in divergent substance use patterns, particularly in terms of drinking network composition and the likelihood of drinking with co-workers outside of work. Conversely, drinking patterns within the same shift and job category may differ, depending upon factors such as responsibilities outside of work (e.g., as parents, students, employees at another job).

Recent research has examined the role played by social capital in shaping the group characteristics and substance using behavior of drinking networks, particularly among young people (Buettner & Debies-Carl, 2012; Theall, DeJong, Scribner, Mason, Schneider, & Simonsen, 2009; Weitzman & Chen, 2005). Social capital refers to the nature and extent of structural linkages that exist both between and within social networks (Halpern, 2005; Putnam, 2000). These internal and external linkages are salient to the issue of drinking networks, including those of restaurant workers, because such networks constitute dynamic systems in terms of their group composition, strength of emotional bonds, and drinking behaviors, among other characteristics. Moreover, particularly for young adults, drinking networks do not exist as isolates; rather, members may participate in more than one network at various points in time, and may self-select via their common association with other social or institutional domains (e.g., workplaces, fraternities).

In his seminal work on the topic, Robert Putnam distinguishes between two types of social capital that are salient to this discussion: bonding networks, which create connections between people sharing similar characteristics, and bridging networks, which establish linkages between disparate groups of people (Putnam, 2000; cf., Buettner & Debies-Carl, 2012). Given the heterogeneity of restaurant workers in terms of race, ethnicity, social class, job responsibilities, and physical location within the work site, a consideration of their drinking networks—including the extent to which they reflect and reinforce social networks on the job—provides a useful lens through which to examine group composition, dynamics, and drinking behaviors.

In this paper, therefore, we explore the drinking patterns of different categories of workers at an alcohol-serving national restaurant chain. Based on qualitative interviews with a nationwide sample of employees who work different shifts and who occupy a variety of job categories, we seek to understand the relationship between job duties and drinking patterns, and by extension drinking networks and social capital. Given worker diversity in terms of age, ethnicity, and occupational aspirations (particularly related to college enrollment), we also seek to identify differences within job categories. Finally, given managers’ understandings of the work stressors and employee characteristics associated with particular jobs within the restaurant, and because they would likely serve as the broker for any subsequent workplace intervention, we examine these supervisors’ perceptions of drinking networks and heavy alcohol use by their employees.

METHOD

This research is part of a larger mixed-method study on alcohol use among workers in a national restaurant chain in which alcohol is served. As in most mixed methods studies, one of the objectives of the qualitative component is to provide context to the survey findings. In the current study, therefore, we provide contextual factors surrounding alcohol use among different categories of workers that may be masked by the quantitative data. Moreover, while the survey data were primarily concerned with individual-level drinking behaviors and consequences, the qualitative interviews focused largely on respondents’ descriptions of, and opinions about, their drinking networks

Although the survey methods and findings are reported in detail elsewhere (Moore, et al., 2009), it is worth noting that our survey sample consisted of 1,294 restaurant workers between the ages of 18 and 29 who were contacted by professionally trained telephone survey interviewers from an employee roster consisting of 4,999 names. Alcohol-related problems were measured with the Alcohol Use Disorders Identification Test (AUDIT; Babor, Higgins-Biddle, Saunders, & Monteiro, 2001), a 10-item instrument originally developed as a brief screener to identify individuals hazardous alcohol consumption, including alcohol dependence, as well as less severe alcohol problems. Our findings in the survey offered evidence of extremely high rates of alcohol misuse among our sample of young adult restaurant workers. Hazardous alcohol consumption patterns were seen in 80% of men and 64% of women. Multivariate analysis showed that workers who were male and were current smokers were more likely to engage in different dimensions of problem drinking as measured by the AUDIT. Additionally, compared to hosts/greeters, those who were categorized as servers/bartenders were at greater risk for hazardous consumption and alcohol dependence (Moore, et al., 2009).

Respondents for the qualitative interview component of the study were chosen randomly from employee lists provided by twenty restaurants in the Midwestern, Southern, and Northeastern U.S. Specifically, we asked to gain access to four restaurants from each region, including at least one urban, one rural, and one suburban location. From employee lists faxed from each restaurant, we randomly selected one of each of the following class of workers: bartenders, servers, kitchen workers and hosts/bussers. Some respondents worked in more than one job category, including servers periodically working as bartenders. However, for the sake of consistency we assigned them to the job category that the company’s human resources department had assigned to them. As few of the hosts/bussers in the first two regions reported that they drank alcohol (nearly all were high school students), we altered our sampling frame to exclude these workers in the remaining interviews. In their place, we interviewed managers and managers in training, in order to capture their perceptions of the drinking behaviors of their employees, and to provide a historical perspective on potential changes in workplace policy and culture as these pertain to alcohol use. Given the high rates of turnover in the restaurant industry, recruitment for the survey and qualitative components of the study were carried out independently, with most of the qualitative interviews administered following the completion of the survey component. As a result, we did not know whether qualitative interview participants also completed the survey, nor did we attempt to link these data.

In all, 76 workers participated in a one hour semi-structured, open-ended interview. We offered $25 as a respondent fee incentive for each interview. The interview guide included questions regarding participants’ social networks, work and non-work related drinking activities, and participants’ theories concerning the high rates of problem drinking among restaurant workers. The study protocol was approved by the Institutional Review Board of the Pacific Institute for Research and Evaluation.

Interviews were audio-recorded and transcribed verbatim. The research team developed the coding manual. Once the manual was developed, all transcriptions were coded by the researchers using a thematic analysis approach (cf., Miles & Huberman, 1994) via the ATLAS.ti software package, Version 5.0 (Muhr, 2006). We utilized standard procedures for ensuring consistency, including parallel coding of initial transcripts until coding agreement was achieved, and periodic audits of subsequent coded transcripts. Recurring themes connected to heavy after-work drinking were identified and are illustrated in the following section. As rates of drinking before or during work were very low, the current study focuses on drinking behaviors that occur either after work or on employees’ days off.

FINDINGS

SETTING

Research participants consisted of employees of a large U.S. casual dining restaurant chain. Casual dining is an industry term for chain restaurants sharing certain characteristics. These include the utilization of wait staff and bussers, the serving of alcohol, a menu consisting of relatively simple foods, and a self-consciously casual atmosphere. More important to this study, chain restaurants tend to have written employee policies regarding drinking before, during, and/or after work. In this regard, they are distinct from independent restaurants, which rarely have such written policies in place. Examples of casual dining establishments in the United States include TGI Friday's^®, Applebee’s^®, Chili’s^®, Olive Garden^®, and Red Lobster^®.

Like most restaurants that have wait service, workers in casual dining establishments are organized into two broad categories of laborers: those who work in the dining room and interact with patrons (e.g., servers, bartenders, bussers), and those who work in the kitchen (e.g., cooks, dishwashers). The reification of these spatial categories is manifested by the fact that “front-of-the-house” and “back-of-the-house” are commonly used shorthand terms both for particular clusters of work responsibilities and the laborers who carry them out.

Restaurant executives from corporate headquarters believed that kitchen workers were at lower risk for problem drinking than their counterparts in the front-of-the-house. As one noted about kitchen workers, “many will work a shift with us and then go to another restaurant to work another shift. Yeah, it’s amazing, they work their tails off…They don’t have time to drink.” However, workers in the front and the back-of-the-house described relatively distinct drinking patterns. These differences are described below.

FRONT-OF-THE-HOUSE: SERVERS AND BARTENDERS

Within each restaurant we studied, employees often used the term “family” to refer to their co-workers, and to the overall social environment of the restaurant more generally. One server, who also occasionally worked as a bartender, compared the informal nature of her workplace relative to those of other restaurants:

Everybody goofs around all the time, half the co-workers there are full-time and they’re always there, it’s like a big family. So everybody knows everything about where everybody hangs out, and it’s like a party place. (At other restaurants), they’re like right by the book, no laughing, they don’t have any fun, which is awful.

However, these familial sentiments in the workplace did not typically translate to after-work socializing between the front-of-the-house and the kitchen staff. Although neither group spoke disparagingly of the other, their different duties, the respective spheres in which they are carried out, and to some degree ethnic and language differences, limited their ability to develop relationships with one another. For example, servers would rarely visit the kitchen during busy work times unless there was a problem with a food order. More often, they would communicate with the Expeditor (commonly referred to as the Expo), the one worker whose primary responsibility is to serve as the point of contact between dining room and the back-of-the-house. Conversely, kitchen workers would rarely venture to the dining area when patrons were present. When asked whether dining room and kitchen staff fraternized together after work, a kitchen worker remarked that this would happen “once in awhile, I mean you got people that hang out together, but for the most part it’s the back-of-the-house and the front-of-the-house. It’s pretty separate.” In terms of social capital, therefore, workers in the front and back-of-the-house participated in relatively distinct bonding networks, with little overlap between their respective groups.

Interestingly, servers included bartenders in their after-work socializing, despite the fact that the latter’s duties are at least superficially like those of cooks; filling orders for drinks in the same way that a cook would prepare food orders. However, unlike interactions between servers and cooks, which are mediated by the Expo, servers and bartenders directly engage with one another in the process of ordering, preparing, retrieving, and serving drinks. Furthermore, two common aspects of their work may contribute to the close relations between bartenders and servers: both rely heavily on gratuities and have extensive contact with patrons. This may explain why, apart from their being younger, hosts/ bussers are not typically included in social activities with the other dining room staff, since they do not receive tips and their interaction with customers is limited to leading them to their tables.

An important way in which workers in the front-of-the-house solidify their relationships with one another is to spend time with each other socially after work. Socializing with co-workers frequently involved drinking together. Workers frequently spoke of participation in drinking networks as a way to build camaraderie and to unwind after work. Here, a manager describes some of the factors that may result in employees drinking together after their shift:

A majority of the drinking has to do with the fact that there’s not really too much to do (in this community). The hours worked, I think has a lot to do with that. And you feel (a sense of) belonging, part of the family. Because like in my restaurant, most of them are college kids, they’re away from their family. So their friends and the people they work with are their surrogate family.

After-work drinking among dining room staff often consisted of going to another restaurant or bar to drink. At the beginning of our research, the company maintained a strict policy against employees drinking alcohol in their own restaurant after their shift, as it was seen as a distraction to co-workers who were still on the clock, and to customers who may be confused by the worker’s transformation from server to patron. Towards the end of our fieldwork, however, the company had changed its policy to one which actively encouraged workers and their friends to patronize their establishments, primarily as a way to generate additional bar revenue. Most workers, nonetheless, preferred to go to a different drinking establishment with co-workers or workers from nearby restaurants, in part because they had already spent many hours at their restaurant, and because the company did not offer them discounts on drinks.

A number of factors contribute to heavy drinking when servers and bartenders go out together after work. First, because most of their pay derives from gratuities, these workers often leave the restaurant with a sizeable quantity of cash in their pocket. Many of the workers we interviewed felt that having that much cash on hand made it difficult to exercise caution when drinking, as described by this bartender:

There’s a lot of high school and a lot of college kids that (work here), they do this to help support themselves through college, and college kids drink a lot, you know? They get off work and they’re like ‘Hey I made like a hundred dollars. Let’s go drink half of it.’ You know, they get excited with all that money in their pocket. They don’t know what to do with it. You get all worked up during work and you’re all fired up and ready to do something, so they go have a couple of drinks after work. And there are some people I know, it turns into fifteen to twenty.

In addition, these workers are part of a broader network of servers and bartenders from the area, many of whom know one another. As a result, workers would often receive substantial discounts on alcohol when they patronized nearby restaurants and bars. Speaking about a neighboring drinking establishment frequented by restaurant staff, a respondent who worked as a server and bartender said, “We know a lot of the bartenders (there), we used to go there a lot. He used to hook us up all the time. By the end of the night we’d end up drinking, three hours of drinking, our bill would be like twenty bucks.”

For bartenders in particular, patronizing and drinking heavily at other establishments has the additional benefit of promoting the restaurant where they work in general, and their bartending in particular. Employee regulations prohibited bartenders from providing free or discounted drinks to patrons, including those from other restaurants. However, some carried company-provided business cards to distribute to workers at other restaurants. Describing his procedure for promoting the restaurant to workers at other establishments, a bartender noted that: “I carry a bartender business card and I'll slip it to (the bartender). Give them a free chips and salsa if they come in and see me.”

As noted above, drinking patterns of front-of-the-house staff are additionally shaped by the large number of college students employed as servers. In contrast, we encountered few students who worked in the kitchen. A substantial body of literature points to college students as being at particular risk for problem drinking (Saltz 2004/2005; Hingson, Wenxing, &Weitzman, 2009). The fact that these students are also employed in an industry with such high rates of problem drinking presents unique and multiple risks. However, different drinking patterns emerged among these student workers. Some, for example, reported that they were not involved in either college- or restaurant-based drinking networks. For these workers, the burdens of working and carrying a heavy course load made late night socializing difficult, except during school breaks. It is worth noting, however, that many of those in this category were commuter students, and lived either with their parents or with a romantic partner.

A small number of college student workers largely eschewed fraternizing with co-workers, preferring instead to spend time with their fellow students. The most common pattern, however, consisted of workers who engaged with both school-based and work-based drinking networks. That is, these workers would drink with co-workers after their shift, while drinking with fellow students on their days off. Thus, depending on the location of an employee’s restaurant relative to their campus, college student workers may only fraternize after work on weekends; otherwise, these workers either go home to study or participate in college drinking cliques. Finally, these largely overlapping networks engage in heavy drinking, and a college student who is an employee—particularly if he or she lives on campus—has multiple opportunities to participate in each. As a college student working as a server noted, “As far as school goes, there’s always a party.” It is worth noting, however, that while college students could participate in both work- and school-related drinking networks, we found little evidence of non-college attending front-of –the-house workers participating in college drinking networks. This is likely because college-attending employees are embedded within both work and school-based social networks. However, non-college attending workers’ exclusion from college-based drinking attendance may also be due to aspirational class distinctions between student employees and their non-degreed colleagues. As discussed below, these distinctions are manifested in the category of “lifer”, a somewhat denigrating term used by managers and front-of-the-house staff to describe workers who have few other career options than to work at the restaurant.

KITCHEN WORKERS

Kitchen workers tended to be more ethnically diverse than participants who worked in the front-of-the-house. While the majority of our total sample was Euro-American, regardless of position, kitchen workers included a somewhat larger portion of African American and Latinos than did those in the front-of-the-house. These demographic characteristics may account for certain differences in drinking patterns, as consumption practices among these groups tend to differ from their white counterparts. More central to the current discussion, however, are three factors that emerged from the qualitative data regarding the distinct drinking patterns of kitchen workers relative to those in the front-of-the-house: a) differences in the way these workers are paid; b) financial concerns; and c) educational attainment, and its association with social class.

Kitchen workers are paid much like those outside of the restaurant industry, biweekly in the form of a paycheck, which is typically direct-deposited in the workers’ bank account. This form of remuneration contrasts with that of front-of-the-house staff, whose sub-minimum wage pay is heavily subsidized by gratuities. As discussed above, gratuities play an important role in shaping the drinking patterns of dining room workers.

Although kitchen workers have the benefit of receiving a predictable level of pay for the number of hours they work rather than being dependent on the unpredictability of tips, these workers’ wages are decidedly modest. Many, in fact, held second jobs, as often as not at another restaurant. Accordingly, kitchen workers’ drinking networks consisted of co-workers from their various jobs, and friends from other aspects of their lives. Although some reported drinking after work with co-workers, drinking in bars on these occasions was relatively uncommon, as these social events were perceived as expensive. More typical is the pattern described by a cook: “You make friends at work. I have people that come to my house, you know. They’ll give me rides home and they’ll come over and hang out for a little while. Talk and socialize. And I have friends that I hang out with. We’ll go out from work.” Interestingly, although both classes of workers reported getting along with one another, there seemed to be relatively little overlap between their drinking networks, whether at parties or at drinking establishments. As this same kitchen worker added, “I socialize more with the back than the front-of-the-house.”

Some employees attributed the relatively distinct social networks of kitchen and dining room staff with the larger number of monolingual Spanish speakers working in their kitchens. For example, in describing the two classes of workers, one Latino kitchen worker said:

Yeah, they’re kind of separate, really. We had an Expo a few years ago that was an American girl who spoke perfect Spanish, and she was friends with a lot of the back house people. But I think that language barrier is kind of holding that up a little bit, and also there’s some difference culturally.

However, even in the many restaurants we visited where kitchen workers spoke English and were of similar ethnic composition to their dining area counterparts, the two sets of workers rarely interacted with one another. Some of the separation between social networks in these instances may be attributable to the different life circumstances of these workers. For example, as mentioned above, we found a preponderance of college students in the front-of-the-house of many restaurants while back-of-the-house workers tended to be self-supporting and not attending school. Finally, some noted that the scope of work in these two spatial domains is best suited to those having different temperaments, as described by a cook: “It takes a certain type of personality to work out front and a certain type of personality to work on the line to deal with the stresses. I’ve had servers say they couldn’t do it, I mean on the line. Me, I can’t be a server. So I guess the personalities are slightly different.”

Regardless of linguistic, ethnic, or perceived personality differences, kitchen workers’ drinking patterns varied substantially from those at the front-of-the-house. For example, because they tended to live paycheck to paycheck, these workers invariably cited financial constraints as the reason for not frequenting bars like their counterparts who work in the dining area. As the survey data show, however, economic challenges did not result in reduced drinking-related problems. Rather, a common practice among these workers is to drink at parties, either in their own home or in the home of friends. These drinking occasions often involve a substantial quantity of alcohol, which can in turn affect the work environment, as described by this kitchen worker:

One of the cooks came in (to work) about two hours late. He got really drunk the night before, and didn’t remember anything that happened. His cell phone was broken and someone broke the windshield of his car. And he had no idea how it happened. So, he was two hours late for work because he couldn’t find his phone and couldn’t figure out what happened to his car. He had a miserable day. He felt like crap all day.

MANAGERS’ RESPONSES TO EMPLOYEE DRINKING

Most of the managers we interviewed had either spent many years working in other restaurants or had worked their way up to a management position. Many, in fact, initially began working for the company as a server while they were in college, and only subsequently decided on a career in the food and beverage industry.

Managers tended to have little direct knowledge about the current drinking behaviors of their employees. As mentioned above, the company had a strict policy against drinking on the job, and we discovered very few infractions of these rules in either the survey or qualitative findings. Likewise, we heard of few cases of employees coming to work while intoxicated. In addition, because the company had a strict policy against managers fraternizing with employees outside of work, these supervisors did not participate in after-work drinking sessions. This was difficult for some managers who had recently been promoted, as they could no longer socialize with those who had been their co-workers:

Right now I call it pure hell for me. I get (invited out by co-workers) constantly, especially since I’m not their age, but because I don’t have a family and I’m single, they feel obliged to ask me to come out all the time and do everything else. But I always have to come up with excuses like, ‘No, I can’t, I gotta work early in the morning.’ ‘No, I can’t, I gotta go meet with my roommate.’ It’s difficult.

The only time that drinking outside of work becomes a problem is when employees come into work with a hangover, a common occurrence according to the managers we interviewed:

I don’t necessarily chastise them about (coming in hungover), but I let them know that they’re coming in like that. I’ve had a couple that hadn’t even slept, and they came in. You send them home. Because obviously if they do reek of, y’know, you can still smell it! ‘You need to go home now.’ They’re not in a good mood, either. I think it’s (that) they’re sick, they need to go home to get some rest.

As implied by the above statement, given the high turnover rates in the restaurant industry overall, and the difficulties of bringing in a replacement worker at the last minute, managers by and large chose to overlook these impairments, as long as the employee was able to carry out his assigned duties.

Like the company’s executives, the managers tended to view kitchen workers as having too many responsibilities in terms of families and second jobs to drink heavily. As most managers had worked at the front-of-the-house before being promoted, they tended to be much more knowledgeable about the overall drinking patterns of those who work in the dining room. Managers largely corroborated our findings in terms of social drinking as a mechanism for social bonding and winding down after a shift, and the role of gratuities in over-spending on alcohol.

Although managers recognized that front-of-the-house employees engaged in heavy drinking, they tended to judge drinking by college student workers differently from those of “lifers”, employees who were not in school and who had worked in the restaurant for an extended period of time. In particular, they tended to view students’ drinking as normative, perhaps because they too had been to college: “It seems like people who go to university get (drinking) out of their skin, and then move on with their careers and lives.” Conversely, although managers may technically be considered lifers themselves (due to their relatively extensive period of time working for the company), they viewed drinking among this category of workers as emblematic of their lack of education and ambition. For example, when asked about why he thought that heavy drinking was so prevalent in the restaurant industry, a manager trainee observed: “We’ve got a lot of good people here, but they haven’t gone to school, they haven’t spent four or five years in college and gotten a degree. So, I don’t want to say that this is the lower half of the totem pole, but that may be some part of it.”

CONCLUSION AND IMPLICATIONS

In this paper, we have described some of the distinct drinking behaviors between dining room staff and those in the back-of-the-house, particularly in terms of their discrete drinking networks. These findings are summarized in Table 1. From a social capital perspective, these drinking cliques act as bonding networks, providing group fellowship and solidarity, but at the expense of excluding those who work in a different part of the restaurant. The formation and maintenance of these distinct social groupings, it should be noted, are not merely a function of one’s job title per se; rather, these networks remain distinct because workers’ occupational position at the front- or back-of-the-house tends to be associated with particular racial-, ethnic- or class-based characteristics, the latter manifested in terms of college attendance.

Table 1.

Factors Affecting Drinking Patterns of Front and Back-of-the-house Employees

	Front-of-the-house	Back-of-the-house
Pay	Hourly salary + tips	Hourly salary
Ethnicity (Language)	Predominantly Anglo-American	Multi-ethnic (English, Spanish)
Currently Attending College	Many	Few
Self-Supporting	Few	Many
Principal Drinking Locales	Bars, restaurants	Parties, home

Given that restaurant workers are at particularly high risk for engaging in problematic drinking, there is a critical need for interventions that target these workers. However, these interventions may not be effective unless they take workers’ job duties—with their concomitantly divergent drinking networks—into account. For example, interventions targeting kitchen workers could focus on avoiding peer pressure to drink heavily at parties. Interventions targeting front-of-the-house workers, in contrast, could encourage workers to deposit their tips into their bank account immediately after their shift, to avoid the temptation of over-spending on alcohol.

Any successful intervention will require the cooperation of managers. However, their different responses to drinking by college students and lifers may place barriers on the success of that intervention. Managers’, and to some degree co-workers’, negative perceptions of those who choose to remain in the restaurant industry clearly reflects the overall de-skilling of labor in chain restaurants in particular, and the de-valuing of occupations requiring manual (cooks, dishwashers) or emotional (servers) labor more generally. However, managers’ perceptions of college students’ drinking as normative and lifers’ as deviant diminish the risky behavior in which both groups engage. Encouraging managers to reflect on these biases, therefore, would likely facilitate the effectiveness of any interventions targeting problem drinking among restaurant workers. Perhaps more important, it may encourage them to reflect upon the dignity of all the workers they supervise, and the skills that they bring to the job.

This study has a number of limitations. First, the study population consisted of employees of a chain restaurant with a relatively well-defined drinking policy. Small, independent restaurants rarely have such codified policies, so our findings may understate the role of drinking on the job as a factor in alcohol-related problems of restaurant workers overall. Second, the qualitative data consisted nearly exclusively of interviews. Although we observed workers on the job, we did not have the opportunity to witness their drinking sessions, so our findings are dependent on self-report. As the high rates of drinking-related problems in our survey data can attest, our qualitative data may thus understate both the prevalence and outcomes of problem drinking among the sample. Nonetheless, the findings provide important insights into the social and structural factors that influence network drinking patterns between distinct classes of restaurant workers, from which effective interventions can be developed.