The Ethnographic Contribution to Understanding Co-worker Relations Randy Hodson Abstract Relations among co-workers are becoming both more important and more complex in modern workplaces as authority over job decisions is shifted from supervisors to quasi-independent teams. The author develops a model of co-worker relations that recognizes these changes and evaluates this model using data content coded from the full population of published book-length workplace ethnographies (N = 204). Confirmatory factor analysis techniques support the existence of three distinct aspects of co-worker relations: cohesive- ness, conflict and peer supervision. The most important determinants of co-worker relations are employee involvement programmes and management behaviour. Returning to specific case studies allows a theoretical elaboration of how employee involvement and management behaviour condition co-worker relations. The author concludes by noting the importance of intellectual exchanges between qualitative and quantitative methods for generating new advances in the study of work and employment relations. 1. Introduction Co-worker relations have long been a core building block in understanding the workplace and workplace relations (Blauner 1964 Crozier 1971 Homans 1950 Pollert 1981 Roy 1954). But such studies were significantly displaced in the latter half of the twentieth century by studies of technology, socio- technical relations and management behaviour. At the beginning of the twenty-first century, co-worker relations are again receiving increased scru- tiny in workplace studies. Researchers have identified at least three recent developments that underwrite this increased attention (Rubery and Grimshaw 2003). First, increasing global competition has necessitated new strategies to increase productivity and reduce costs. The development of self-monitoring teams is one response to these pressures. Second, increasing workforce education allows the assignment of more and more supervisory Randy Hodson is at the Ohio State University. British Journal of Industrial Relations doi: 10.1111/j.1467-8543.2007.00670.x 46:1 March 2008 0007���1080 pp. 169���192 �� Blackwell Publishing Ltd/London School of Economics 2008. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

duties to front-line workers. Third, the explosion of microchip technologies has allowed increasingly flexible production procedures. And self-monitoring teams have been identified as one mechanism for maximizing flexibility and avoiding bureaucratic rigidity under flexible production (Batt 2001). These developments have combined to increase the complexity of co-worker rela- tions and to increase their importance for organizational success. Given this heightened focus on co-worker relations, it is useful to review and integrate what we know about co-worker relations and with a view towards understanding contemporary changes. Several questions emerge in this process. First, what are the key dimensions of co-worker relations? Second, what are the major influences on these relations? Third, what is the role of management in the new and more socially active workplace? And what can management do that is supportive of positive workgroup relations? This article is devoted to providing preliminary answers to these questions. We turn first to co-worker relations and the role of management behaviour in shaping these relations and then to a consideration of other organizational influences on co-worker relations. The empirical data we use to address these questions are the accumulated body of workplace ethnography. I analyse these case studies using both quantitative and qualitative metho- dologies in order to realize the benefits of both traditions. In this process, I first temporarily decontextualized the ethnographic data by content coding it and then recontextualized it in the process of theoretically exploring specific cases. 2. Co-worker relations Three dimensions of co-worker relations emerge from the workplace and employment relations literatures as potentially fruitful avenues for investiga- tion: cohesiveness, conflict and peer supervision. First, through group cohe- siveness and solidarity co-workers can provide both instrumental and affective social support. The study of workgroup cohesion was a mainstay of early organizational studies (Homans 1950 Pollert 1981 Sayles 1958). Although the study of group cohesion has been partly displaced by studies of management styles in recent decades, it still remains a key building block of the study of employment relations and is experiencing revitalization with the increased focus on teams (Z��rraga and Bonache 2005). Second, co-worker relations can be problematic, and can be a source of conflict, tension and bullying in the workplace (Bartel and Saavedra 2000 Jehn 1995). Co-worker conflict has been an occasional topic of study in the workplace, but this focus has never rivalled that of group cohesion. Reports from workplace ethnographies, however, suggest that infighting and conflict are at least as important to the daily experience of work as the more com- monly studied themes of cohesion and solidarity (Jackall 1978 Sims 2005). Third, co-workers frequently monitor, formally or informally, the behav- iour and productivity of fellow employees ��� in essence they engage in peer 170 British Journal of Industrial Relations �� Blackwell Publishing Ltd/London School of Economics 2008.

supervision (Barker 1999 Sewell 1998). This phenomenon has become increasingly common with the spread of team forms of production (Appel- baum and Batt 1994) and represents a potentially important emerging com- ponent of co-worker relations in the contemporary economy (Ezzamel and Willmott 1998). I discuss these three aspects of co-worker relations in greater detail in the following sections. Cohesiveness Cohesiveness can be defined as the extent to which workers seek each other out for social contact and support (Pollert 1981 Sayles 1958). Group cohe- siveness and solidarity can mitigate feelings of alienation that arise from meaningless work. Group cohesiveness has often been difficult to measure using conventional survey techniques because cohesiveness is a group property and such phenomena are difficult to explore through face-to-face interviews or telephone surveys that entail an inherent methodological indi- vidualism. Because cohesiveness is a characteristic of groups, field methods have been particularly important in its study (Van Maanen 1998). Cohesiveness at the workplace includes a willingness of co-workers to defend each other in the face of challenges, most generally from management, but also sometimes from other groups of workers or from customers (Tsui and Gutek 1999). Workgroup cohesion is manifest across a range of behav- iours in the workplace including strong group boundaries, group leadership, mutual defence, and social activities and friendships at work (Mannon 1991). An example of strong friendships emerging in a cohesive workplace is pro- vided by an ethnography of an electronics manufacturing factory: Individuals frequently socialize with each other throughout the day about work- related and other matters, including favorite television programs, romantic involvements, and weekend activities. . . . [Even those recently hired] note how easily and quickly they formed close ties with colleagues, who often have shared interests that extend beyond the job. . . . David, a sales representative, claimed that he and his coworkers are like ���brothers and sisters���. (Tucker 1999: 22, 33���34) Co-worker Conflict Co-worker conflict can be defined as a pattern of chronic conflict, destructive gossip and even interference either within or between workgroups. Relations between co-workers are not always positive and chronic conflicts can become a part of everyday life in some workplaces. Antisocial behaviour at the workplace can include such behaviours as ���criticizing people at work���, ���start- ing an argument with someone at work��� and ���griping about co-workers��� (Robinson and O���Leary-Kelly 1998: 663). The incidence of workplace con- flict may actually be on the increase due to the shift towards more participa- tive workplace structures with more dense interactional patterns (Jehn 1995). Indeed, the association of group conflict with increased group interaction has Ethnographic Contributions to Understanding Co-worker Relations 171 �� Blackwell Publishing Ltd/London School of Economics 2008.

led to the suggestion that conflict and cohesion may be fellow travellers on the road to a more participative workplace (Barker 1999). Gossip between workgroups is a particularly common avenue through which co-worker conflicts are vented (Noon and Delbridge 1993 Tebbutt and Marchington 1997) and is evidenced in an ethnography of temporary clerical workers: Not only were individual temps the butt of contemptuous and often bitchy remarks made by the permanents, but the whole group of us were sometimes subject to generalised insults and cool behaviour. Temps were often discussed by the perma- nent staff. This usually happened when all the temps had left at 4.30 p.m. (McNally 1979: 169) Active interference is less common than gossip, which is a safer activity because it does not so blatantly violate workplace norms but interference is not unknown in the workplace as the following episode from an ethnography of a slaughterhouse illustrates: Owen did not like me any better the second week. One morning when my whizard [electric knife] started to vibrate violently so that I couldn���t cut with it, he eventu- ally told me to tighten a screw to adjust it, but mostly he was silently hostile. At one point he threw several uncut pieces of teat line in the tub where I threw my finished pieces, which would have put me in trouble if I hadn���t seen them and dug them out. (Fink 1998: 24) Peer Supervision A third aspect of co-worker relations, and one potentially growing in impor- tance, is peer supervision (Guest and Peccei 2001). Peer supervision can be defined as the regular participation by co-workers in supervising, training and enforcing either formal or informal work rules on their fellow employees. Peer supervision is not a new invention in the workplace. Workgroups, especially ones whose tasks are interdependent, have long been involved in peer training and discipline. Such activities have a long tradition in craft settings, for example. Informal work rules can entail both keeping up with standards and not exceeding these standards (Roy 1954 Webb and Palmer 1998). Peer training and discipline often first emerge in the process of social- izing new workers to appropriate job performance, as the following example from an ethnography of ironworkers illustrates: He did everything wrong, of course. . . . He was forever on the wrong side of the pieces, and it still amazes me that he wasn���t knocked over the side. . . . Jiggs continually screamed at him to be careful. Several times I pulled and held pieces away from him and told him I wasn���t going to let him have his end unless he stopped bouncing around. (Cherry 1974: 186) In management-initiated teams, training is often explicitly coupled with surveillance (Delbridge 1998). Thus, under modern team organizations of 172 British Journal of Industrial Relations �� Blackwell Publishing Ltd/London School of Economics 2008.

production, increased employee involvement can ironically mean greater autonomy from direct managerial supervision but, simultaneously, greater peer pressure and closer overall surveillance (Marks et al. 1998). Manage- ment may be less visibly present, but workgroups can be extremely effective at exacting high levels of effort from their members. If workers resist their team���s demands, they risk being made to feel unworthy as team members. Criticism and ostracism by one���s peers are powerful forms of social control. Team-based organizations of work can thus provide the basis for an even tighter control of work life than management systems based on bureaucratic rules or supervisory control (Barker 1999). Traditional peer support and training are thus increasingly coupled with increasing team surveillance and pressures for production. Peer supervision is an explicit component of total quality management (TQM) systems, which have achieved widespread popularity across a range of workplaces (Guest and Peccei 2001). In workplaces employing TQM, employees are routinely expected to fill out reports evaluating the perfor- mance of other team members. In this way, responsibilities previously held by management are shifted to the workgroup (Barker 1999 Sewell 1998). Two hypotheses summarize our expectations concerning co-worker relations: Co-worker Hypothesis 1: Co-worker relations are constituted by three separate and distinct facets ��� cohesiveness, conflict and peer supervision. Co-worker Hypothesis 2: All three aspects of co-worker relations are positively related, including conflict. The latter hypothesis follows from the observation that workplaces can be differentiated between socially active and socially isolating environments and the corollary that all facets of co-worker relations, both positive and nega- tive, are likely to be higher in socially active workplaces than in socially isolating workplaces. 3. Management behaviour Management behaviour provides the framework and sets important prece- dents for social relations among co-workers. Trustworthy management behaviour, including respect for implicit contracts about the treatment of employees, is a precondition for productivity and meaning in work (Elger and Smith 1998 Hodson 2005). Trustworthy management behaviours have been found to be important both for organizational citizenship and for minimizing conflict and aggression in the workplace (Adler 2001). What behaviours constitute trustworthy management? Normatively expected behaviour for management involves at least two domains ��� providing technically viable production technologies and protocols and respecting workers and their interests and rights (Rubery and Grimshaw Ethnographic Contributions to Understanding Co-worker Relations 173 �� Blackwell Publishing Ltd/London School of Economics 2008.