The Organizational and Social Foundations of Worker Resistance

by Vincent J. Roscigno, Randy Hodson
Citation
Title:
The Organizational and Social Foundations of Worker Resistance
Author:
Vincent J. Roscigno, Randy Hodson
Year: 
2004
Publication: 
American Sociological Review
Volume: 
69
Issue: 
1
Start Page: 
14
End Page: 
39
Publisher: 
Language: 
English
URL: 
Select license: 
Select License
DOI: 
PMID: 
ISSN: 
Abstract:

The Organizational and Social Foundations of Worker Resistance

Vincent J. Roscigno Randy Hodson

Ohio State University Ohio State University

The study of worker resistance has tended to focus either on organizational attributes that may alter actors 'capacity to respond or on influential shop-floor social relations. This divide, partially driven by analytical and methodological preference, is also a

function of different theoretical traditions. In this article, we suggest that organizational attributes and interpersonal relations in the workplace, in concert with union presence and collective action history, may be simultaneously but also conditionally meaningfiul for workers and their potential resistance strategies. Findings, derived from analyses of unique data on 82 workplace ethnographies and that merge Qualitative Comparative Analysis (QCA) techniques and more conventional quantitative methods, largely support these expectations. Most notably, the impact of workplace organization and even union presence on worker resistance varies depending on social relations on the shopfloor. Where there is union presence and signiJicant interpersonal conflict with supervisors, the likelihood of collective resistance in the form of strike action is heightened. This pattern also holds for certain more individualized forms of worker resistance (i.e., social sabotage, work avoidance, and absenteeism). More central to individual resistance, howevel; are workplace contexts characterized by poor organization and a lack of collective action legacy. We conclude by discussing the implications of our results for

future analyses of workplace social relations, workplace structure, and collective and

individual resistance-oriented actions.

orker resistance to real or perceived short- falls of the contemporary workplace is a commonly occurring phenomenon. Despite some declines in recent decades, collective response in the form of union activity and strike action remains considerable and worthy of atten- tion. Each year in the United States, for instance, more than 200,000 workers take part in work

Direct correspondence to Vincent J. Roscigno, Department of Sociology, Ohio State University, 300 Bricker Hall, 190 N. Oval Mall, Columbus, OH 432 10-1353 (Roscigno. l@osu.edu). The material in this article is based on work supported by the National Science Foundation under grant 01 12434. Any opin- ions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The authors thank Cliff Brown, Marc Dixon, Steven Lopez, and the editors and anonymous reviewers of ASR for helpful comments on earlier drafts of this article.

stoppages and strike action (Bureau of Labor Statistics 2001). Worker resistance can and often does take on individualized forms as well, such as sabotage, theft, and work avoidance-forms that are seldom studied given their subtle and often covert character (Jermier, Knights, and Nord 1994). Such actions, rather than being the mere product of delinquency, often reflect broader grievances about work organization and treatment on the job.

Worker resistance, particularly in its collec- tive form, is patterned by broader societal dynamics pertaining to split labor markets (e.g., Brown and Boswell1995; Bonacich 1972), the extent of elite cohesion (e.g., Jehs and Perrow 1977; Pichardo 1995), legal-political structures (e.g., Burawoy 1985; Low-Beer 1978; McCammon 1993, 1994), and local cultural, kinship, familial and ascriptive attributes (Cornfield and Kim 1994; Lee 1998). Especially important will be the presence of unions, the resources they bring to bear, and collective action legacies (e.g., Kelly and Kelly 1991; Rubin 1996; Rubin, Griffin, and Wallace 1983). Yet, the grievances upon which collective and individual resistance strategies are founded are ultimately forged at the point of production- a fact highlighted in classical theoretical treat- ments (e.g., Marx 1971) and literature on rank and file activism (e.g., Brecher 1972; Kimeldorf

1999; Montgomery 1979; Scott 1985; see also Foucault 1988). Simply, workers' sense of dig- nity and satisfaction are formulated on an ongo- ing basis, and relative to what occurs in their particular workplaces.' Therefore, the work- place and its dynamics are the natural starting point for understanding how, why, and when workers contest.

Questions nevertheless remain about whether grievances and resistance unfold as a function of workplace organization or interpersonal mis- treatment on the shop floor. Are workers more likely to strike in the case of abuse on the job, or are they more likely to protest incoherent workplace organization? Is an individual work- er more likely to engage in sabotage or theft in a bureaucratized, more controlled work envi- ronment, or are such actions more likely to be triggered in the face of ongoing conflict with managers? A fundamental divide in the work- place literature exists on these questions. While differences in conclusions can be partly attrib- uted to the methodological tendencies of orga- nizational versus job-level research, or survey versus ethnographic designs, we believe that theoretical assumptions of each tradition are also involved.

In this article, we build upon and extend existing perspectives on collective and individ- ual resistance, with a specific focus on the work- place as an often contested arena. Our discussion begins with the influence of workplace orga- nization, and then we describe how interpersonal

This is not to suggest, of course, that inequality in other institutional areas (e.g., family, education, etc.) or dynamics outside of particular workplaces (e.g., segregation) may not intrude upon workplace resistance. They certainly may. What occurs in work- places organizationally and socially, however, is more paramount. That is, day-to-day and more proximate workplace organization and social relations will hold the greatest implications for the resistance strategies undertaken in any given workplace.

THE FOUNDATIONSOF WORKER RESISTANCE

dynamics also may be important. Our theoret-

ical discussion delineates how and why the

bifurcated nature of this literature is problem-

atic, and then offers an alternative understand-

ing-one in which the effects of workplace

organization and more proximate social relations

are viewed in a conditional manner, and with-

in the context of union presence and collective

mobilization history. The data, comprising 82

workplace ethnographies, provide detailed infor-

mation on worker-supervisor interpersonal rela-

tions and considerable variation in workplace

organizational structure. Our unique use of

Qualitative Comparative Analysis (QCA) tech-

niques in concert with quantitative methods

allows for the examination of unique configu-

rations of organizational and interpersonal

dynamics that either diminish or exacerbate

collective and individual resistance. Indeed,

findings reveal cross-level configurations that

are conducive to both collective resistance, in

the form of strike participation, and more indi-

vidualized forms, such as sabotage, theft, and

absenteeism. We conclude by discussing our

results, the need for simultaneous examination

of both organizational structures and social rela-

tions in the workplace, and what our findings

suggest relative to prior and future research on

worker well-being and insurgent action.

WORKPLACE ORGANIZATION

Associations between workplace organization, worker well-being, and resistance-oriented action have always been at the core of socio- logical theory (Collins 1981). Marx (1967, 1971), of course, clearly denoted the impact of structural and technological innovations, with the assumption that such transitions would cre- ate conflict and eventual resistance by alienat- ing workers from their creative potentials, from the labor process and product, and from their fel- low workers. Durkheim (1984) and Weber (1968), respectively, similarly noted inherent tensions associated with the "forced division of labour" and the ever-constraining impact of bureaucratization on workers.

It should thus come as no surprise, given these rich, early insights, that the issue of work- place organization and worker response has continued to demand the attention of sociolo- gists, particularly since the emergence of Taylorism and scientific management in the early twentieth century. Indeed, Taylor (191 1) himself recognized that resistance would occur, but suggested that the promise of higher mate- rial rewards and careful selection of employees could mitigate any disruptive potential. Take, for instance, his now classic (and quite derogatory) discussion of introducing specialization into pig iron work, and the importance of selecting the right employees:

Now one of the very first requirements for a man who is fit to handle a pig iron as a regular occu- pation is that he shall be so stupid and so phleg- matic that he more nearly resembles in his mental make-up the ox than any other type. The man who is mentally alert and intelligent is for this very reason entirely unsuited to what would, for him, be the grinding monotony of work of this character. Therefore the workman who is best suited to han- dling the pig iron is unable to understand the real science of doing this class of work. He is so stu- pid that the word "percentage" has no meaning to him, and he must consequently be trained by a man more intelligent than himself into the habit of working in accordance with the laws of this science before he can be successful. (Taylor 19 11:41)

Unlike Taylor, more contemporary social sci- entists have been more critical of workplace organizational transformation by highlighting the consequences for workers themselves. Dunlop (1958) and Blauner (1964), for instance, were concerned with resulting declines in work- er autonomy and a shift in power from workers to managers and supervisors. Braverman (1974) was worried about the deskilling of workers via specialization an4 consequently, the dissection of mental and physical labor. R. Edwards (1 979) and Burawoy (1979, 1985) note that increasing bureaucratic control of workers over time not only may depress levels of worker power and satisfaction, but also may result in significant conflict on the shop floor. More likely, howev- er, bureaucratic workplaces with pronounced organization and divisions of labor will mitigate the potential for class consciousness and action by constraining workers to their job tasks and by channeling broader grievances and conflicts through organizational channels (Wright 2002).2

One critique that can be made of the organiza- tional literature is that it tends to treat resistance in strictly classic terms, as solely collective in nature, and often driven by class polarization and class iden- tity among workers. While we do not disagree that such manifestations reflect class resistance, they are but one form. Indeed, important theorizing on the

Along with bureaucratization, a second crucial characteristic of organizations has impli- cations for worker well-being and resistance- an organization's ability to maintain coherent and integrated production activities (Barnard 1950; Dunlop 1958; Roethlisberger and Dickson 1939). Organizational norms specify that man- agement is responsible for maintaining a coher- ent and effective system of production that allows employees to meet the demands of their jobs (Whitener et al. 1998). Coherent organi- zational procedures are essential for organiza- tional effectiveness (Bass 1985), the maintenance of management legitimacy (Della Fave 1980), and an organizational climate in which confidence and good will can replace rigid or coercive systems of coordination, inspection, and evaluation (Burawoy 1985; Pfeffer 1998). Without effective organization, workplaces become characterized by chaos and abuse (Juravich 1985).

Students of organizations suggest that a wide range of positive consequences follow from organizational coherence and integration (Barker 1999; Moore 1962; Smith 2001). These include increased citizenship on the part of employees and the creation of a more cooper- ative and less conflictual workplace (Pfeffer 1998). We thus expect organizational coher- ence to play a prominent role in mitigating both formal and informal worker resistance. In effect, organizational coherence and integration create

topic of class action suggests that collective resistance in the form of reformatory action is a viable, not to mention more practical, alternative for workers (Giddens 1982; Mann 1973; Rubin 1986). Such action typically emerges at key temporal moments, and oftenprecedes overt manifestations of class sol- idarity (Fantasia 1988; Letwin 1998; Roscigno and Danaher 2001). Furthermore, workers have at their disposal a repertoire of individual actions that can be, and often are, used to exert class interests on the shop floor (Halle 1984; Jermier 1988; Molstad 1986). Following Hodson (1996:722), such actions are "attempts to defend or regain dignity in the face of work organizations that violate workers' interests, limit their prerogatives, or otherwise undermine their autonomy." Without taking into account these alter- native forms of resistance, as we do in the analyses that follow, sociological accounts of workplace change and worker response (or lack thereof) become overly determined and devoid of primary forms of human agency.

at least a limited sphere of shared interests between employees and management. Where such coherence and integration are missing, employees may define management as illegiti- mate (Halaby 1986). Employees also may feel that their individual and collective interests in taking pride in secure and stable work are threat- ened (Hodson 2001a).

SOCIAL RELATIONS AT WORK

The literature on workplace organization, pre- viOuslydiscussed accords prior it^ to structure above and beyond social dynamics occurring on the shop floor. Poor interpersonal relations in the form of conflict and supervisory abuse, if they are dealt with at all, tend to be viewed as byproducts of organizational structure rather than causal agents in their own right (e.g., B~~~~~~1985; ~ d 1992). We would not deny that associations at organizational and interpersonal levels exist. Far from it. One can easily see, for instance, how a high level of bureaucracy might decrease the need for, or likelihood of, more abusive and conflictual manager-worker relations. It is important to note, however, that there is social and interpersonal variability internal to any par- ticular organizational form-facts that must be considered when generating theoretical models of worker behavior and the realities of work- places. Variations in social relations, and worker treatment and conflict with managers more specifically, may hold implications for workers' sense of injustice above and beyond workplace structural arrangements. Recent sociological research concurs on this point. Vallas (1987:252), for instance, analyzes intrinsic and extrinsic attributes of workplaces simultane- ously, concluding that "it is not the tasks work- ers perform, but the broader treatment they receive at the hands of management that impinges on their level of class consciousness" (see also Vallas 2003). Hodson (1999) similar- ly finds that conflict with and abuse by man- agers erodes worker citizenship, even with work characteristics and organizational structural variations controlled. Juravich (1985) provides ethnographic detail pertaining to the point we are making. Here, in a wire manufacturing company, the dynamic nature of workplace interpersonal relations, worker interpretation of abusive managerial

THEFOUNDATIONS OF WORKER RESISTANCE

q

behavior, and resistance in the form of sabotage become quite apparent:

[Bobby] was originally called to make a small adjustment on the depth of the machine's applica- tor. It was a simple adjustment accomplished by loosening a single screw. In a normally equipped shop it would have been a five-minute job, but Bobby could not find the proper screwdriver. We searched all the toolboxes. but the screwdrivers were either too large or had been ground at the ends. Bobbv asked Carroll [the boss1 if he could buy a screwdriver at the hardware store down the street. Carroll refused and told him to grind one

of the ones we had, Bobby tried but ended up stripping the screwhead so badly that nothing could get it out. Then Carroll came to the floor and in typical fashion chewed Bobby out in front of every- body. After Carroll left, Bobby brought the applicator over to the bench and . . . used a ten- pound copper n~allet to smash a machine Part that cost hundreds of dollars to replace. (Juravich

~~~d~

1985:135-136)

As this example suggests, workers hold a normative sense of what constitutes proper man- agerial ethics. Such ethics include good lead- ership, knowledge of the labor process, and fair and respectable treatment of employees (Brody 1960; Halaby 1986; Moore 1962).

Whether social relations at work exert an independent effect on workers with implica- tions for resistance is a difficult question to answer. Theorizing on legitimacy, interpreta- tion, and the emergence of insurgent action, however, does provide some basis for pre- dicting that social relations at work make a difference. This literature suggests, most gen- erally, that insurgency at group and individ- ual levels requires not only grievances about structural inequalities and constraints, but also an understanding and interpretation that puts a malicious or greedy face on the suf- fering that is occurring (Gamson 1995). Such causal interpretation provides potential actors with not only a more concrete target, but also an essentially moral justification for acts of resistance and contention (Roscigno and Danaher 2001; Snow and Benford 1992; see also Della Fave 1980, 1986). It remains unclear, however, whether patterns of mana- gerial behavior and worker response uncov- ered in qualitative accounts of the shop floor are, in fact, due to unmeasured organization- al differences. The methodology typically employed is understandably case-oriented, which is detailed and useful in providing rich insight into the active nature of workers and workplaces. However, such methodology usu- ally constrains observations to one or a small set of environments. Thus, the wide continu- um of organizational practices (from poorly to carefully organized, and from informal to highly bureaucratic), within which the micropolitical context of manager-worker relations are played out and possibly even conditioned, is largely missing.

ORGANIZATIONAL AND SOCIAL

INTERPLAY WITHIN UNION AND

HISTORICAL CONTEXTS

Sound sociological theory brings together awareness of structure and action and their potentially reinforcing andlor conditional nature (Giddens 1984; Lawler, Ridgeway, and Markovsky 1993). For our purposes, this entails not only simultaneous consideration of work- place organization structure and interpersonal relations, but also their possible joint impact on worker collective and individual action within union and historical contexts.

There is good reason to expect that organi- zational structure and social relations condi- tion one another and, thus, have contingent effects on worker grievances and resistance strategies. First, as Burawoy (1985) has noted in one of the most developed treatments, worker-supervisor relations on the shop floor may be driven partially by organizational structure. Thus, all else being equal, workers will likely prefer consistency and integration in the structure of the organization and the behavior of the managers. Bureaucratic, well- organized work sites, for instance, procedu- rally constrain managers, at least to some extent, from adopting abusive tones and also provide institutional outlets for conflict reso- lution. Such a straightforward theoretical link- age, nevertheless, overlooks the often loose coupling between organizational setup and the capacity of actors (including workers and managers) to act independently of those struc- tures. Moreover, causal and analytic priority in this association is given to organizational structure, even though such structure is fun- damentally dependent on individuals com- plying with and, in fact, re-creating it. To be sure, the constraint of action by organizational structure may be the case much of the time. However, our theoretical models must also acknowledge agency3 on the shop floor, its impact on organization, and possible organi- zational and interpersonal inconsistencies that may emerge (Vallas 2003).

Table 1 presents standard predictions regard- ing worker resistance, and then our conditional framework. The workplace organizational model suggests that well-run, bureaucratic work environments will experience less resist- ance due to institutionalized grievance proce- dures, greater control over the method and pace of production, and clear delineation ofjob tasks. A focus purely on social relations, in contrast, would suggest that poor interperson- al relations especially between workers and their supervisors will trigger resistance- oriented action.

We recognize the possibility that organiza- tional and interpersonal factors may act inde- pendently in fostering resistance. Our conception, however, views these workplace features as mutually dependent. Well-organ- ized, bureaucratic workplaces with little inter- personal conflict and abuse, for instance, will generate lower levels of resistance, while poor- ly organized more informal work environments with poor social relations will display higher lev- els. Note, however, that our conditional con- ception also allows for disjuncture across levels. We base this on our previous assumption that organizational actors, including managers and workers, have significant agency. Flexibility in how one interprets organizational requirements and procedures can generate a loose coupling of organizational structure and interpersonal behavior. Such would be the case in a well organized and bureaucratic work environment with significant worker-manager conflict, or in a workplace that is poorly organized but has rel- ative stability due to good management on the shop floor. Such situations will likely generate resistance, although its extent may be mitigat-

By agency, we are referring to the objective capacity of individuals to act collectively or indi- vidually in a manner that either reinforces or under- mines prevalent social relations and organizational structure. For the purposes of the analyses that fol- low, we suspect that significant agency is often expressed by workers and managers even in the face of some constraint and possible sanctions that expres- sions of agency might entail.

Table 1. Theoretical Models and Emuirical Predictions Pertaining to Worker Resistance

Workplace Organization Model Workplaces characterized by formal organizational structures and procedures will
  exhibit less worker resistance. This may be due to several factors, including for-
  mal organizational capacity to deal with grievances, greater workplace bureaucratic
  control, and clear specification and constraints on job duties and responsibilities.
Social Relations Model Interpersonal relations in the workplace (e.g., abuse by managers and conflict
  between managers and workers) will provoke resistance by violating workers'
  normative expectations and thus helping to legitimate or justify resistance-
  oriented action.
Conditional Model The impact of workplace organization on worker resistance will be conditional on worker-manager relations, and vice versa. Specifically, well-organized, bureaucratic
  environments may foster lower levels of worker resistance, although this will like-
  ly vary depending on levels of managerial abuse and worker-manager conflict.
  Workplaces characterized by poor and informal organization, in contrast, will
  experience heightened resistance at both collective and individual levels, although
  this may be buffered by good shop floor social relations.
Union and Legacy Model Workplaces characterized by historical class identity and union presence will be
  more amenable to labor organization and activity. Direct and more aggressive
  forms of individual resistance may be more notable. Such effects, however, may
  be conditional upon organizational and interpersonal dynamics within workplaces.

ed by positive organizational or interpersonal union's ability to tap into worker consciousness features of those very workplaces. and radicalism already forged at the point of pro-

Beyond acknowledging organizational and duction. Rubin, Griffin, and Wallace (1983) social dynamics and the contingent impact they similarly recognize the somewhat tenuous rela- may have, theoretical formulations of worker tionship between unions and activism, sug- resistance must also be sensitive to preexisting gesting that pre-existing militancy on the shop differences in labor organization, class identi- floor along with political opportunity may con- ty, and legacies of collective action (Cotgrove dition the association and perhaps even pre- andvamplew 1972; Griffin, Botsko, Wahl, and cede union involvement. Isaac 199 1 ;Vallas 1987). For this reason, Table A historical legacy of resistance, while often 1 includes an additional prediction concerning related to union presence, may be influential in union presence and historical legacy effects. its own right. Drawing from broader theoretical

Union presence may bolster the likelihood of arguments pertaining to the sequential nature of worker resistance (Ashenfelter and Johnson social causation (for instance, see Abbott 1995), 1969; Edwards 1996; Rubin 1986). This expec- recent research has begun to delineate the impor- tation of a positive impact parallels broader the- tance of path dependence, or how temporally orizing and research on social movement prior political processes and policies influence organizations and the role they play in provid- those that follow (e.g., Pierson 2000, 2001). ing movements and movement participants with Formal and informal resistance behaviors may financial and informational support, networks, similarly be shaped by successive processes. ideological framing, and recruitment, all of Past insurgency, for instance, may be meaning- which are essential to get a movement off the ful for future action through the establishment ground and ensure its survival (e.g., Isaac and of interpersonal and organizational networks Christiansen 2002; Jenkins and Eckert 1986; (Minkoff 1997; Shin 1994) along with the forg- Staggenborg 1989). Historically grounded ing of oppositional frameworks, identity, and analyses of unions, however, reach a more abeyance structures that can be explicitly acti- nuanced interpretation. Kimeldorf's (1 999) vated at a later point (Taylor 1989). Research on detailed account of striking longshoremen and past and contemporary worker resistance, using hotel and restaurant workers in the early 1900s, a variety of methodologies, concurs with these for instance, denotes a somewhat variable union possibilities. Specifically, a legacy of challenge role-a role that often depends on a given and its level of success or failure have implica- tions for future organizing (Griffin, McCammon, and Botsko 1990), the degree of worker quiescence (Edwards and Hyman 1994; Shalev 1992), and the development of critical consciousness among workers (Fantasia 1988; Kelly and Kelly 199 1).

Union and historical legacy effects, as noted above, certainly may be independent of those resulting from the internal features of work- places. We suspect, however, that union presence and mobilization history will have varying effects depending upon the actual, current expe- riences of workers on the shop floor-a conditional expectation derived from historically grounded analyses and interpretations of union activity, shop floor experiences, and worker contestation (e.g., P. Edwards 1978; Isaac and Griffin 1989; Kimeldorf 1999; Roscigno and Danaher 2004; Rubin, Griffin and Wallace 1983; Snyder 1975).

DATA

Our data are derived from the systematic cod- ing of all book-length workplace organization- al ethnographies pertaining to the United States and England (N= 82). Each represents an aver- age of over a year in the field with at least as much additional time spent in analysis and writ- ing. The accumulated record of organizational ethnographies is based on more than 160 years of Ph.D. level observation and interpretation. Until recently, this rich data resource has remained largely unanalyzed by social scientists studying organizations (see Harnrnersley 1997; Schwartzman 1993). The systematic analysis of data from a comprehensive set of organiza- tion ethnographies takes advantage of the depth and range of observation offered by ethnogra- phies, while avoiding the limits of analyzing a single case (Ragin 2000). Indeed, such data allow us to move beyond the bikcation between case study analysis and large sample analysis evident in much prior work. Systematic coding of workplace ethnographies allows for detail pertaining to managerial behavior and worker resistance strategies, while also providing con- siderable variation in the organizational struc- ture and significant representation across occupation and industry (Table 2).

The current analysis thus brings to bear data representing sustained in-depth observation of the workplace and workplace relations of a sort that is rarely utilized outside of specific case study settings. There are, nevertheless, certain limitations worth mentioning, not the least of which is the underlying assumption that these data constitute a realist account of the organi- zational structure and worker behavior existing in each workplace. Each ethnographer, in sum- marizing his or her experiences, chose to report certain events as typical. These events then form the basis of our codings (Van Maanen 1998; Lee 1999). The prevalence of a common frame of reference in these workplace studies allows the coding of key indicators across volumes (see Schwartzman 1993; but see also Blee and Billings 1986). The current analysis thus sum- marizes the available ethnographic evidence on the workplace and goes well beyond tradition- al analyses of ethnographic data by introducing probabilistic logic and affording comparative

leverage across organizational types.

Generalizations derived from these data and applied to the population of all organizations must admittedly be made with some caution. The cases analyzed do not necessarily reflect a representative sample of all organizations. Rather the data represent the population of avail- able ethnographic evidence on organizations, and conclusions from their analysis must be interpreted in that light.4 The significant varia- tions in industrial representation (Table 2), as well as the generally wide representation along the dimensions of workplace organization and social relations, nevertheless bolster our confi- dence in the utility of these data relative to analyses of single cases.

WORKER RESISTANCE MEASURES

Consistent with our previous discussion, it is important to measure both collective forms of resistance and more informal, individual strate- gies. The ethnographic data described earlier provide both, and the relevant variables are described in Table 3. For analytic purposes,

The codesheet, coding protocol, and data are available at http:llwww.soc.sbs.ohio-state.edu/ rdh/Workplace-Ethnography-Project.htm1.As with any content analysis project, we may have made errors in the interpretation of the texts or in the cod- ing of the data. The data, however, are available for public scrutiny and analysis and we welcome sug- gestions, criticisms, and alternative views on the recorded data.

Table 2. Distribution of Ethnographies by Industry and Occupation (N= 82)
Cases
YO (n)
Industry  
Extractive 8.5  
Construction 3.7  
Non-Durable Foods 4.9  
Textile Products 3.7  
Paper Production 1.2  
Chemical Production 8.5  
Lumber and Wood Products 2.4  
Metal Products 4.9  
Machinery/Electrical Products 3.7  
Transportation Equipment 8.5  
Professional Equipment 4.9  
Transportation/Communications and Other Public Services 9.8  
Sanitation Services 1.2  
Retail Trade 4.9  
Finance/Insurance/Real Estate 11.0  
Business and Repair Services 1.2  
Personal Services 8.5  
Professional and Related Services 4.9  
Public Administration 3.7  
Occupationa  
Professional 8.5  
Managerial 8.5  
Clerical 9.8  
Sales 1.2  
Skilled 13.4  
Assembly 31.7  
Unskilled 9.8  
Service 12.2  
Farm 4.9  
a Data in the "Occupation" section is from within the "Industry" population total of 82.  

each of these outcomes as well as all indepen- occurrence was not observed or was rarely dent variables are measured dichotomously. observed. For those workplaces coded 1, the

Collective resistance is indicated by a strike particular form of individual resistance was occurrence during the period of observation. observed often or was even prevalent in the As noted in Table 3, approximately 2 1 percent particular workplace. In nearly half of the work- of the establishments included in the sample places observed, social sabotage and absen- experienced such a strike. As noteworthy, if not teeism were witnessed as relatively more so, is the prevalence of individual resist- commonplace. In more than half, significant ance strategies, including social sabotage, work work avoidance was observed. Theft and play- avoidance, playing dumb, absenteeism, and ing dumb are less pronounced, yet their con- theft.5 Each of these measures is coded 0 if the sistent observation in 20 to 27 percent of the

In preliminary analyses we also examined be more constrained by the technology, the organi- machine sabotage. Incidences of machine sabotage, zation of production, and the resulting opportunities however, are rare relative to the other forms of indi- and limits on workers' actions. Given this uniqueness, vidual resistance. This may be a function of the seri- both in terms of its actual statistical occurrence and ousness of such behavior which, unlike the others, is the processes that likely guide its occurrence, machine a legally criminal act. We also suspect that machine sabotage and its analyses are not reported alongside sabotage is quite distinct in that such behaviors will those of other individual resistance strategies.

22 AMERICAN SOCIOLOGICAL REVIEW
Table 3. Definitions, Coding, and Means for Dependent Variables
Variable Definition Coding Mean
Collective Resistance Striking Individual Resistance Social Sabotage Work Avoidance Playing Dumb Absenteeism Whether there was a strike during the period of observation Undermining of superiors through mocking and ridicule Avoiding work andlor work tasks Pretending not to understand particular job tasks or organizational procedures Absenteeism as a response to workplace 0 = none 1 = strike occurred 0 = notirarely observed 1 = observed oftedprevalent 0 = notirarely observed 1 = observed oftedprevalent 0 = notirarely observed 1 = observed oftedprevalent 0 = notirarely observed ,207 ,427 ,573 ,270 .45 1

problems Theft Stealing while on the job

workplaces investigated reveals that these are by no means rare occurrences.

ORGANIZATIONAL AND SOCIAL RELATIONS MEASURES, AND UNION PRESENCE/STRIKE HISTORY

The use of QCA, which we describe in the upcoming analytic strategy section, forces us to focus on a limited number of independent vari- ables that are closely tied to the theoretical con- cerns addressed previously. We thus focus attention on arguably the two most important organizational structural attributes, bureaucra- cy and coherent organization, and on two key indicators of workplace social relations, man-

1 = observed oftedprevalent 0 = notirarely observed ,195 1 = observed oftedvrevalent

agement abuse and workplace conflict. Along with these, and consistent with our interest in possible union and historical legacy variations, are indicators of union presence and strike his- tory at the workplace. These variables are described in Table 4.

Bureaucratic organizations, which comprise approximately 60 percent of the sample, are those wherein operational control of daily pro- cedures resides in written rules. Such rules include day to day operations of production as well as grievance procedures and avenues. While virtually all contemporary workplaces entail at least nominal bureaucracy, this indicator is scored "yes" only for those cases in which the actual de facto control of the work process is

Table 4. Organizational and Social Relations Measures and MacroiHistorical Elements

Variable

Organizational Attributes Bureaucratic (BUREAU)

Good Organization (GOODORG) Social Relations Managerial Abuse (ABUSE) Workplace Conflict (CONFLICT)

Union and Legacy Measures Union Presence

(UNION) History of Strikes (HIST) Definition

Workplace bureaucratically organized- operational control of daily procedures resides in written rules

Coherence and integration of production practices

Verbal, emotional, or physical abuse by supervisor of individual employees Ongoing conflict between workers and

supervisors

Union representation in the particular workplace

Workplace has experienced strikes in the past

Coding Mean
0 = adequate or less 1 = good or exceptional ,351
0 = never or rarely 1 = sometimes or common 0 = never or rarely 1 = sometimes or common ,402 .354
0 = no union presence 1 = union presence 0 = no strikes in the past 1 = strikes in the vast .598 ,171

governed by bureaucratic rules. The use of bureaucratic rules for governing promotions and other job-related benefits is not sufficient for an organization to be considered bureau- cratically organized. An example from an ethno- graphic study of an insurance company illustrates a significant use of bureaucratic rules in the daily organization of work tasks:

The largest source of dissatisfaction for Kevin,

however, is the bureaucratic framework of Servall,

with its routinized work patterns and its restrict-

ed autonomy: ". . . I'm in a framework, a corporate

framework, where I have to abide by their rules and

regulations for everything, which gets to me

because of all the bureaucratic junk that I have to

go through to complete something. I know there's

a faster way to do something, but I have to follow

their ways, which is frustrating sometimes." (Burris

1983:157)

The measure of bureaucracy is distinct from that capturing good workplace organization. Good organization entails a coherence and inte- gration of production practices including but not limited to the availability of materials, the effi- ciency of work flow, and the organization of work specific tasks. Those establishments that were good or exceptional on this dimension were coded 1, while others were coded 0. Distributions across these two organizational measures suggest considerable variation in the workplace organizational structure in this sam- ple of ethnographies.

Managerial abuse and conflict are arguably the most important dimensions of workplace social relations, and they have implications for worker satisfaction, grievance interpretation, and resistance behaviors. Our indicator of abuse includes whether or not verbal, emotional, or physical abuse of employees by supervisors was witnessed as a common occurrence during the observation period. In the quote from a wiring harness manufacturer described previ- ously, the example of a supervisor yelling at a worker in front of other workers illustrates such abuse. Specific incidents of abuse are distinct from our measure of conflict, which taps into ongoing, systematic animosities between groups of workers and supervisors. A notable 40 percent of workplaces examined had significant levels of supervisory abuse toward individual employees, while 35 percent of workplaces dis- played evidence of ongoing conflict between supervisors and workers.

Conflict in the form of gossip, backbiting, and character assassination in a situation of chron- ic mismanagement is reported in an ethno- graphic study of a British apparel factory:

There was no suggestion from the women in John's department that management had either the right or the ability to manage. Instead, the women were constantly critical of management. They asked, 'When are they going to manage? After all, it's what they get paid for and it's a dam sight more than we get.' The [lead workers] especially, were very critical of management: Gracie: The trouble with this place is we never know what's happening and it's my bet that man- agement don't know either. . . . Jessie: Either we've got no work or there's a bloody panic on here. I ask you, what do management do with their time? I reckon I could do better myself than this lot. This place never runs smoothly. . . . Edna: I agree, they tell you one thing, you get ready to do it and then it doesn't arrive. We could do better ourselves, I don't know what this lot get paid for. (Westwood 1984:25-26)

Conversely, an example of infrequent con- flict with managers is provided by an ethno- graphic study of copy machine repair workers, which reports that "most technicians enjoy informal, casual relationships with their immediate managers, most of whom have recently been technicians themselves" (Orr 1996:68).

Along with organizational and social fea- tures, we include indicators of union presence and strike history for reasons explained pre- viously. Union presence denotes whether there is union representation at the particular work- place. Approximately 60 percent of the work- places examined have some form of union presence, while 17 percent of the establish- ments had experienced strikes in their respec- tive histories, according to the ethnographic accounts. Although ethnographic accounts may underestimate strike history, due to lim- itations in the ethnographers' knowledge of the particular workplace history, we suspect that this measure remains a reasonable indicator. Indeed, if an ethnographer garnered informa- tion about past strikes from a recorded histo- ry or especially from workers themselves, then that history likely continues to be con- sequential for workers' perceptions, consciousness, and available repertoires of resistance.

ANALYTIC STRATEGY AND RESULTS

We analyze configurations of workplace organ- ization and workplace social relations, along with union presence and strike history, using a combination of QCA techniques, based on Boolean logic (Ragin 1987), and more conven- tional quantitative methods. This multi-method strategy affords us the ability to adopt QCA's case-oriented logic and reductive power, while simultaneously extending its typical (deter- ministic) usage in a more probabilistic man- ner.

QCA forces the analyst to consider all pos- sible combinations of causal factors an4 with its comparative algorithmic logic, eliminates redundant and superfluous information. The benefits of QCA, which lie in its ability to spec- ify configurations of variables relative to all theoretical possibilities and their implications for various social outcomes, are being increas- ingly recognized in social science research. Recent articles using this method, for instance, have focused on wage policies and social wel- fare programs (Amenta and Halfmann 2000), the emergence of the social security system (Hicks 1994a), strikebreaking and split labor markets (Brown and Boswell 1995; Brueggemann and Boswell 1998), patterns of union growth and decline (Ebbinghaus and Visser 1999; Griffin et al. 1991), labor policies in Southern textile mills (Coverdill, Finlay, and Martin 1994), the success or failure of left-lib- ertarian political parties (Redding and Viterna 1999), and the success or failure of mobiliza- tion drives among the homeless (Cress and Snow 2000).

QCA is uniquely suited to address our theo- retical arguments, especially the potential con- ditional relations described earlier, given its ability to analyze complex conjunctures of caus- es in relation to a particular outcome. The method, however, is not without certain limita- tions-limitations that have received consider- able attention recently (e.g., Abbott 200 1;Hicks 199413; Lieberson 200 1). These include being constrained to a limited number of independent variables because of the conditional logic of QCA and the number of configurations gener- ated. The inclusion of large numbers of inde- pendent variables makes interpretation exponentially unwieldy. Consequently, the researcher using QCA is forced to specify and focus on variables deemed theoretically impor- tant to the processes outlined. One might inter- pret this as a methodological weakness in that an array of control variables cannot be includ- ed as in the typical regression model. We believe, however, that the benefits of QCA- theoretical rigor in choosing variables in the first place, its case-oriented logic, and the spec- ification of potentially complex, conditional configurations-outweigh the costs (see also Boswell and Brown 1999; Griffin et al. 1991 ; Ragin 2000).

A more serious limitation, in our view, is the typically deterministic character of QCA results. In common usage, results are derived from a log- ical reduction of configurations that are posi- tively related to the outcome. That is, configurations of variables are usually generated in relation to the outcome always being coded "1" (i.e., yes or present), and then reduced by the program by logically teasing out redundan- cy and irrelevant factors. While useful in denot- ing conditions under which an event always occurs, this approach to using QCA does not make full use of the data and does not capture tendencies, variations, and divergences from absolutes in the real world. Specifically, infor- mation on configurations associated with a neg- ative (0) outcome on the dependent variable or configurations associated with a contradictory outcome (where some cases in the configuration are coded 1on the outcome of interest and other cases are 0) is usually not presented or consid- ered. This is unfortunate, as configurations that are associated with 1,0, and contradiction rep- resent the actual degree of variation in the rela- tion between the explanatory pattern and the dependent ~ariable.~

Acknowledging especial- ly contradictory configurations and their rela- tion to the dependent variable of interest can introduce probabilistic possibilities and inter- pretations into the typically restrictive QCA model. Ragin (2000: 133) concurs on this point,

It is for this reason and others that the first step in QCA, the generation of a truth table displaying the observed configurations and the outcome variable, is arguably the most important step (Boswell and Brown 1999). It helps analysts establish the coverage of their data, relative to all theoretically possible cases. Moreover, considering contradictory and negative cases within a truth table is essential to "guard against making spurious inferences and to reinforce conclu- sions drawn from positive cases" (Griffin et al. 1991).

suggesting how researchers can and should explore non-deterministic configurational pat- terns through the use of percentages, probabil- ities, and even t tests. In doing so, one can arguably show how the prevalence of an out- come for a particular configuration of attributes differs from that of another group or from some predefined standard set by the researcher.

Our modeling takes these criticisms and sug- gestions to heart by utilizing QCA to generate and then logically reduce all configurations rep- resented in the data by eliminating superfluous information. The result is essentially a set of organizational typologies-typologies that denote unique combinations of attributes in the data. More specifically, these configurations denote the minimum number of factors (i.e., variables) and configurations of factors need- ed to logically cover all positive (I), negative (0), and contradictory cases in the data. The theo- retical benefit here lies in the preservation of a case-oriented focus-a focus that underlies QCA-and the need for theoretical clarity in selection of variables. And, relative to more commonly used quantitative methods, such as OLS or logistic regression, we begin with the assumption that there exist cases (types of work- places) with meaningful combinations of attrib- ute~.~

7As suggested by Griffin et al. (1991) and Ragin 1987), the assumptions of QCA are fundamentally distinct from more conventional quantitative meth- ods that attempt to tease out effects of individual variables, thus abstracting or stripping variables from the cases within which they exist. "Real cases, how- ever, are comprised of attributes and their relational configurations, and it is the precise constellation of these relationships which makes one case different from another" (Griffin et al. 1991 :133; emphasis in original). Conventional quantitative modeling can certainly capture unique configurational combina- tions with the introduction of interaction effects, although the typical usage tends to focus on effects of individual variables rather than on cases that fall into some form of typological category. Moreover, to adopt a full configurational approach in OLS or logistic regression, for instance, would require a test of all possible interactional combinations. In our case, with six independent variables, this would amount to testing all possible two-way, three-way, four-way, five-way, and six-way interactions, many of which may contain redundant information and some of which do not actually exist in the data.

THE FOUNDATIONS OF WORKER RESISTANCE

a5

The typical use of QCA relies on a second reductive step in which further Boolean alge- braic reduction identifies any deterministic cor- respondence between configurational attributes and a positive (or a negative) outcome on the dependent variable. Instead of relying on this second step, which is limited in dealing with probabilistic patterns and the degree of varia- tion across outcomes as noted previously, we undertake a two-pronged multi-method strate- gy wherein (1) QCA's initial reduction procedure is used to identify nonredundant configurations in the data, and (2) we then analyze these con- figurations and their associations with collec- tive and individual resistance using more conventional, quantitative method^.^ Specifically, we use t tests to make proportion- al comparisons between the distribution of out- comes for a given configuration and that of the outcomes for cases not captured by that con- figuration. We ask, simply, is a given configu- ration associated with the dependent variable, and is this association significantly and statis- tically distinct from cases not captured by the configuration? For ease of interpretation, ratios of configuration percentages to nonconfigura- tion percentages on the dependent variable are used, with asterisks denoting statistically sig- nificant effects.

Beyond addressing positive, negative, and contradictory associations, and thus providing a more probabilistic interpretation of patterns, our strategy has the added benefit of offering some comparative leverage when examining multiple dependent variables. By selecting pos- itive, negative, and contradictory configura-

The cost of our approach is that it is limited in making causal, as opposed to associational, claims. This is why we do not use the typical QCA descrip- tive terminology of "necessary" or "sufficient" in our discussion or results. Nevertheless, since organiza- tional structures and social relations in the work- place, not to mention union presence and strike history, were most assuredly intact prior to observa- tion of resistance-oriented behavior, it seems rea- sonable to allow some causal interpretation. Moreover, our consistent findings even when retest- ing associations using logistic regression techniques bolster the likelihood that the associations we are describing may be of the causal variety. Assuredly, though, any ultimate test of causal claims necessitates longitudinal data.

tions for the initial reduction, QCA begins with the entire truth table (all observed configura- tions; see Appendix Table The configura- tions generated with QCA will thus be the same across dependent variables. Yet, the pattern of associations across configurations for the dif- ferent outcomes, derived from our quantitative comparisons, may differ. This allows us to draw substantive conclusions pertaining to the con- figurational relationship to collective action (i.e., striking) relative to more individual-level resistance strategies.

We begin by examining collective resistance manifest in the form of strike occurrence (Table 5). The key question is, what factors and/or configurations of factors enhance or diminish the likelihood of strike action? The notation used to report configurations follows QCA con- ventions. The presence of a particular work- place attribute is indicated by capital letters, while the absence of that characteristic is denot- ed by lowercase letters. An asterisk indicates "and," and denotes that all of the conjoined

Here, we are reducing all configurations in the data, not necessarily those strictly tied to some value of our dependent variables. Because our reduction begins with the entire truth table, all positive, nega- tive, and contradictory configurations in the observed data are covered. Along with this point of clarifica- tion, it is also essential to be explicit about combi- nations that logically exist, but that are not evidenced in the data, since this has implications for general- izability claims (Griffin et al. 199 1). It is for this rea- son that Boswell and Brown (1999) suggest that researchers using QCA make use of a "diversity index," or a ratio of observed to theoretically possi- ble combinations. In the case of the analyses to fol- low, the data coverage is reasonable given the number of independent variables being used. Indeed, "The more complex the model, the greater number of unusual combinations of variables for which cases could not exist" (Boswell and Brown 1999: 161; see also Ragin 1987). Exclusive of union presence and strike history, the number of observed configura- tions is 26 of 32 possible configurations, yielding a ratio index of 8 1.3. With the addition of union pres- ence and strike history, the number of observed con- figurations with six independent variables becomes 33 of 64 possible, or a diversity index of 54. In the results section, we briefly discuss reduced configu- rations that do not actually exist in our data, and what these mean relative to our analyses and con- clusions.

attributes are part of the configuration. We also report the total number of cases captured by the configuration and the distribution (raw nurn- ber and percentage) on the outcome. For reasons of interpretation, and as noted previously, we standardize this percentage by the average for cases not in the configuration, thus generating a more interpretable ratio. Values greater than one indicate a positive association and a gen- erally greater likelihood of a strike occurrence among cases in the configuration (relative to those cases not in the configuration). In contrast, values less than one suggest a depressant effect. Where such effects are statistically significant, they are so denoted with asterisks. To highlight the strength of our multi-method strategy and the robustness of findings relative to strike occur- rence, we reestimate the positive significant configurational associations in Table 5 using logistic regression (Table 6), and we include in these models the controls that also may poten- tially shape strike activity but that are not as cen- tral to our theoretical concerns.

The second portion of the analysis (Table 7) examines the same configurations and their respective associations with individual resist- ance in the form of social sabotage, work avoid- ance, playing dumb, absenteeism and theft. As with the strike analysis, statistically significant differences are highlighted. Findings pertaining to individual resistance are discussed relative to the predictions outlined previously, and rela- tive to the analysis of strikes.

COLLECTIVE RESISTANCE

Table 5 reports levels of strike occurrence in relation to reduced configurations. Notably, each of the nine configurations includes at least three elements, and five highlight the necessary coexistence of some organizational and some social attribute in the workplace. The inclusion of indicators specifying union presence and strike history is validated by the inclusion of at least one of these attributes in each of the nine configurations.

The first configuration we report (CON- FLICT * UNION * BUREAU) indicates work- places with significant conflict on the shop floor, union presence, and a bureaucratized work structure. Of the 18 cases that hold all three attributes, 11 (approximately 61 percent) experienced a strike during the period of obser- vation. Since it is important to interpret this within the context of other cases in the sample that do not have these configurational attributes, we focus on the ratio. The resulting ratio of 6.5 suggests that those workplaces characterized by the constellation of conflict, union presence, and bureaucracy are six and a half times more likely than other workplaces to witness a strike occurrence-a large effect, to be sure.1° The t test results reveal this difference is statistically significant, well beyond the .001 1evel.l

CONFLICT and UNION are necessary attrib- utes in not only the first configuration, but the second configuration (CONFLICT * UNION * ABUSE * goodorg) as well. Within these work- places, individualized abuse and poor organi- zation also prevail, and strikes are approximately

4.4 times more likely compared to cases not captured by the configuration. The combination

lo Ratios were calculated by dividing the config- uration mean by the mean of cases not falling into the configuration. The ratio reported above, for instance, was derived from the following: 61.1 119.38 = 6.5. Although non-configuration means are not reported in the tables, they are available from the authors on request.

'' In Table 6, reported momentarily, we test whether these configurations are significant when modeled in logistic regression. We also tested whether other con- figurations reported in Table 5 are significant in logistical regression, which they are not. We attrib- ute this in part to weaker associations with striking, but also the relatively small (82) sample size.

THE FOUNDATIONS OF WORKER RESISTANCE

of conflict and union presence, noted in each of our first two configurations and that eventual- ly culminates in strike action, is captured in the following:

In building up an organization on the section the steward enforces unwritten agreements from his supervisor. When the supervisor is placed under pressure by his superiors he often breaks these secret understandings. Jack [the union steward[ committed his feelings to print. He filled in a pro- cedure report calling the supervisor a 'perpetual liar' and a 'deceitful bastard'. The supervisor went to law, but he wasn't allowed to push it too far. Higher management persuaded him that the case was better dropped and Jack Jones escaped his chance to testify in the dock. . . . The establishment of a steward in a particular section was clearly related to the attempt by the workers to establish job control in that section. If the steward wasn't up to the job he was replaced, or he stood down leav- ing the section without a steward for a while. Where a steward stuck with the job, he and the men on his section were involved in a perpetual battle with foremen and management. (Benyon 1975:144-145)

As the example above suggests, union pres- ence in a conflict-laden environment may be meaningful due to the presence of shop stew- ards and their place in addressing worker griev- ances in an ongoing manner-a manner that may galvanize worker loyalties over time. Consistent with some of the themes we provid- ed at the outset, we also attribute such effects to greater labor union and strike legitimacy,

Table 5. Reduced Configurations and Variations in Strike Occurrence

Configurations CONFLICT * UNION * BUREAU CONFLICT * UNION * ABUSE * goodorg CONFLICT * UNION * HIST * GOODORG * abuse UNION * conflict * hist UNION * conflict * abuse

* goodorg conflict * abuse * hist conflict * goodorg * hist bureau * goodorg * hist bureau * unions * hist

Total # of # of
Cases zeros ones
18 7 11
7 2 5
4 2 2
21 19 3
8 8 0
29 29 0
34 32 2
20 17 3
16 16 0

Strikes in Mean Ratio Configuration (configuration to Example (%) non-configuration) ethnography

61.1 6.5 15*** Wedderburn (1972)

71.4 4.420** Seider (1984)

50.0 2.600 Walker (1957)

13.6 ,585 Cavendish (1982)

0.0 ,000 Finlay (1 988)

0.0 .000*** Burris (1983)

5.9 .188** Paules (1991)

15.0 ,867 Foster (1969)

0.0 .OOO* Juravich (1985)

* p < .05; **p< .01; ***p < ,001 (one-tailed test): denotes statistically significant difference between configura- tion mean and mean of cases not captured by the configuration.

typically fostered on an ongoing basis by stew- ards and workers through networks and meet- ings (Dixon and Roscigno 2003). This possibility is reflected in the following obser- vation from the same plant:

The stewards in the Paint Trim Assembly plant met each other regularly. They ate their meals together in the works canteen and drank together after meetings. They were friends. Occasionally they arranged social evenings to which they took their wives. On all these occasions they joked and told stories about people and events, about the city and the factories. More often they told stories about the plant, about the early days when the PTA plant was first unionized by the Transport and General Workers Union. These stories in particu- lar were always told to newcomers (the same sto- ries were told to me dozens of times during the months when I cam to know them) particularly to new shop stewards and activists and almost inevitably at times of crisis. (Benyon 1975:74-75)

Ongoing conflict on the shop floor is also cen- tral and is typically driven by a lack of man- agement competence. Worker response, particularly when the problem is viewed as ongoing and systematic, is likely to be collec- tive in nature. Such is the case in the following excerpt, where the workplace also subsequent- ly experienced a strike during the period of observation:

An untactful supervisor had in the past tried to take a strong line against maintenance men he consid- ered to be dilatory by making official complaints to his plant manager. The result was that it seemed extraordinarily difficult to repair electrical faults that occurred while he was on duty; often it was only when his shift had finished and the next had taken over that the job was done. It might be said that he had tried to use his formal powers as super- visor instead of informal influence, and that he failed dismally. (Harris 1987: 149)

Table 6 speaks to the strengths of the findings thus far and to the credibility of our analytic strategy more generally by reexamining the two significant configurations we have just described, using logistic regression and includ- ing a number of other potentially influential controls. The controls include whether the case is in England as opposed to the United States, whether the workplace is in an area of high unemployment, the race and gender composi- tion of the workplace, establishment size, and the era in which the ethnographic observation took place.12 Rather than using interaction terms in the traditional sense, which would be very cumbersome for specifying third and fourth level interactions, each configuration in these models is measured as a dichotomous indicator of whether all configurational attributes are present (1 = yes; 0 = no). Along with greater interpretative ease, this measurement strategy is consistent with the case-oriented focus outlined in our analytic strategy section.

The configurational associations with strike action hold, even with the addition of the con- trols in the second equation of each model. Moreover, the configurational patterns remain strong and statistically significant. This rein- forces confidence in the findings reported thus far. It also highlights the benefits of using QCA to reduce and generate configurational patterns, and then quantitative techniques to decipher associations between types of configurations and the outcome of interest.

Among the controls, the cases that were observed in England, workplaces characterized by relatively high minority concentration, and, to a lesser extent, the era in which the ethno- graphic observation occurred, influence strike likelihood. The influence of England may cap- ture the impact of broader national history and collective action legitimacy, above and beyond union presence and strike history at a singular workplace (Griffin et al. 199 1 ;Kelly and Kelly 1991;Scruggs and Lange 2002). Minority com- position, while it may lead to labor fragmenta- tion due to discrimination and associated competitive processes, may increase collective action given that "negatively privileged groups"

l2 Of the 82 cases in our sample, 3 1 percent are in England and 27 percent are in areas of high unem- ployment. The mean establishment size is 2363, while the average race and gender composition of these workplaces is 18 and 24 percent, respectively. Sixty percent of the ethnographic observation occurred post 1970, approximately 25 percent occurred between 1950 and 1970, and the remainder (15 per- cent occurred prior to 1950. Although not significant in these models, considering historical era effects makes sense given the possibility that conditions of political opportunity may differ (Tarrow 1998). Moreover, broader movements during certain eras may permeate workplaces and radicalize workers (Isaac and Christiansen 2002).

Table 6. Logistic Regression Estimates (Log-Odds) of Configurational Influence on Likelihood of a Strike

(1). , (2)\, (1)~, (2). .
CONFLICT * UNION * BUREAU 2.721*** 2.485***    
CONFLICT * UNION * ABUSE * goodorg     2.422** 1,871*

England High Unemployment Percent Female Percent Minority Establishment Size Era (Referent: Pre- 1950)

1950-1969

1970-Current

Constant -2.269 -1.847 -1.729 -1.461

*p < .05; **p < .01; ***p < ,001 (one-tailed test).

may be more likely to join unions and collec- "If they'd leave us alone, we can take care of the work and make money for the company. We did a

tively mobilize in the first place (Oberschall thousand yards today. But I've done better. As

1993; see also Cornfield and Kim 1994;

long as Carmen [his employer] leaves Earl [the

Zingraff and Schulman 1984).

superintendent] alone we can turn out the work."

Recall that, given the co~lfigurational logic

(Applebaum 1981:63)

used, no one attribute can be interpreted outside the context of the other configurational attrib- The four remaining configurations, reported utes. With this point in mind, we now turn back at the bottom of the table, include no apparent to Table 5 and the impact of union presence strike history (4 configurations), the absence of within the fourth and fifth configurations. conflict (2 configurations), poor organization (2 Notably, and consistent with the conditional configurations), and little bureaucratic struc- expectations highlighted at the outset, union ture (2 configurations). All four configurations, presence in and of itself is not enough to bol- generally capturing unorganized workplaces, ster strike action. Rather, it appears to be the are negatively associated with strike occurrence, copresence of union organization with existent and three of the four patterns are statistically dis- shop floor conflict, reported previously, that is tinct compared to cases not in that particular most meaningful. Indeed, union presence in the configuration. In two of these scenarios, the absence of ongoing conflict, characteristic of our first where there is no conflict, abuse, or histo- fourth (UNION * conflict * hist) and fifth ry of strikes and the second, where there is no (UNION * conflict * abuse * goodorg) config- bureaucracy, unions, or strike history, no strikes

urations appears, if anything, to be associated whatsoever are observed. In the third, charac- with a diminished likelihood of strike action. terized by poor organization but without conflict This possibility is exemplified in the case of

or strike history, the likelihood of strike occur- unionized construction workers who do not

rence is 80 percent less than cases without these strike over the period of observation and who,

configurational attributes. The fourth configu- in the absence of conflict and supervisory abuse,

ration, which entails limited bureaucracy, no express satisfaction in their productivity.

strike history, and poor workplace organiza- At the sewage treatment plant, there were several tion, is only slightly less likely than other cases occasions when the men placed a thousand yards to have a strike occurrence. We interpret these of concrete in a single day. No one from manage- configurations and their negative associations ment told them to do it. The superintendent, fore-

to be largely capturing the reverse of what was

men, and key journeymen decided and planned it

reported at the top of the table-that is, while

on their own initiative. One evening, at the local

union presence and conflict generally combine

bar, in July 1976, after one of the thousand-yard

with other organizational attributes to increase

pours, Pete expressed the pride and satisfaction that comes from extraordinary accomplishment, and strike propensity, their relative absence reduces

said: strike likelihood. Clearly, however, the absence of strike history across all four of the final con- figurations stands out as unique.

In general, our findings suggest that ongoing conflict between managers and workers is fun- damental to forging collective resistance, but only in concert with union presence and some- times with bureaucratic and poor workplace organization and abuse. Such associations, as Table 6 reports, remain strong and robust when retested with traditional techniques that allow for the inclusion of arguably important con- trols. Unions and the organization of production are certainly meaningful as well, although their effects are quite contingent on social relations on the shop floor. Bureaucracy, for instance, is associated with an increased likelihood of strike action, but only in the presence of conflict and unions. Union presence, interestingly, follows the same pattern. Where unions coincide with conflict and sometimes bureaucracy or poor organization, strike likelihood is magnified con- siderably. Where conflict is low, in contrast, union presence is negatively associated with strike action (albeit these effects are not statis- tically significant).I3

What remains unclear is the extent to which the configurations reported above foster or sub- due more individualized forms of worker resist- ance. That is, does the copresence of worker-manager conflict and unions also height- en the likelihood of individual resistance? Might

l3 The following equation denotes reduced con- figurations that theoretically exist, but do not exist in our data:

unions HISTORY +
HISTORY conflict GOODORG +
unions CONFLICT BUR +
HISTORY ABUSE conflict +
unions ABUSE BUR GOODORG +
UNIONS ABUSE CONFLICT bur GOODORG +
UNIONS history CONFLICT bur GOODORG +
HISTORY abuse CONFLICT bur goodorg

Although the absence of cases in our data with these configurational attributes sets some limits on our conclusions and ability to generalize, we suspect that the actual existence of such cases in the real world are limited, at least relative to the configurations that do exist in our data and that are reported here. We nev- ertheless take issues of generalizability seriously, and temper our conclusions keeping in mind our inability to establish the consequences of these missing configurations.

well-run, bureaucratic work contexts mitigate, through the control of workers, individual resist- ance strategies such as theft, sabotage, and work avoidance? And, might unions and strike histo- ry diminish the need to express grievances through more individual routes? It is to these questions that we now turn.

INDIVIDUAL RESISTANCE

In this section we examine individual forms of resistance that workers have at that disposal and embed our understanding of their extent and forms in the organizational and social attributes of workplaces. Table 7 reports a QCA analysis of five discrete resistance strategies, namely social sabotage, work avoidance, playing dumb, absenteeism, and theft. Like the modeling of strike occurrence, our use of the first step in QCA allows us to logically reduce configura- tions by removing redundant and non-neces- sary configurational components. We then rely on quantitative comparative techniques to delin- eate significant associations for cases within a given configuration relative to those that fall out- side of the configuration. For reasons of brevi- ty, our results here report only the configuration, the percent of cases in the configuration that experienced the particular form of resistance, and (in parentheses) the ratio of the configura- tion mean relative to the mean of cases not in the configuration.

Patterns for informal resistance indicate con- siderable similarity with those pertaining to strike prevalence, particularly in the configu- rations toward the top of the table. One pattern that differs significantly relative to strikes, how- ever, has to do with the seventh (conflict * goodorg * hist) and eighth (bureau * goodorg

* hist) configurations toward the bottom of the table. Prior results suggested that these config- urational patterns reduce, on average, the like- lihood of collective worker resistance in the form of strike action. Here, however, we see that lack of collective mobilization history com- bined with poor organization and a lack of sys- tematic conflict leads to more individualized forms of resistance, with relative consistency across our indicators. Workplaces character- ized by little bureaucracy, poor organization, and no collective mobilization history, for instance, are nearly twice as likely to experience social sabotage (1.83), playing dumb (2.19, absenteeism (1.86), and theft (1.86). They are also about 30 percent more likely to suffer from high levels of work avoidance by their employ- ees (1.32). A generally similar pattern holds for poorly organized workplaces, without conflict or strike history, with the exception of absen- teeism. Four of the associations across the two configurations being described reach statistical significance. Unlike conflict-ridden union con- texts, where individual resistance can be inter- preted more easily as being directed against management, workers in unorganized work- places may be venting their more general frus- trations toward the disorganization of the labor process itself.

The amplified patterns of individual resist- ance in poorly organized, nonbureaucratized work settings can be attributed to lack of orga- nizational control over workers' time, move- ment, and interactions, and a sense of hstration over the poor organization of work that employ- ees may experience daily in their work lives. Combined with a lack of collective response his- tory, one can envision a scenario where these other forms of resistance become workers' only recourse. Such possibilities are exemplified in the following passage, where workers in a poor- ly run and organized workplace engage in social sabotage and disparagement of managers in their free time.

We spent the better portion of the next two weeks slurpin' bad coffee, chain-smoking and concoct- ing this ridiculous idea for a short film. . . . Even though the line was only budging out about one truck per hour, we were required to stay put near our jobs. . . . The movie we were discussing was to be a violent blue-collar docudrama called No Need for a Grievance Procedure. It would be a collection of short pieces that chronicled the systematic execu- tions of our least favorite shoplords.. . . In one scenario Dave [one of the workers] snuck up to the glassed-in office where Henry Jackson [a hated foreman] sat and shrewdly welded the door shut from the outside. Dave then comman- deered an abandoned fork tmck and bore down on the tiny office, scooping it right off the factory floor. Raising the office several feet above the ground, he jostled the cubicle violently. Inside the office, Henry Jackson was slammed to and fro like a mannequin in a cyclone. (Hamper 1991:125)

Conversely, it may be the case that well- organized establishments keep workers tied to particular work stations and limit their capaci- ty to engage in social sabotage, theft, or work avoidance. Take, for instance, this account of an assembly line worker's job in an automated steel mill.

I have to watch my two strippers and make sure bar and pipe get kicked into the right trough for the stripper.I have to watch the conveyor down to the reheat furnace to make sure we don't pile up or let a pipe hit the bumper. And sometimes if the bar doesn't strip, it will go down into the furnace. I also watch the inserter. . . . There just isn't a chance to talk or walk around; I have to keep watch all the time (Walker 1977:32;emphasis in original).

The remaining configurations speak some- what to these possibilities, although the pat- terns are mixed. Bureaucratized workplaces characterized by conflict and the presence of a union (CONFLICT * UNION * BUREAU) experience relatively high levels of absen- teeism (1.71) and work avoidance (1.22), but average social sabotage (1.05) and lower lev- els of theft (.51) and playing dumb (.36). It seems that in such contexts, previously shown to be positively associated with strikes, more extreme forms of individual resistance may be controlled by bureaucracy or siphoned into collective expression. Remaining individual options for workers include general avoidance of work tasks or absence from the workplace altogether.

Mixed patterns are also observed in our third configuration (CONFLICT * UNION * HIST * GOODORG * abuse), which likewise couples conflict and unions with specific orga- nizational features (and also strike history and little individualized abuse). Within workplaces characterized by these attributes, social sabo- tage, work avoidance, and playing dumb occur less than in other workplaces, while absen- teeism and theft are more pronounced. Interestingly, the likelihood of social sabotage and work avoidance mounts when poor organ- ization prevails and when conflict is coupled with individualized abuse by supervisors (CONFLICT * UNION * ABUSE * goodorg). Work avoidance and reprisals in response to earlier conflict with managers are evidenced in the following worker interaction, which occurs at a bar near the factory, during the shift's din- ner break:

"What happens if we just don't go back on time?
They can dock us, but I bet they wouldn't fire us
all."
"Couldn't even start the line," Alfred says. "We got
the briner, the capper, and the pasteurizer all right
here. Right?"

Carl then points out that he is assigned as backup to all three machines. We think about it for a moment. Then Johnson T.turns and yells, "Hey, at the bar-send us another pitcher." It is the best beer I will drink all year. (Turner 1980:57-58)

Patterns, both similar and varying, among the first, second, and third configurations sug- gest that individual resistance strategies prima- rily involve direct response to managers (social sabotage), response to managerial orders (avoid- ance), or escape from the situation entirely (absenteeism). Such responses are most likely to occur when there is also some solidarity in the form of union presence, limited bureau- cratic control, or simply poor organization on the shop floor. And, as the configurations toward the bottom of the table (particularly the seventh and eighth) suggest, poor organization and lit- tle bureaucracy in and of themselves, even with- out conflict, can open the door to a plethora of individual resistance strategies.

CONCLUSION

Sociological theorizing on worker resistance and class action has highlighted the importance of workplace organizational structure and, to a lesser extent, social relations on the job. Moreover, prior analyses have been concerned largely with collective rather than individual resistance strategies that workers may use. Such modeling often downplays the importance and possibility of agency on the shop floor-agency that may be exercised by managers in their daily conduct and that is often used by workers as they combat the harshness of their jobs on a day-today basis. This article has addressed these issues, along with limitations in the literature, by (1) offering a broader theoretical conception of resistance, (2) simultaneously considering orga- nizational and interpersonal contexts within which workers are embedded, and (3) doing so with a configurational and case-oriented logic that also considers the potentially conditioning role of union presence and collective action legacy.

Our analyses, which use a unique integration of QCA reduction techniques and quantitative comparative methods, reveal interesting orga- nizational configurations-configurations that appear to vary in their associations with both collective and individual resistance. Findings also suggest three broad clusters of workplace types, defined by both their configurational commonalities and the relations of those com- monalities to the various forms of resistance we examine.14 Contentious workplaces, captured by the first, second and third configurations report- ed in our results, are characterized by system- atic and ongoing interpersonal conflict with management on the shop floor, worker solidar- ity and resources in the form of union pres- ence, and sometimes, bureaucratic structure. Cohesive workplaces (captured by the fourth and fifth reported configurations), are similar- ly characterized by worker representation, but with limited or little conflictlabuse. Finally, unorganized workplaces, which are apparent in the remaining configurations toward the bottom of Tables 5 and 7, are workplaces with gener- ally poor if not chaotic organization on the shop floor and little worker representation either con- temporarily or historically.

Each of the three workplace typologies- contentious, cohesive, and unorganized-as we suggest below, hold unique implications for understanding and interpreting worker resist- ance. Although our construction of "types" and our conclusions about them are tempered by the potential representativeness of the data we analyze, the considerable variation and repre- sentativeness of these data across lines of indus- try, organizational structure, and workplace interpersonal dynamics provide reasonable con- fidence. Moreover, the unique data we use and the methods employed have allowed us sys- tematically to examine important questions and issues that neither large organizational survey

l4 Others in the literature have similarly generat- ed typologies of organizations as theoretical framing tools for helping to understand workplace change, organizational variations, and their implications for workers. Hodson (2001b) for instance, denotes the importance of disorganized, unilateral, andpartici- pative organizations, while Adler and Borys (1996) distinguish between enabling and coercive bureaucracies. While we find these quite useful especially for thinking about the organization of work and vari- ations in bureaucratic potential, we believe the typolo- gies of workplaces we define here, generated from modeling of both organizational and interpersonal processes in relation to resistance, better incorporate the relative importance of social-interactional dynam- ics.

THE FOUNDATIONS OF WORKER RESISTANCE

designs nor single ethnographic analyses can address alone.

Within contentious workplaces, the co-joined attributes of manager-worker conflict and union presence (along with bureaucracy or poor orga- nization) appear to have simultaneous and strong associations with both collective resistance in the form of strike action and certain individual resistance strategies, most notably social sabo- tage, work avoidance, and absenteeism. Conflict in conjunction with union presence and bureau- cracy, in fact, appears to increase the likelihood of strike action more than sixfold. Strikes are more than four times more likely where conflict and union presence exist, where abuse and poor organization prevail, and where bureaucratic organization is weak. In both scenarios, we find higher-than-average levels of sabotage, work avoidance, and absenteeism, and lower-than- average levels of playing dumban interesting contrast perhaps related to the presence of unions and the security and confidence it affords to workers in what are obviously volatile con- texts. Clearly, collective and more individualized forms of resistance are not mutually exclusive in terms of their emergence or the factors that drive them.

One important insight that emerges in light of these results is that social relations on the shop floor play a meaningful role in prompting both collective and individual manifestations of class resistance. Here, the importance and fluidity of social relations with management as well as workers' abilities to respond to those relations is made quite explicit. Especially important is sys- tematic and ongoing conflict with managers, which seems to be associated with a lack of man- agement competence and appears to prompt both striking and certain individual resistance behav- iors. To interpret these relations as simply con- sistent with traditional social relations perspectives on worker satisfaction or action, however, would be problematic. Conflict's role in prompting resistance behaviors, as we suggest- ed in our earlier theoretical development and as confirmed by our results, can only be interpret- ed or understood within a particular organiza- tional, union, and historical context.

Bureaucracy appears to have little effect on col- lective resistance in and of itself; a fact that runs counter to well-known and commonly accepted assumptions within the workplace organization tradition. Even in the first configuration, where bureaucracy appears, it is its joint presence with conflict and union presence that magnifies both strike potential and individual resistance aimed particularly at managers. An4 it is within unor- ganized workplaces that a plethora of more indi- vidualized resistance strategies also seem to be realized. As our final table reports, all forms of individual resistance are magnified in such a context, although configuration differences are only statistically distinct for two individual resist- ance outcomes. Such findings seem to suggest that it is not necessarily organizational constraint that matters, as is typically assumed, or even only conflict with managers to which workers may respond, as in the case of contentious work- places. Rather, doing one's job in a disorganized workplace environment invokes stress and uncer- tainty, to which individual workers also clearly react. Individual resistance may thus appear in both contentious and unorganized workplaces, although the processes that invoke it seem to be quite distinct.

Beyond the analysis of organizational struc- tural, social relations, and their conditional effects on resistance, our findings also reveal some noteworthy patterns pertaining to union presence and strike history. Union presence appears in five of the reported configurations. In contentious workplaces, where unions coin- cide with conflict on the shop floor, associations with strike action are positive. Notably, two of these associations remain robust using tradi- tional logistic techniques and with the inclusion of potential influential controls. And, as noted previously, certain forms of individual resistance (i.e., social sabotage, work avoidance, and absenteeism) are more prevalent as well in these workplaces. In contrast to these patterns are those pertaining to cohesive workplaces, with- in which there is similarly union presence but with limited or nonexistent interpersonal con- flict on the shop floor. Here, associations with strike activity are actually negative. This sug- gests to us a quite contingent effect of unions, certainly with regard to worker collective action. Others have made the case, and we concur, that the effect of unions will vary geographically and historically depending upon the mobilization campaign, union strategy, elite response, and internal worker divides (e.g., Brown and Boswell 1995; Brueggemann and Boswell 1998). We extend this contingency argument by suggesting that union influence will also be fundamentally tied to the lived experiences of conflict on the shop floor (Fantasia 1988).

Unions, in and of themselves, can certainly bring resources to bear on organizing and mobi- lizing workers. Yet, if a union's appeal does not resonate with the organizational and interper- sonal realities of the shop floor, as experienced by workers themselves, union potential for mobi- lizing workers will be mitigated. And, it would be erroneous to assume that union presence has causal priority in this process. It may very well be the case, as suggested in recent historical analyses of worker mobilization campaigns, that shop floor conflict and resulting critical con- sciousness on the part of workers is a precursor to effective union involvement (e.g., Kimeldorf 1999; Roscigno and Danaher 2004; Rubin, Griffin, and Wallace 1983). Our results, while admittedly limited in their ability to uncover dynamic aspects of these situations, suggest rather straightforwardly that collective action is most likely to emerge in situations where shop floor experiences and union presence align.

Union presence and a historical legacy of collective action are associated with individual resistance strategies as well, and in informative ways. As was the case with strike occurrence, union presence coupled with conflict tends to be positively associated with individual resist- ance, particularly social sabotage, work avoid- ance, and absenteeism. This pattern holds, perhaps even more so, in poorly run, unionized workplaces without significant conflict or abuse. Clearly, the identity dynamics that unions and union membership may afford workers may be playing a role in fostering these associations. Even more notable and consistent, however, is individual resistance within unorganized work- places, including those that have no apparent strike history. Given the general lack of union presence and collective action in such contexts, it makes sense that informal resistance will be at the core of workers' protest repertoires. Again, rather than such resistance being solely an effort to regain dignity in the face of personal insult and conflict with managers, resistance in such settings may be as much a function of frustra- tion, boredom, and personal stress resulting from organizational chaos.

Both organizational structure and social rela- tions in the workplace are meaninghl for work- er resistance, but almost always in an interwoven manner. Moreover, the interplay of these work- place attributes will be conditioned by union presence and collective mobilization history. Differences in organizational structure and orga- nizational structural constraint may open the window to collective or individual forms of resistance. However, ultimately, how workers respond also will be conditioned by more prox- imate and systemic interpersonal conflicts with managers and by resulting worker interpreta- tions of the problems they are facing. In this regard, union presence and a history of collec- tive action may afford workers a behavioral repertoire that is more, although not exclusive- ly, collective in form.

The challenge for future research, given the arguments and findings presented in this arti- cle, lies in acknowledging some of the vari- ability and flexibility in how managers interact with workers under varying organizational struc-

Aupendix: Table 1. Truth Table for Configurations

tures, and the grievance outlets available to workers. Crucial as well will be the broader conceptions of resistance-conceptions that systematically include the possibility of work- er agency and that theoretically incorporate the broader tool kit of resistance strategies at col- lective and individual levels that workers may, and indeed do, draw upon. Future efforts to delineate relevant processes, configurations of organizational characteristics and social rela- tions, and worker actions across union, histor- ical, or geographic contexts will no doubt require a blending of case study insights and comparative analytic techniques.

Explanatory Measures   Distribution of Resistance Measures within Each Configuration
                Social Work Play    
B G C A U H Cases (n) Strike Sabotage Avoidance Dumb Absenteeism Theft

B = Bureaucracy; G = Good Organization; C = Conflict; A =Abuse; U = Union Presence; H = Strike History.

Vincent J. Roscigno is Associate Professor of Sociology at Ohio State University. His main fields of interest are stratification, sociology of education, social movements, and labor: Some of his more his- torical labor work, with William Danaher, waspub- lished recently in American Sociological Review (2001) and is also forthcoming in The Voice of Southern Labor: Radio, Music, and Textile Strikes, 1929-1 934 (Forthcoming, University of Minnesota Press). Other recent mobilization work, with Marc Dixon, focuses on contemporary unionization and strikes (American Journal of Sociology 2003). Along with his interest in social movements and labor, Roscigno remains involved in several projects per- taining to education, stratification, and work. He also is currently coordinating a large quantitative and qualitative project on employment and housing dis- crimination, drawing from archived Civil Rights Commission files.

Randy Hodson is Professor of Sociology at Ohio State University. His research interests include work- er citizenship and resistance, management behavior, and coworker relations. He is also engaged in research on economic transformations in Eastern Europe and China. His recent books include Dignity at Work (2001, Cambridge) and Worlds of Work: Building an International Sociology ofwork (coauthored with Daniel B. Cornfield, 2002, Kluwer/Plenum). He is also coauthor with Teresa A. Sullivan ofThe Social Organization of Work, 3rd edi- tion (2001, Wadsworth) and editor ofthe JAI/Elsevier Science annual series on Research in the Sociology of Work. For more information see his web site (http://www.soc . sbs. ohio-state. edu/rdh/).

REFERENCES

Abbott, Andrew. 1995. "Sequence Analysis: New Methods for Old Ideas." Annual Review of Sociology 21:93-113.

. 2001. "Review of Fuzzy-Set Social Science, by Charles C. Ragin." Contemporary Sociology 30:330-1.

Adler, Paul S. and Bryan Borys. 1996. "Two Types of Bureaucracy: Enabling and Coercive." Administrative Science Quarterly 42:61-89.

Amenta, E. and D. Halfmann. 2000. Wage Wars: Institutional Politics, WPA Wages, and the Struggle for U.S. Social Policy." American Sociological Review 65:506-28.

Appelbaum, Eileen R. and Rosemary Batt. 1994. The New American Workplace. Ithaca, NY: Industrial and Labor Relations Press.

Applebaum, Herbert. 198 1. Royal Blue: The Culture of Construction Workers. New York: Holt.

Ashenfelter, 0.and G.E. Johnson. 1969. "Bargaining Theory, Trade Unions, and Industrial Strike Activity." American Economic Review 59:35-49.

Barker, James R. 1999. The Discipline of Teamwork: Participation and Concertive Control. Thousand Oaks, CA: Sage.

Barnard, Chester I. 1950. The Functions of the Executive. Cambridge, MA: Harvard University Press.

Bass, Bernard M. 1985.Leadership and Performance Beyond Expectations. New York: Free Press.

Benyon, Huw. 1972. Perceptions of Work: Variations Within a Factory. London, England: Cambridge University Press. . 1975. Working for Ford. East Ardsley, England: E.P. Publishing. Blauner, Robert. 1964. Alienation and Freedom. Chicago, IL: University of Chicago. Blee, K. and D. B. Billings. 1986. "Reconstructing Daily Life in the Past: An Hermeneutical Approach to Ethnographic Data." Sociological Quarterly 27:443-62. Bonacich, Edna. 1972. "A Theory of Ethnic Antagonism: The Split Labor Market." American Sociological Review 37:547-59. Boswell, Terry and Cliff Brown. 1999. "The Scope of General Theory: Methods for Linking Inductive and Deductive Comparative History." Sociological Methods and Research 28: 1 54-85. Braverman, Harry. 1974. Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century.New York: Monthly Review.

Brecher, Jeremy. 1972. Strike! San Francisco, CA: Straight Arrow Books.

Brody, David. 1960. Steelworkers in America: The Nonunion Era. Cambridge, MA: Harvard University Press.

Brown, Cliff and Terry Boswell. 1995. "Strikebreaking Solidarity: The Great Steel Strike of 191 9." American Journal of Sociology 100:1479-519.

Brueggemann, John and Terry Boswell. 1998. "Realizing Solidarity: Sources of Interracial Unionism During the Great Depression." Work and Occupations 25:43&82.

Burawoy, Michael. 1979. Manufacturing Consent. Chicago, IL: University of Chicago Press. . 1985. The Politics ofProduction. London, England: Verso. Bureau of Labor Statistics. 2001. "Major Work Stoppages in 2000." News Release, February 9.

Burris, Beverly H. 1983. No Room at the Top: Underemployment and Alienation in the Corporation. New York: Praeger.

Cavendish, Ruth. 1982. Women on the Line. Boston, MA: Routledge and Kegan Paul.

Cheng, Y. and Arne L. Kalleberg. 1996. "Employee Job Performance in Britain and the United States." Sociology 30: 1 15-29.

Cole, G.D.H. 1918. "Recent Developments in the British Labor Movement." The American Economic Review 8:485-504.

Collins, Randall. 198 1. Sociology Since Midcentury:

Essays in Theory Cumulation. New York:

Academic Press.

Cornfield, Daniel B. and Hyunhee Kim. 1994. "Socioeconomic Status and Unionization Attitudes in the United States." Social Forces 73:521-32.

Cotgrove, Stephen and Clive Vamplew. 1972. "Technology, Class and Politics: The Case of the Process Workers." Sociology 6: 169-85.

Coverdill, James E., William Finlay, and Jack K. Martin. 1994. "Labor Management in the Southern Textile Industry." Sociological Methods and Research 235-85.

Cress, Daniel M. and David A. Snow. 2000. "The Outcomes of Homeless Mobilization." American Journal of Sociology 105: 1063-104. Crozier, Michel. 1964. The Bureaucratic Phenomenon. Chicago, IL: University of Chicago Press. Della Fave, L. Richard. 1980. "The Meek Shall Not Inherit the Earth." American Sociological Review 45:955-71. . 1986. "Toward and Explication of the Legitimation Process." Social Forces 65:47&500. Dixon, Marc and Vincent J. Roscigno. 2003. "Status, Networks, and Social Movement Participation: The Case of Striking Workers." American Journal of Sociology 108: 1292-1327. Dunlop, John T. 1958. Industrial Relations Systems. Carbondale, IL: Southern Illinois University Press. Durkheim, Emile. [I8931 1984. The Division of Labor in Society. New York: The Free Press. Ebbinghaus, Bernhard and Jelle Visser. 1999. "When Institutions Matter: Union Growth and Decline in Western Europe, 1950-95." European Sociological Review 15:135-58. Edwards, Paul K. 1978. "Time Series Regression Models of Strike Activity: A Reconsideration with American Data." British Journal of Industrial Relations 16:320-35. . 1992. "Industrial Conflict: Themes and Issues in Recent Research." British Journal of Industrial Relations 30:361404. . 1996. "The Struggle for Control of the American Workplace 1920-1985." Pp. 150-203 in The Nature of Industrialization, edited by Peter Mathies and John A. Davis. Oxford, England: Blackwell. Edwards, Paul K. and Richard Hyman. 1994. "Strikes and Industrial Conflict: Peace in Europe?" Pp. 250-80 in New Frontiers in European Industrial Relations, R. Hyman and A. Ferner. Oxford, England: Blackwell. Edwards, Richard. 1979. Contested Terrain. New York: Basic. Fantasia, Rick. 1988. Cultures of Solidarity: Consciousness, Action, and Contemporary American Workers. Berkeley, CA: University of California Press. Finlay, William. 1988. Work on the Wate@ont: Worker

THE FOUNDATIONS OF WORKER RESISTANCE

Power and Technological Change in a West Coast Port. Philadelphia, PA: Temple University Press. Foster, Charles. 1969. Building with Men. London, England: Tavistock. Foucault, Michel. 1988. Politics, Philosophy, Culture. New York: Routledge.

Gamson, William A. 1995. "Constructing Social Protest." Pp. 85-106 in Social Movements and Culture edited by Hank Johnston and Bert Klandermans. Minneapolis, MN: University of Minnesota Press.

Giddens, Anthony. 1982. "Class Structuration and Class Consciousness." Pp. 157-74 in Classes, Power, and Conflict: Classical and Contemporary Debates, edited by A. Giddens and D. Held. Berkeley, CA: University of California Press.

. 1984. The Constitution of Society. Cambridge, England: Cambridge University Press. Gouldner, Alwin. 1964. Patterns of Industrial

Bureaucracy. New York: Free Press. Graham, Laurie. 1995. On the Line at Subaru-Isuzu. Ithaca, NY: Industrial and Labor Relations Press.

Griffin, Larry J., Christopher Botsko, Ana-Maria Wahl, and Larry Isaac. 1991. "Theoretical Generality, Case Particularity: Qualitative Comparative Analysis of Trade Union Growth and Decline." International Journal of Comparative Sociology 32: 110-36.

Griffin, Larry J., Holly J. McCammon, and Christopher Botsko. 1990. "The 'Unmaking' of a Movement: The Crisis of U.S. Trade Unions in Comparative Perspective." Pp 169-1 74 in Changes in Societal Institutions, edited by M.T. Hallinan,

D.M. Klein, and J. Glass. New York: Plenum.

Griffin, Larry, Michael Wallace, and Beth Rubin. 1986. "Capitalist Resistance to the Organization of Labor Before the New Deal: Why? How? Success?" American Sociological Review 51:147-67.

Halaby, Charles N. 1986. "Worker Attachment and Workplace Authority." American Sociological Review 51:634-49.

Halle, David. 1985. America $ Working Man. Chicago, IL: University of Chicago Press.

Hammersley, M. 1997. "Qualitative Data Archiving: Some Reflections on Its Prospects and Problems." Sociology 31:13 142.

Hamper, Ben. 1991. Rivethead: Tales from the Assembly Line. New York: Warner. Harris, Rosemary. 1987. Power and Powerlessness in Industry. London, England: Tavistock.

Hicks, Alexander. 1994. "Qualitative Comparative Analysis and Analytic Induction: The Case of the Emergence of the Social Security State." Sociological Methods & Research 23:8&113.

Hodson, Randy. 1996. "Dignity in the Workplace Under Participative Management: Alienation and Freedom Revisited." American Sociological Review 61: 719-38.

. 1999. "Management Citizenship Behavior:

A New Concept and an Empirical Test." Social Problems 46:460-78. . 2001a. Dignity at Work. New York: Cambridge University Press.

. 2001b. "Disorganized, Unilateral, and Participative Organizations: New Insights from the Ethnographic Literature." Industrial Relations 40:204-30.

Isaac, Larry W. and Larry J. Griffin. 1989. "Ahis- toricism in Time-Series Analyses of Historical Process: Critique, Redirection, and Illustrations from U.S. Labor History." American Sociological Review 54:873-90.

Isaac, Larry and Lars Christiansen. 2002. "How the Civil Rights Movement Revitalized Labor Militancy." American Sociological Review 67:72246.

Jackall, Robert. 1978. Workers in a Labyrinth: Jobs and Survival in a Bank Bureaucracy. Montclair, NJ: Allanheld and Osmun.

Jenkins, J. Craig and Craig M. Eckert. 1986. "Channeling Black Insurgency: Elite Patronage and Professional Social Movement Organizations in the Development of the Black Movement." American Sociological Review 5 1 :8 12-29.

Jenkins, J. Craig and Charles Perrow. 1977. "Insurgency of the Powerless: Farm Worker Movements (1946-1972)" American Sociological Review 53:191-208.

Jermier, John M. 1988. "Sabotage at Work." Pp. 101-135 in Research in the Sociology of Organizations, edited by Nancy DiTomaso. Greenwich, CN: JAI Press.

Jermier, John, David Knights, and Walter Nord. 1994. Resistance and Power in Organizations. London, England: Routledge.

Juravich, Tom. 1985. Chaos on the Shop Floor: A Worker S View of Quality, Productivity, and Management.Philadelphia, PA: Temple University Press.

Kelly, John and Caroline Kelly. 199 1. "'Them and Us': Social Psychology and 'The New Industrial Relations.'" British Journal ofIndustria1 Relations 29:2548.

Kimeldorf, Howard. 1999. Battling for American Labor: Wobblies, Craft Workers, and the Making of the Union Movement. Berkeley, CA: University of California Press.

Klein, D., M. Hallinan, and J. Glass, eds. 1990. Change in Societal Institutions. New York: Plenum.

Kunda, Gideon. 1992. Engineering Culture: Control and Commitment in a High-Tech Corporation. Philadelphia, PA: Temple University Press.

Lawler, Edward, Cecilia Ridgeway, and Barry Markovsky. 1993. "Structural Social Psychology and the Micro-Macro Problem." Sociological Theory 11:268-90.

Lee, Ching Kwan. 1998. Gender and the South China Miracle: Two Worlds ofFactory Women. Berkeley: University of California Press.

Lee, T. W. 1999. Using Qualitative Methods in Organizational Research. Thousand Oaks, CA: Sage.

Letwin, Daniel. 1998. The Challenge of Interracial Unionism: Alabama Coal Miners, 1878-1921. Chapel Hill, NC: University of North Carolina Press.

Lieberson, Stanley. 2001. "Review of Fuzzy-Set Social Science, by Charles C. Ragin." Contemporay Sociology 30:33 14.

Low-Beer, John. 1978. Protest and Participation: The New Working Class in Italy. New York: Cambridge.

Mann, Michael. 1973. Consciousness and Action Among the Western Working Class. London, England: Macmillan.

Marx, Karl. 197 1. "The Economic and Philosophic Manuscripts of 1844." In Karl Marx, edited by

Z.A. Jordan. London, England: Michael Joseph. . [18871 1967. Capital, Volume I. New York: International Publishers.

Mayer, Roger C., James H. Davis, and F. David Schoorman. 1995. "An Integrative Model of Organizational Trust." Academy of Management Review 20:709-34.

McCammon, Holly J. 1993. "From Repressive Intervention to Integrative Prevention: The U.S. State's Legal Management of Labor Militancy, 1881-1978." Social Forces 71569-601. . 1994. "Disorganizing and Reorganizing Conflict: Outcomes of the State's Legal Regulation of the Strike Since the Wagner Act." Social Forces 72:101149.

Minkoff, Debra C. 1997. "The Sequencing of Social Movements." American Sociological Review 62:779-99.

Molstad, Clark. 1986. "Choosing and Coping with Boring Work." Urban Life 15:2 15-36.

Montgomery, David. 1979. Workers' Control in America: Studies in the History of Work, Technology, and Labor Struggles. New York: Cambridge University Press.

Moore, W.E. 1962. The Conduct of the Corporation. New York: Random House.

Morrill, Calvin. 1995. The Executive Way: Conflict Management in Corporations. Chicago, IL: University of Chicago Press.

Obershcall, Anthony R. 1993. Social Movements: Ideologies, Interests, and Identities. New Brunswick, NJ: Transaction.

Orr, Julian. 1996. Talking about Machines: An Ethnography ofa Modern Job. NewYork: Cornell University Press.

Paules, Greta Foff. 199 1. Dishing it Out: Power and Resistance among Waitresses in a New Jersey Restaurant. Philadelphia, PA: Temple University Press.

Pfeffer, Jeffrey. 1998. The Human Equation: Building Profits by Putting People First. Boston, MA: Harvard Business School Press.

Pichardo, Nelson A. 1995. "The Power Elite and Elite-Driven Countermovements: The Associated Farmers of California During the 1930s." Sociological Forum 10:2149.

Pierce, Jennifer L. 1995. Gender Trials: Emotional Lives in Contemporary Law Firms. Berkeley: University of California Press.

Pierson, Paul. 2000. "Increasing Returns, Path Dependence, and the Study of Politics." American Political Science Review 94:25 1-67.

. 2001. New Politics of the We'elfare State. Oxford, England: Oxford University Press. Ragin, Charles. 1987. The Comparative Method. Berkeley, CA: University of California Press. . 2000. Ragin, C. C. 2000. Fuzzy-Set Social Science. Chicago, IL: University of Chicago Press.

Redding, Kent and Jocelyn S. Viterna. 1999. "Political Demands, Political Opportunities: Explaining the Differential Success of Left-Libertarian Parties." Social Forces 78:491-510.

Roethlisberger, F. J. and William J. Dickson. 1939. Management and the Worker. Cambridge, MA: Harvard University Press.

Roscigno, V J. and W. Danaher. 2001. "Media and Mobilization: The Case of Radio and Southern Textile Worker Insurgency, 1929-1 934." American Sociological Review 66:2 148. . 2004. The Voice of Southern Labor: Radio, Music, and Textile Strikes, 1929-1934. Minneapolis, MN: University of Minnesota Press. Rubin, Beth A. 1986. "Class Struggle American Style: Unions, Strikes, and Wages." American Sociological Review 51:6 18-3 1. Rubin, Beth A., Larry J. Griffin, and Michael Wallace. 1983. "'Provided Only That Their Voice Was Strong:' Insurgency and Organization of American Labor from NRA to Taft-Hartley." Work and Occupations 10:32547. Schwartzman, H. B. 1993. Ethnography in Organizations. Qualitative Research Methods Series #27. Newbury Park, CA: Sage. Scott, James C. 1985. Weapons of the Weak. New Haven, CT: Tale University Press. Scruggs, L. and I?Lange. 2002. "Where Have All the Members Gone? Globalization, Institutions, and Union Density." Journal ofPolitics 64: 126-53. Seider, Maynard, 1984. A Year in the Life of a Factory. San Pedro, CA: Singlejack Books. Sennett, Richard. 1998. The Corrosion of Character: The Personal Consequences of Work in the New Capitalism. New York: Norton. Shalev, Michael. 1992. "The Resurgence of Labour Quiescence." Pp. 102-32 in The Future ofLabour Movements, edited by M. Regini. London, England: Sage. Shin, Gi-Wook. 1994. "The Historical Making of Collective Action: The Korean Peasant Uprisings of 1946." American Journal of Sociology 99:1596-624. Smith, Vicki. 2001. Crossing the Great Divide:

THE FOUNDATIONS OF WORKER RESISTANCE

Worker Risk and Opportunity in the New Economy.

Ithaca, NY Cornell University Press.

Snow, David A. and Robert D. Benford. 1992. "Master Frames and Cycles of Protest." Pp 133-55 in Frontiers of Social Movement Theory, edited by Aldon Morris and Carol Mueller. New Haven, CT: Yale.

Snyder, D. 1975. "Institutional Setting and Industrial Conflict: A Comparative Analysis of France, Italy, and the United States." American Sociological Review 40:259-78.

Staggenborg, Suzanne. 1989. "Organizational and Environmental Influences on the Development of the Pro-Choice Movement." Social Forces 68:20440.

Tarrow, Sidney G. 1998. Power in Movement: Social Movements and Contentious Politics. New York: Cambridge University Press.

Taylor, Frederick W. 19 1 1. The Principles of Scientific Management. New York: Harper and Row.

Taylor, Verta. 1989. "Social Movement Continuity: The Women's Movement in Abeyance." American Sociological Review 54:76 1-75.

Turner, Steve. 1980. Night Shift in a Pickle Factory. San Pedro, CA: Singlejack Books.

Vallas, Steven P. 1987. "The Labor Process as a Source of Class Consciousness: A Critical Examination." Sociological Forum 2:237-56.

. 2003. "The Adventures of Managerial Hegemony: Teamwork, Ideology, and Worker Resistance." Social Problems 50:204-25.

Van Maanen, J., ed. 1998. Qualitative Studies in Organizations. Thousand Oaks, CA: Sage.

Walker, Charles R. [I9571 1977. Toward the Automatic Factory: A Case Study of Men and Machines. New Haven, CT: Yale University Press.

Weber, Max. [I9141 1968. Economy and Society. New York: Bedminster.

Weber, R. P. 1990. Basic Content Analysis, 2d ed. Quantitative Applications in the Social Sciences #49. Newbury Park, CA: Sage.

Wedderburn, Dorothy. 1972. Workers 'Attitudes and Technology.London: Cambridge University Press. Westwood, Sallie. 1984. All Day Every Day: Factory and Family in the Making of Women's Lives.

London: Pluto.

Whitener, Ellen M., Susan E. Brodt, M. Audrey Korsgaard, and Jon M. Werner. 1998. "Managers as Initiators of Trust: An Exchange Relationship Framework for Understanding Managerial Trustworthy Behavior. Academy of Management Review 23:513-30.

Wright, Erik Olin. 2002. "The Shadow of Exploitation in Weber's Class Analysis." American Sociological Review 675332-53.

Zingraff, Rhonda and Michael D. Schulman. 1984. "Social Bases of Class Consciousness: A Study of Southern Textile Workers with a Comparison by Race." Social Forces 63:98-116.

Comments
  • Recommend Us