The Relationship Between Task ... - Wiley Online Library

3 downloads 2870 Views 150KB Size Report
perceived amount and clarity of interdependency in managers' jobs affect role ... demands–control model as greater job control reduced role ambiguity when ...
Journal of Management Studies 44:2 March 2007 0022-2380

The Relationship Between Task Interdependency and Role Stress: A Revisit of the Job Demands–Control Model*

Sze-Sze Wong, Gerardine DeSanctis (deceased) and Nancy Staudenmayer (deceased) Nanyang Technological University, Singapore; Duke University, Durham, NC; Duke University, Durham, NC abstract Drawing from Karasek’s job demands–control model, this study investigated how perceived amount and clarity of interdependency in managers’ jobs affect role stress, and the extent to which job control moderates these relationships. Results show that amount of interdependency was positively associated with role conflict, and clarity of interdependency was negatively associated with role ambiguity. There was also support for the job demands–control model as greater job control reduced role ambiguity when clarity of interdependency was low. Although higher job control produced lower role ambiguity when both clarity and amount of interdependency were low, higher job control did not produce lower role ambiguity when clarity of interdependency was low and amount of interdependency was high, suggesting that the buffering value of job control on reducing role stress is contingent on the task interdependencies that managers confront.

INTRODUCTION Management of interdependencies constitutes a major activity of everyday work life. Research studies across a wide variety of professions indicate that people spend the majority of their work life working with and coordinating with others (Ancona and Caldwell, 1992; Eccles and Crane, 1988; Kellogg et al., 2006; Perlow, 1999; Wong, 2004). It is perhaps a paradox of our era that greater professionalization and flexibility in the workplace are coincidental with greater interdependence in the nature of work. Increasingly it is the case that in order to accomplish one’s work goals, coordination with others is required. Task interdependence is part and parcel of today’s organization, and forces such as accelerating globalization and geographical dispersion of work (Davidow and Malone, 1992; Ghoshal and Bartlett, 1990; Doh, 2005; Levy, 2005; Malone and Rockart, 1991), shorter product life cycles (Brown and Eisenhardt, 1997), greater knowlAddress for reprints: Sze-Sze Wong, Nanyang Business School, Nanyang Technological University, S3-B2B-47, Nanyang Avenue, Singapore 639798 ([email protected]). © Blackwell Publishing Ltd 2007. Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

The Job Demands–Control Model

285

edge intensity of organizational work (Davenport and Prusak, 1998) and networked forms of organizing (Lui and Ngo, 2005; Rubery et al., 2002) will no doubt further the interdependent nature of work. Research on the interdependent nature of work has focused primarily on the design of interdependencies and the impact of task interdependence on coordination flow and effectiveness (e.g. Crowston, 1997; Thompson, 1967; Wageman, 1995). Few studies have examined how characteristics of task interdependence in contemporary work settings affect the psychological states of managers. Although the study of job demands on job stress and strain is extensive in the organizational stress literature (e.g. De Croon et al., 2004; Karasek, 1979; McGrath, 1976), the majority of these studies have focused on task characteristics such as workload and pace of work (e.g. de Rijk et al., 1998; Fox et al., 1993; Van Yperen and Hagedoorn, 2003). Given that a substantial portion of managerial work involves communicating with interdependent parties (Bartlett and Ghoshal, 1987; Torrington and Weightman, 1987), it is surprising that so little attention has been directed towards examining how task interdependency affects managerial psychological health or the strategies that might be used to cope with task interdependency. There is reason to believe that the characteristics of task interdependency are having consequential effects on workers, especially job role stress. Organizations today are seen as webs or networks of lateral cooperative linkages (Ghoshal and Bartlett, 1990; Nohria, 1992). Advances in telecommunications technology are enabling globalization of companies’ operations and more fragmented organizational structures (DeSanctis et al., 1999; Fulk and DeSanctis, 1995; Rubery et al., 2002). Increasing complexity of products and services has come about with the emergence of cooperative alliances among organizations (Badaracco, 1991; Liebeskind et al., 1996). Collectively, these forces have led to an organizational workplace that is characterized more by a fluid and complex web of internal and external task interdependencies, and less by a static and clear blueprint of interdependencies. Managers today confront large numbers of interdependencies in their work, with less apparent clarity of work design, precisely when more coordination is needed (e.g. Hill, 2004; Kogut and Zander, 1992). Examples of interdependencies in modern work abound. A new product development manager has to consider with whom to consult and inform when there is a change in the product design. A software programming manager needs to know with whom and when to coordinate regarding updates and fixes to the specific software modules. A financial analyst has to select among sources when building a forecast model for a bond fund. A human resource manager must decide with whom to coordinate in various business units to enhance the success of a human resource policy. A marketing manager needs to know whom to seek and send specific kinds of information about market and customer trends. These examples underlie our central question: What are the psychological effects of managing task interdependencies – a central component of managerial work – on managers today? Interdependent work can no doubt bring benefits such as interpersonal stimulation and learning (see Jehn et al., 1999), but it may also imply costs in the sense that the psychological health of managers is hampered by the challenges of managing more interdependencies (e.g. Stamper and Johlke, 2003) whose structures are less clear. Drawing from Karasek’s (1979) job demands–control model, our research goal is to investigate how characteristics of task interdependency affect role stress of managers, and © Blackwell Publishing Ltd 2007

286

S-S. Wong et al.

how job control moderates the relationship between task interdependency and role stress. The remainder of our paper is organized as follows. First, we briefly review the literature on job stress and task interdependency, and present our conceptual model. Next, we describe our study’s methodology and present the results of our statistical analyses. We conclude with a discussion of the theoretical and practical implications of our findings for future research. THEORETICAL BACKGROUND AND FRAMEWORK Karasek’s (1979) job demands–control model has been an influential theoretical foundation for various studies of job stress (e.g. Cooper, 2000; Fox et al., 1993; Van Yperen and Hagedoorn, 2003). The theoretical argument underlying this model is that individual physiological strain results from the interactive effects of one’s job demands and the amount of job control available to one’s job. Specifically, the theory posits that in order to minimize physiological strain, job demands should be matched to job control such that when job demands are high, job control should correspondingly be high. High job control allows job incumbents to adapt to demands by developing appropriate behavioural response patterns. Job incumbents in demanding jobs who can proactively manage how their work is done can channel their energy in constructive ways, thus reducing the stress brought on by job challenges. Although Karasek’s original work focused on physiological strain, his model has since been extended to mental or psychological stress (e.g. De Croon et al., 2004; Fox et al., 1993; Xie, 1996). A large proportion of research on the job demands–control model has focused on nurses (e.g. de Rijk et al., 1998; Fox et al., 1993; Schaubboeck and Merritt, 2003; Van Yperen and Hagedoorn, 2003) and production workers (e.g. Parker and Sprigg, 1999; Wall et al., 1996). Some studies have supported the proposed interaction effect (e.g. Fox et al., 1993), and others have demonstrated no such effect (e.g. Landsbergis, 1988). As a result, some researchers in this area have adopted a contingency approach by investigating the extent to which the job demands–control relationship is moderated by individual-level characteristics such as self-efficacy (Schaubboeck and Merritt, 2003) and proactivity (Parker and Sprigg, 1999). However, it may be premature to abandon the job demands–control model for a model with more individual-level variables. It is plausible that in nursing and production jobs where incumbents are used to relatively high levels of formalization in management of their tasks, greater job control could be perceived as a burden rather than as a stress-reducing mechanism. Theories such as Holland’s theory of vocational choice (Holland, 1973, 1985) and Schneider’s theory of attraction–selection–attribution (Schneider, 1987) have suggested that individual differences are associated with occupations. For example, managers have been found to prefer job autonomy while nonmanagers prefer shorter working hours and higher job security (e.g. Savery, 1988). To the extent that there are differences in work demands and individual preferences among occupations, individuals in different occupations (e.g. managers vs. non-managers) may respond differently to greater job control. For instance, managers who are accustomed to greater job decision latitude may respond to greater job control in a way consistent with Karasek’s model. © Blackwell Publishing Ltd 2007

The Job Demands–Control Model

287

In addition, research on Karasek’s model has largely focused on job demands such as workload and work pace (e.g. de Rijk et al., 1998; Fox et al., 1993; Van Yperen and Hagedoorn, 2003). Interestingly, there have been no studies, to the best of our knowledge, that have applied the job demands–control model to the social nature of work demands, that is, job challenges arising from managing interdependencies with other people in the workplace. Task interdependency is the degree to which the accomplishment of a work assignment is determined, influenced, and/or controlled by another entity such as a person, organizational unit, or firm (Thompson, 1967; Van de Ven et al., 1976). Task interdependency places demands on job incumbents to coordinate with others in the delivery and/or receipt of resources such as materials or knowledge. Task interdependencies bring about pressures such as timely exchange of resources (Galbraith, 1974), managing cooperative social relations (Tjosvold, 1988), and effective conflict resolution (McCann and Ferry, 1979). Hence, it would be valuable to examine the extent to which Karasek’s model applies in situations where job demands consist of managing a high amount and/or low clarity of task interdependency; and the extent to which job control can be an effective stress-reducing mechanism in these situations. Managing task interdependency is at the heart of a manager’s job, as evidenced by the significant portion of time a manager spends in communicating and coordinating information to people internal and external to the firm (e.g. Mintzberg, 1971; Torrington and Weightman, 1987). As we argued earlier, managers confront increasing amounts of interdependency in the modern workplace, and the structure of these interdependencies is often unspecified. If managers perceive their interdependencies as high in volume and low in clarity, then we may expect them to experience heightened levels of role stress. Hence, extending Karasek’s model to examine task interdependency, we propose that perceived amount of and clarity of interdependence are likely to affect managers’ perceptions of role conflict and role ambiguity – two well established dimensions of role stress (House et al., 1983; Rizzo et al., 1970). Further, we can expect that this relationship is moderated by the degree of job control that managers perceive in their jobs. Task Interdependency and Role Stress Interdependencies vary in their amount as well as in their nature (Staudenmayer, 1997). The amount of interdependency is defined as the degree to which one relies on others to accomplish one’s work. Hence, perceived amount of interdependency can increase when the number of interdependent parties increases and/or the number of interdependency relationships with existing parties increases. The amount of interdependency that one perceives can be due to the inherent design of the technology of the task (Thompson, 1967) and/or the voluntary behaviours of oneself or other people (Shea and Guzzo, 1987). It is important to distinguish between objective and subjective amount of interdependency as some studies have shown that even individuals who predominantly perform their work independently experience varying levels of interdependence due to varying perceptions (Shea and Guzzo, 1987; Tjosvold, 1986; Wageman, 1995). In this study, we are focusing on perceived amount of interdependency as it is managers’ perceptions of their job demands that are likely to affect their perceptions of role stress. © Blackwell Publishing Ltd 2007

288

S-S. Wong et al.

Clarity of interdependency is defined as the degree to which an individual is certain about whom he or she is reliant on and for what purpose(s). Whereas traditional hierarchical structures and formalized job descriptions have led to relatively well-specified and clearly designed task interdependencies in the past, more fluid and global operations today have inadvertently led to less well defined working relationships, particularly for general managers. As a result, hidden interdependencies arise, making coordination more difficult (Staudenmayer, 1997). Exactly with whom a manager must interface in order to complete a task, or how that interaction is to be managed, is typically unspecified. Clarity of interdependency may emerge with experience, but the dynamic nature of work will tend to make even experienced managers sensitive to low clarity in task interdependency. Managers who confront low clarity of interdependency in their work are more likely to experience high stress. Amount and clarity of task interdependency are distinct dimensions. A manager in a software development project will perceive a high amount of interdependency, if the job requires coordination across a large number of software module developers, but these interdependencies may be well-specified if the different software modules and persons in charge are clearly designated. In such a setting, with whom one must coordinate is clear even though the amount of task interdependency is high. Alternatively, a financial analyst who provides investment forecasts to, say, an investment officer of a bank, may operate fairly independently, with few task interdependencies. However, with whom he must coordinate to obtain information may vary considerably over time. Hence, although the amount of task interdependency is low, the clarity of interdependency is also low. We propose that perceived amount and clarity of interdependency lead to different types of role stress. Role conflict is defined as the degree of incompatibility between the expectations of other parties or between aspects of a single role. Role ambiguity is defined as the degree of uncertainty associated with the goals and means to perform one’s role (Kahn et al., 1964). The likelihood of encountering conflicting expectations is greater when there is a larger set of interdependencies to coordinate, so we can expect managers to experience greater role conflict when they perceive higher amounts of interdependency in their jobs. Feelings of role conflict are likely to arise insofar as managers perceive that they have to struggle to reconcile opposing requirements from different interdependent parties. Clarity of interdependency presents a very different type of job challenge. When managers are unsure about with whom and what to coordinate, they are likely to experience ambiguity about the expectations of their role and how to fulfil their role responsibilities in the organization. Certainty about how to carry out one’s role is likely to require high clarity in one’s task requirements. To the extent that there are uncertainties about one’s job interdependencies (i.e. low clarity of interdependence), one is likely to perceive greater uncertainty about how to perform one’s role (i.e. higher role ambiguity). Thus, we can expect managers with lower clarity of interdependency to experience greater role ambiguity. Hypothesis 1: The higher the perceived amount of interdependency, the higher the role conflict. Hypothesis 2: The lower the perceived clarity of interdependency, the higher the role ambiguity. © Blackwell Publishing Ltd 2007

The Job Demands–Control Model

289

Job Control as Moderator of Task Interdependence – Role Stress Relationship Job control is defined as the extent to which one has the capacity to influence or manipulate the various facets of one’s job (Ganster and Fusilier, 1989). Job control has been demonstrated to be a stress-reducing strategy in demanding jobs where there are high workload and fast work pace as higher job control allows job incumbents to transform the arousal triggered by high job demands into constructive behaviours (Karasek, 1979; Van Yperen and Hagedoorn, 2003). Following this logic, we argue that perceived job control can reduce managerial role stress because managers are able to take proactive moves to address the challenges in their job. For instance, with greater job control managers can decide to reduce or eliminate less important interdependencies (so as to reduce role conflict) and/or make changes to their work methods (so as to reduce role ambiguity). In contrast, when managers perceive low job control, they may feel locked into a hopeless situation, with little latitude to manage the demands in their work environment. Hence, we hypothesize that greater perceived job control is likely to reduce the role stress triggered by task interdependence as managers perceive greater latitude to cope with the coordination demands. Hypothesis 3: When there is greater job control, a higher perceived amount of interdependency is less likely to be positively associated with role conflict. Hypothesis 4: When there is greater job control, lower perceived clarity of interdependency is less likely to be negatively associated with role ambiguity. Although greater job control is likely to be beneficial in providing managers with more autonomy to address job demands, its value as a stress-reducing mechanism is limited if job demands escalate to a level that strains job incumbents’ ability to constructively utilize the increased autonomy to meet job challenges. Karasek’s job demands–control prediction is built on the premise that incumbents in high demand jobs can develop constructive behaviours, when given increased job control to respond to these demands. However, to the extent that the job demands are so overwhelming that it becomes cognitively difficult for a manager to grasp what actions must be taken to manage these demands, having high decision latitude may be of little utility. Thus, we predict that when managers confront a high amount of interdependency and low clarity of interdependency, greater job control is less likely to be effective in reducing role stress. In this job situation cognitive demands are exacerbated and managers must make sense of a larger set of ambiguous interdependencies. Greater job control is less likely to be helpful because it is cognitively more difficult for the manager (even if he or she has high autonomy) to decide how to reduce conflicting expectations from interdependent parties due to high ambiguity about with whom and what to coordinate. Similarly, it would be cognitively difficult for the manager to redesign his or her work in ways that would clarify interdependencies when the large set of interdependencies constrains the extent to which the manager is free to change his or her work methods to reduce role ambiguity. In contrast, under less demanding task interdependency situations (i.e. low amount and © Blackwell Publishing Ltd 2007

290

S-S. Wong et al.

clarity of interdependency), higher job control is likely to be effective in reducing role stress because the manager is better able to constructively apply the higher latitude provided to make sense of interdependencies; there are greater degrees of freedom to restructure work relationships to reduce role conflict and role ambiguity when the manager is interdependent with a smaller set of parties. Hypothesis 5: The relationship between perceived amount of interdependency and role conflict will be stronger for the combination of low perceived clarity of interdependency and high job control and weaker for the combination of low perceived clarity of interdependency and low job control. Hypothesis 6: The relationship between perceived clarity of interdependency and role ambiguity will be stronger for the combination of high perceived amount of interdependency and high job control and weaker for the combination of high perceived amount of interdependency and low job control.

METHOD Sample and Data Gathering Procedure Executive MBA students who were enrolled in a weekend executive MBA programme at a US university were solicited to complete a survey on ‘Managing in an Interdependent World’, which we administered at the end of class sessions, all on the same day. Completion of the survey was anonymous, voluntary and unrelated to the course grade. The survey took about 20 minutes to complete. A total of 125 of 181 surveys were returned, thus generating a response rate of 69 per cent. We omitted 12 surveys due to missing or erroneous data, leaving a final analytical sample of 113 (effective response rate of 62 per cent). As respondents were enrolled in a weekend executive MBA programme, they were working full-time while attending the programme. Respondents were managers from a range of industries and had considerable organizational experience. Thirty-two per cent of the respondents worked in manufacturing companies, 16 per cent in chemicals and pharmaceutical companies, and 15 per cent in electronic or computing companies. Others were from consulting, banking and finance, retail, construction, and various service industries. Sixty-four per cent of the respondents worked in firms that employed at least 1000 employees. Approximately 62 per cent of the respondents had at least 10 years of working experience, with the average working experience being 13 years. Seventy-nine per cent of the respondents were male.

Measures Task interdependency. In order to ensure that respondents had a clear and common understanding of the task interdependency concept, the following instruction preceded the questions measuring amount and clarity of interdependency in the survey: © Blackwell Publishing Ltd 2007

The Job Demands–Control Model

291

Now think about the degree and types of interdependency you experience in your job. Your work is independent to the extent that it does not require reliance on others to complete a task. Your work is interdependent to the extent that you depend on other people to perform a task or other people depend on you to accomplish their work. Of course, these relationships may involve people in your workgroup or firm, but they may also extend outside the company. We would like you to consider all cases when answering questions in this section. Amount of interdependency was measured by first asking the respondent to ‘allocate the percentage of your work that is independent versus interdependent with others’, and next to indicate ‘how many people (either inside or outside your workgroup or company) are you interdependent with on a daily basis’. Because amount of interdependency is a function of both the number of interdependent parties and the degree of interdependence with each party, we measured perceived amount of interdependency as a multiplicative score of percentage of work that is interdependent and the number of interdependent parties. On average, the respondents reported that 72 per cent (s.d. = 24) of their work was interdependent with others. On average, they reported that they were interdependent with 17 people (s.d. = 21) on a daily basis. Responses to the two items were first standardized, then multiplied to form the perceived amount of interdependency scale. Clarity of interdependency was measured by two items: (1) ‘To what extent are you clear about whom you are interdependent with when performing your work?’; and (2) ‘To what extent are you clear about where to obtain the resources (such as physical resources, supplies, data, information, etc) needed to perform your work?’ These two questions were scaled on a five-point Likert scale, ranging from 1 = not at all to 5 = to a very great extent. Cronbach’s alpha for this measure was 0.72. Role stress. Role conflict and role ambiguity scales were drawn from House et al.’s (1983) study. Their scales were derived from the pioneering work on role ambiguity and conflict by Rizzo et al. (1970). Role conflict was measured by eight items, and role ambiguity was measured by five items. All the items were scaled on a five-point scale, ranging from 1 = strongly disagree to 5 = strongly agree. Cronbach’s alphas for the role conflict and role ambiguity measures were 0.78 and 0.86 respectively. Component items can be obtained from the authors upon request. Job control. We used Ganster’s (1989) validated measure of job control. Ganster’s original scale had 22 items, each asking the subject how much control they possessed over the various facets of their work. We trimmed the scale to 13 items, removing those items that were not applicable to the managers in our sample; these included questions about control over the scheduling and duration of rest breaks, control over when vacations are taken, control over the physical conditions of one’s work station, or control over the ability to decorate or personalize the work area. All the items were scaled on a five-point Likert scale, ranging from 1 = very little to 5 = very much. The observed Cronbach’s alpha for this scale in our sample was 0.85. Component items can be obtained from the authors upon request. © Blackwell Publishing Ltd 2007

292

S-S. Wong et al.

Control variables. Given that the managers in our sample were from different industries, we created an industry dummy variable (1 = service-based industry, 0 = productionbased industry) to control for the different working environments of the managers. Specifically, we might expect managers in service-based industries to experience greater role conflict and ambiguity due to the less routine nature of work in such industries. We also included the degree of job experience as a control variable (measured by the number of months a manager has been in his or her current job) as managers with greater job experience may experience lower role stress (Wageman, 1995). Gender was also added as a control variable as there could be gender effects in perceptions of role stress (Paul, 1974). Finally, the size of the manager’s workgroup was included as a control variable. We asked participants to indicate ‘the number of people they work together with on a regular basis and are identified by others as a formal work group in the company’.[1] Workgroup size is likely to increase perceived role stress as the likelihood of conflicting demands from co-workers increases with size. Confirmatory Factor Analysis We performed a confirmatory factor analysis (CFA) on the task interdependency variables, role stress variables and job control variable to verify our hypothesized factor structure and to ensure convergent and discriminant validity. We tested the measurement model using the maximum likelihood LISREL 8.54 program. The chi-square statistic, c2 of the CFA was 622.41 (p < 0.001, df = 368) and the goodness-of-fit index (GFI) was 0.72. Although c2 was significant and GFI was not high, the other indices revealed a reasonably good fit. Specifically, the RMSEA was 0.07, indicating a satisfactory fit (Brown and Cudeck, 1989). The non sample-dependent indices such as the comparative fit index (CFI), non-normed fit index (NNFI) and incremental fit index (IFI) were 0.89, 0.88 and 0.89 respectively, reflecting an acceptable fit (Bentler, 1990). The relatively high values of these supporting indices suggest a satisfactory measurement model fit (Anderson and Gerbing, 1988). As the amount of interdependency was a single-item measure, we estimated its reliability to be 0.30 and constrained its pattern coefficient to 0.84. However, for all the other indicators, their estimated pattern coefficients as posited on the underlying factor were significant (greater than twice the standard error of the indicator), thus satisfying the test for convergent validity (Anderson and Gerbing, 1988). To test for discriminant validity, we determined whether the confidence interval around the correlation estimate between any two factors included 1.0 (Anderson and Gerbing, 1988). Results indicated that none of the confidence intervals between pairs of factors included 1.0, thus achieving the test for discriminant validity. RESULTS Table I reports the means, standard deviations, reliabilities, and Pearson correlations among the variables. All the variables have acceptable reliabilities, with the Cronbach alpha values ranging from 0.72 to 0.86. As seen in Table I, amount of and clarity of interdependency were not significantly related, thus suggesting that they are indeed © Blackwell Publishing Ltd 2007

3.37 3.26 2.28

4.05

0.57 0.73 0.98

0.70

0.49 – 38.46 21.89 38.82

s.d.

a N = 113. *** p < 0.001, ** p < 0.01, * p < 0.05, † p < 0.10.

7. 8. 9.

6.

0.37 – 41.69 19.42 16.60

1. 2. 3. 4. 5.

Industry Gender Job experience Workgroup size Amount of interdependency Clarity of interdependency Job control Role conflict Role ambiguity

Mean

Variable

0.08 0.00 -0.02

-0.16† 0.03 0.09

0.01 0.00 0.10

0.11

-0.05

-0.03

n.a. -0.09 -0.06

3

n.a. -0.03 -0.06 -0.07

2 n.a. 0.09 0.21* 0.07 0.15

1

Table I. Correlations, means, and standard deviationsa

n.a. 0.19*

-0.11 0.25** 0.00

-0.02

4

0.10

n.a.

-0.02 0.22* 0.05

5

0.24** -0.18† -0.37***

(0.72)

6

(0.85) -0.43*** -0.27**

7

(0.78) 0.37***

8

(0.86)

9

The Job Demands–Control Model 293

© Blackwell Publishing Ltd 2007

294

S-S. Wong et al.

distinct dimensions of task interdependency. The pattern of correlations between the task interdependency dimensions and the role stress dimensions was consistent with our hypotheses, thus providing some initial support for our arguments. Hierarchical regression analyses were conducted to test our hypotheses. In order to avoid multicollinearity, our predictor variables (i.e. amount of interdependency, clarity of interdependency and job control) were standardized and then multiplied to form the interaction terms (Aiken and West, 1991). Table II reports the results of the hierarchical regression analyses. With the exception of work group size, the control variables were not significantly related to role conflict (see Model 1a). When the task interdependency and job control variables were added to the regression model (see Model 1b), the amount of variance explained in role conflict significantly increased by 21 per cent (p < 0.001). Consistent with Hypothesis 1, amount of interdependency was positively and significantly related to role conflict (b = 0.19, p < 0.05). Hence, Hypothesis 1 was supported. Similarly, the set of control variables was not significantly associated with role ambiguity (see Model 2a). When the task interdependency and job control variables were added to the regression model (see Model 2b), the amount of variance explained in role ambiguity significantly increased by 18 per cent (p < 0.001). As we expected, clarity of interdependency was negatively and significantly related to role ambiguity (b = -0.34, p < 0.001), thus supporting Hypothesis 2. Hypothesis 3 and Hypothesis 4 propose interaction effects between task interdependency and job control on role conflict and role ambiguity respectively. As shown in Model 1c, the interaction term of amount of interdependency and job control was not significant in the regression model for role conflict (b = -0.02, p > 0.10). Hence, Hypothesis 3 was not supported. However, as shown in Model 2c, the interaction term of clarity of interdependency and job control was positively and significantly related to role ambiguity (b = 0.24, p < 0.01). Interestingly, even though we did not hypothesize a joint effect of amount of interdependency and job control on role ambiguity, their interaction term was positively and marginally significantly related to role ambiguity (b = 0.17, p < 0.10). Following Aiken and West’s (1991) probing procedures for interaction terms, we regressed role ambiguity on clarity of interdependency at high (one standard deviation above the mean) and low (one standard deviation below the mean) values of job control. Results showed that at a low level of job control, clarity of interdependency was significantly and negatively related to role ambiguity (b = -0.49, p < 0.001). However, at a high level of job control, clarity of interdependency was not significantly related to role ambiguity (b = -0.16, p > 0.10). This pattern of findings suggests that greater job control reduces managerial perceived role ambiguity when interdependencies are unclear. Hypothesis 4 was therefore supported. The interaction effect of job control and clarity of interdependency on role ambiguity is plotted in Figure 1. Hypothesis 5 and Hypothesis 6 propose a three-way interaction effect of amount of interdependency, clarity of interdependency and job control on role conflict and role ambiguity respectively. Hypothesis 5 was not supported as the three-way interaction term reported in Model 1d predicting role conflict was not significant (b = -0.11, p > 0.10). The three-way interaction term in the regression model for role ambiguity (see © Blackwell Publishing Ltd 2007

-0.01 0.03 0.04 0.22*

1.35 4, 108

0.05

0.24 0.14 0.17 0.00 0.00

0.06

-0.11

0.26 0.21*** 5.25*** 7, 105

-0.39*** 0.06

0.06

0.22 0.13 0.15 0.00 0.00

0.19*

-0.04 -0.02 0.09 0.14

*** p < 0.001, ** p < 0.01, * p < 0.05, † p < 0.10.

R2 D R2 F df

Amount ¥ clarity Amount ¥ clarity ¥ job control

Amount ¥ job control Clarity ¥ job control

Amount of interdependency Clarity of interdependency Job control

(Constant) Industry Gender Job experience Workgroup size

s.e.

s.e.

0.06

0.06

0.22 0.13 0.16 0.00 0.00

-0.02 0.05

0.26 0.00 4.07*** 9, 103

0.09 0.06

-0.40*** 0.07

-0.11

0.19*

-0.04 -0.02 0.09 0.13

b s.e.

0.06

0.12

0.22 0.14 0.16 0.00 0.00

-0.12 -0.11

-0.06 0.04

0.27 0.01 3.38*** 11, 102

0.09 0.13

0.11 0.06

-0.40*** 0.07

-0.13

0.31†

-0.04 -0.01 0.08 0.13

b

0.11 0.08 -0.04 -0.04

b

0.59 4, 108

0.02

0.33 0.20 0.23 0.00 0.00

s.e.

b

b

s.e.

Model 2a

Model 1d

Model 1b

Model 1a

Model 1c

Role ambiguity

Role conflict

Table II. Hierarchical regression results

0.09

0.31 0.18 0.22 0.00 0.00

s.e.

-0.20*

0.20 0.18*** 3.78*** 7, 105

0.09

-0.34*** 0.09

0.08

0.08 0.04 0.01 -0.08

b

Model 2b

0.09

0.30 0.18 0.21 0.00 0.00

s.e.

0.17† 0.24**

-0.27**

0.26 0.06* 4.08*** 9, 103

0.13 0.09

0.09

-0.34*** 0.08

0.06

0.13 0.00 0.02 -0.12

b

Model 2c

0.16

0.29 0.18 0.21 0.00 0.00

s.e.

-0.05 -0.29*

0.03 0.22*

-0.29**

0.31 0.05* 4.11*** 11, 101

0.12 0.17

0.15 0.09

0.09

-0.37*** 0.09

0.17

0.15+ 0.01 0.01 -0.12

b

Model 2d

The Job Demands–Control Model 295

© Blackwell Publishing Ltd 2007

296

S-S. Wong et al. 3.5

3

Role Ambiguity

2.5

2

1.5

1

0.5

0 – 1 s.d.

+1 s.d. Clarity of Interdependency High Job Control

Low Job Control

Figure 1. Interactive effect of clarity of interdependency and job control on role ambiguity

Model 2d) was significant (b = -0.29, p < 0.05). Figure 2 displays the three-way interaction effect on role ambiguity. The pattern is consistent with our predictions. At high amount and low clarity of interdependency, the difference in predicted role ambiguity between the high and low job control group was not significant (b = -0.09, p > 0.10). However, at low amount and low clarity of interdependency, the predicted role ambiguity value in the high job control group was significantly lower than that of the low job control group (b = -0.75, p < 0.001). Hence Hypothesis 6 was supported. DISCUSSION Work stress does not only derive from the tasks we do, but also from the web of interactions we have to manage – the many others on whom we are dependent in order to complete our work (Lichtenberg, 1998). Managers today confront a large and complex web of task interdependencies, propelled by decentralization, globalization, and distributed operations of firms. For instance, the managers in our sample were keenly aware of the importance of interdependence in their work. They reported, on average, that 72 per cent of their work was interdependent with other people and only 28 per cent was independent. They estimated, on average, that they were interdependent on a daily basis with at least 17 other people. While there has been extensive research on the design and coordination needs of task interdependency (Crowston, 1997; Galbraith, 1977; Wageman, 1995), there is little understanding of how managerial perceptions of interdependency affect role stress, or how mechanisms such as job control enable managers to cope with this stress. Our study has extended research on the job demands–control model beyond consideration of workload and work pace (e.g. de Rijk et al., 1998; Fox et al., 1993; Van Yperen © Blackwell Publishing Ltd 2007

The Job Demands–Control Model

297

(a) 3.5

3

Role Ambiguity

2.5

2

1.5

1

0.5

0

–1s.d.

+1s.d. Clarity of Interdependency

Low Job Control

High Job Control

(b) 4

3.5

Role Ambiguity

3

2.5

2

1.5

1

0.5

0 –1s.d.

+1s.d. Clarity of Interdependency Low Job Control

High Job Control

Figure 2. Interactive effect of amount of interdependency, clarity of interdependency and job control on role ambiguity. (a) High amount of interdependency. (b) Low amount of interdependency

and Hagedoorn, 2003) to include the impact of the more social demands of the modern manager’s job. We examined the relationship between perceived amount and clarity of task interdependency on managerial role stress. Our findings indicate that task interdependency is, indeed, associated with increased role stress. Specifically, managers who perceived greater amount of interdependency experienced greater role conflict, and © Blackwell Publishing Ltd 2007

298

S-S. Wong et al.

managers who perceived less clarity of interdependency experienced greater role ambiguity. Task interdependency creates demands to coordinate with others (Kiggundu, 1983; Thomas, 1957), so managers experience greater role stress when they are subjected to conflicting expectations from interdependent parties and/or face uncertainty about with whom and what to coordinate. Job demands relating to the amount and clarity of interdependence in work thus have important potential implications for managerial mental health. Further, our findings suggest that the effects of task interdependency on role stress are not universal but vary as a function of two important dimensions – amount of interdependency and clarity of interdependency. Managers can experience different types of role stress depending on how variation in these dimensions prevails in their work. In line with Karasek’s (1979) job demands–control model, we found that greater perceived job control reduces managers’ perceived role ambiguity when there is low clarity of interdependency. Although the importance of job control as a stress-reducing mechanism has been questioned in some studies (e.g. Parker and Sprigg, 1999; Schaubboeck and Merritt, 2003), our findings supported its moderating effect on the relationship between clarity of interdependency and role ambiguity. Managers who perceive greater job control feel that they have greater autonomy to (re)structure their work relationships (e.g. increase communication) and/or (re)design their work methods (e.g. streamline their work) to clarify the web of interdependencies in their job, which can in turn alleviate perceived role ambiguity. Low clarity of interdependency does not raise role ambiguity when job control is high; in this way high job control can mitigate the otherwise negative consequences of low clarity of interdependency. It is reasonable to expect that managers, who in general are accustomed to greater decision latitude in their work, would respond positively to greater job control under stressful job demands as they can exploit the greater autonomy to reduce the pressures of their work environment. However, when managers confront both unclear and high amounts of interdependency, job control becomes less effective as a stress-reducing mechanism. We found that at a high amount and low clarity of interdependency, managers in both high and low job control groups perceived similar levels of role ambiguity. Job control was effective in role ambiguity reduction only for managers who faced a small set of unclear interdependencies (i.e. low amount and low clarity of interdependency). Our findings suggest that job control is ineffective in reducing role stress when the job demands create such a complex and uncertain work environment (as in the case when managers have to manage a huge web of ambiguous interdependencies) that managers do not know how to make use of the increased autonomy to address these demands. In contrast, in relatively less demanding task interdependency conditions, managers with higher job control experienced lower role ambiguity, implying that managers are able to constructively employ the autonomy to manage job demands when the demands are more tractable. Whereas many studies have challenged the buffering role of job control based on individual differences (e.g. Parker and Sprigg, 1999; Schaubboeck and Merritt, 2003; Xie, 1996), our findings suggest a more fundamental refinement to Karasek’s job demands–control model: job control has a buffering value contingent upon the amount and clarity of task interdependency faced by incumbents. We did not find support for an interaction between amount of interdependency and job control or for a three-way interaction between the task interdependency dimensions © Blackwell Publishing Ltd 2007

The Job Demands–Control Model

299

and job control on role conflict. Job control was not perceived as an effective stressreducing mechanism when the source of stress was derived from a high amount of interdependency. If the amount of task interdependency arises from the organizational structure (e.g. formal coordination channels between departments) and how one’s job scope is formally designed (e.g. marketing manager has to span various departmental boundaries), managers may not be free to terminate conflicting interdependencies. Even though there could be greater latitude to decide when to interact with others or which work methods to use (i.e. greater job control), it may be that managers have little organizational power or influence to reduce the number of interdependencies in their job insofar as these interdependencies are structurally derived. In particular, when asked to report the extent to which the interdependencies in their job arise from the way tasks are defined and scheduled in their firm, 84 per cent of the managers in our sample rated at least 3 on a scale ranging from 1 to 5. Hence, to the extent that organizational workflow is formally designated, managers may perceive greater job control as ineffective in reducing conflicting interdependencies. An area worthy of future research would be to explore how different types of control may be appropriate to manage different types of job demands. Limitations and Future Directions There are three important limitations of our study. First, all the data in this study were self-reported, thus increasing the risk of common method variance. Despite this risk, perceptual measures, which are by definition self-reports, are in fact core to our theory. As outlined earlier, our theory centres on perceived interdependency. Perceived interdependency has been posited to be distinct from objective interdependency (Tjosvold, 1986) due to reasons such as bounded rationality and uncertainty (McCann and Ferry, 1979). Following this logic, managers facing the same objective job demands could experience different mental stress due to varying perceptions of amount of and clarity of interdependency. As such, we focused on perceived as opposed to objective interdependency. Our outcome variables – role conflict and role ambiguity – are also less amenable to objective measurement (in contrast to physiological stress variables) and have been measured in the role stress literature by self-reports (House et al., 1983; Rizzo et al., 1970). Hence, the use of self-reports is consistent with our theoretical focus. Notwithstanding, concerns about common method variance remain and are a limitation of this study. Second, the reliance on cross-sectional data to test our hypotheses rules out conclusions about causality. However, the theorized impact of job demands on stress outcomes is consistent with extensive studies of work stress (e.g. Schaubboeck and Merritt, 2003; Van Yperen and Hagedoorn, 2003). In contrast, there has been no theoretical grounding for the reverse causal relationship, that is, the impact of role stress on perceived job demands. Regardless, longitudinal research would be valuable to establish the causal relationship between task interdependency and role stress. Third, we targeted a group of managers for study, arguing that most research on the job demands–control model has targeted nurses and production workers. Future research might explicitly compare occupational groupings and test the relative power of © Blackwell Publishing Ltd 2007

300

S-S. Wong et al.

the model, extended to include task interdependency. In this way we can more clearly determine the applicability of the model for different kinds of occupations. Finally, we examined the characteristics of the overall set of task interdependency faced by our respondents. We did not distinguish between the types of interdependency (e.g. internal versus external interdependency, lateral versus vertical interdependency) a manager has to manage. It is conceivable that managerial role stress may vary as a function of the types of interdependency a manager has to manage. For instance, a manager who has a higher number of external interdependencies to coordinate may experience greater role ambiguity because external interdependencies tend to be more fluid and less structured. The direction of the interdependent relationship, whether the focal person is the one who demands information or the one who supplies information, is another critical dimension in need of analysis. Given that the types of interdependencies managers have to manage are becoming more varied, it would be important for future research to investigate how different types of interdependency impact managerial role stress. Managerial Implications Our study shows how task interdependency in modern managerial work can contribute to role stress. As such, it is important for organizations to find ways to structure the interdependencies in managers’ jobs in ways that reduce perceptions of high volume of interdependency and/or unclear interdependency. One approach to this process that may be useful is to visually map the various interdependencies in a manager’s job. This can be constructive in illuminating the web of interdependencies a manager has to manage, especially when the work environment is fluid and might serve as a base for potentially identifying and removing interdependencies with conflicting demands. Managing interorganizational interdependencies will inevitably bring conflicting demands into play. Even in the event that interdependencies with conflicting demands cannot be terminated, at least relevant organizational support can be rendered to help managers deal with incompatible demands from different parties. Although job control is an effective stress-reducing mechanism when clarity of interdependency is low, we found that it is most effective as a stress-reducing mechanism in managerial jobs when the set of unclear interdependencies is small. Managers working in such a work environment should be given the latitude to structure their work in a way that can elucidate the various interdependencies they have to manage. However, we suspect that it would be challenging to increase managerial perception of greater job control when the nature of task interdependency (e.g. low clarity) heightens feelings of low control (since perceived interdependency, by definition, means a perception that one has to rely on others to perform one’s tasks). Perhaps one way to increase perceived job control is not only to allow latitude to adjust work methods but also to provide the power to terminate and/or create interdependencies. Autonomy to seek and reduce one’s interdependencies can illuminate one’s web of interdependencies more clearly than formal workflow charts when there are complex and frequent changes in operations. Overall, our study highlights that organizations may face a profound dilemma in the future related to the increasingly interdependent nature of managerial work. In order to © Blackwell Publishing Ltd 2007

The Job Demands–Control Model

301

remain competitive, organizations will need to promote behaviours that enable managers to take on a portfolio of multi-dimensional interdependencies (e.g. interdepartmental and/or interorganizational interdependencies) – but this work allocation can be detrimental to mental health. An important future challenge of organizational and job design will be to somehow reconcile organizational needs for multiple interdependency management with managerial needs for reduced role stress.

NOTES *Nancy Staudenmayer passed away in November 2000 after a long struggle with illness. Nancy was instrumental in inspiring our interest in the study of interdependency, a topic in which she had deep interests and insights. Both Gerry DeSanctis and I revived this research study in 2004, but sadly, Gerry passed away in August 2005 after losing the battle with cancer. Both Gerry and Nancy were my mentors while I was a doctoral student at Duke University and, through them, I have witnessed (and learned) the spirit of genuine scholarship – unwavering passion for learning and pursuit of knowledge. As colleagues, they were exceedingly generous with their insights and committed to contributing to every research collaboration they were involved in. But, most importantly, they will also be remembered as caring friends – always warm, thoughtful and supportive. [1] Based on this definition, we expect the manager’s work group to include the manager’s direct reports as well as peer colleagues who work regularly with the manager in the same group.

REFERENCES Aiken, L. S. and West, S. G. (1991). Multiple Regression: Testing and Interpreting Interactions. Thousand Oaks, CA: Sage. Ancona, D. G. and Caldwell, D. F. (1992). ‘Bridging the boundary: external activity and performance in organizational teams’. Administrative Science Quarterly, 37, 634–65. Anderson, J. C. and Gerbing, D. W. (1988). ‘Structural equation modeling in practice: a review and recommended two step approach’. Psychological Bulletin, 103, 411–23. Badaracco, J. L. Jr (1991). The Knowledge Link: How Firms Compete through Strategic Alliances. Boston, MA: Harvard Business School. Bartlett, C. A. and Ghoshal, S. (1987). ‘Managing across borders: new strategic requirements’. Sloan Management Review, 28, 7–17. Bentler, P. M. (1990). ‘Comparative fit indexes in structural models’. Psychological Bulletin, 107, 238–46. Brown, M. W. and Cudeck, R. (1989). ‘Single sample cross-validation indices for covariance structures’. Multivariate Behavioral Research, 24, 445–55. Brown, S. L. and Eisenhardt, K. M. (1997). ‘The art of continuous change: linking complexity theory and time-paced evolution in relentlessly shifting organizations’. Administrative Science Quarterly, 42, 1–34. Cooper, C. L. (Ed.) (2000). Theories of Organizational Stress. Oxford: Oxford University Press. Crowston, K. (1997). ‘A coordination theory approach to organizational process design’. Organization Science, 8, 157–75. Davenport, T. H. and Prusak, L. (1998). Working Knowledge: How Organizations Manage What They Know. Boston, MA: Harvard Business School Press. Davidow, W. H. and Malone, M. S. (1992). The Virtual Corporation. New York: Harper Business. De Croon, E. M., Sluiter, J. K., Broersen, J. P. J., Blonk, R. W. B. and Frings-Dresen, M. H. W. (2004). ‘Stressful work, psychological job strain, and turnover: a 2-year prospective cohort study of truck drivers’. Journal of Applied Psychology, 89, 442–54. de Rijk, A. E., Le Blanc, P. M., Schaufeli, W. B. and De Jonge, J. (1998). ‘Active coping and need for control as moderators of the job demand-control model: effects on burnout’. Journal of Occupational and Organizational Psychology, 71, 1–18. DeSanctis, G., Staudenmayer, N. and Wong, S. (1999). ‘Interdependence in virtual organizations’. In Cooper, C. and Rousseau, D. (Eds), Trends in Organizational Behavior: The Virtual Organization. New York: Wiley, 6, 81–104. © Blackwell Publishing Ltd 2007

302

S-S. Wong et al.

Doh, J. P. (2005). ‘Offshore outsourcing: implications for international business and strategic management theory and practice’. Journal of Management Studies, 42, 695–704. Eccles, R. G. and Crane, D. B. (1988). Doing Deals: Investment Banks at Work. Boston, MA: Harvard Business School Press. Fox, M. L., Dwyer, D. J. and Ganster, D. C. (1993). ‘Effects of stressful job demands and control on physiological and attitudinal outcomes in a hospital setting’. Academy of Management Journal, 36, 289–318. Fulk, J. and DeSanctis, G. (1995). ‘Electronic communication and changing organizational forms’. Organization Science, 6, 337–49. Galbraith, J. R. (1974). ‘Organization design: an information processing view’. Interfaces, 4, 28–36. Galbraith, J. R. (1977). Organization Design. Reading, MA: Addison-Wesley. Ganster, D. C. (1989). Measurement of Worker Control. Final report no. 88-70187. Cincinnati, OH: National Institute of Occupational Safety and Health. Ganster, D. C. and Fusilier, M. R. (1989). ‘Control in the workplace’. In Cooper, C. L. and Robertson, I. (Eds), International Review of Industrial and Organizational Psychology. Chichester: John Wiley, 235–80. Ghoshal, S. and Bartlett, C. A. (1990). ‘The multinational corporation as an interorganizational network’. Academy of Management Review, 15, 603–25. Hill, L. A. (2004). ‘New manager development for the 21st century’. Academy of Management Executive, 18, 121–6. Holland, J. L. (1973). Making Vocational Choices: A Theory of Vocational Personalities and Work Environments. Englewood Cliffs, NJ: Prentice Hall. Holland, J. L. (1985). Making Vocational Choices: A Theory of Careers. Englewood Cliffs, NJ: Prentice Hall. House, R. J., Schuler, R. S. and Levanoni, E. (1983). ‘Role conflict and ambiguity scales: reality or artifacts?’. Journal of Applied Psychology, 68, 334–7. Jehn, K. A., Northcraft, G. B. and Neale, M. A. (1999). ‘Why differences make a difference: a field study of diversity, conflict, and performance in workgroups’. Administrative Science Quarterly, 44, 741–64. Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoek, J. D. and Rosenthal, R. A. (1964). Organizational Stress. New York: Wiley. Karasek, R. (1979). ‘Job demands, job decision latitude, and mental strain: implications for job redesign’. Administrative Science Quarterly, 24, 285–306. Kellogg, K. C., Orlikowski, W. J. and Yates, J. (2006). ‘Life in the trading zone: structuring coordination across boundaries in postbureaucratic organizations’. Organization Science, 17, 22–44. Kiggundu, M. N. (1983). ‘Task interdependence and job design: test of a theory’. Organizational Behavior and Human Decision Processes, 31, 145–72. Kogut, B. and Zander, U. (1992). ‘Knowledge of the firm, combinative capabilities, and the replication of technology’. Organization Science, 3, 38397. Landsbergis, P. A. (1988). ‘Occupational stress among health care workers: a test of the job demands–control model’. Journal of Organizational Behavior, 9, 217–40. Levy, D. L. (2005). ‘Offshoring in the new global political economy’. Journal of Management Studies, 42, 685–93. Lichtenberg, R. (1998). Work Would Be Great if it Weren’t for the People. New York: Hyperion. Liebeskind, J. P., Oliver, A. L., Zucker, L. and Brewer, M. (1996). ‘Social networks, learning, and flexibility: sourcing scientific knowledge in new biotechnology firms’. Organization Science, 7, 428–43. Lui, S. S. and Ngo, H. Y. (2005). ‘An action pattern model of inter-firm cooperation’. Journal of Management Studies, 42, 1123–53. Malone, T. W. and Rockart, J. F. (1991). ‘Computers, networks, and the corporation’. Scientific American, 265, 92–9. McCann, J. E. and Ferry, D. L. (1979). ‘An approach for assessing and managing inter-unit interdependence’. Academy of Management Review, 4, 113–19. McGrath, J. E. (1976). ‘Stress and behavior in organizations’. In Dunnette, M. (Ed.), Handbook of Industrial and Organizational Psychology. Chicago, IL: Rand McNally, 1351–95. Mintzberg, H. (1971). ‘Managerial work: analysis from observation’. Management Science, 18, B97–110. Nohria, N. (1992). ‘Is a network perspective a useful way of studying organizations?’. In Nohria, N. and Eccles, R. G. (Eds), Networks and Organizations: Structure, Form and Action. Boston, MA: Harvard Business School Press, 1–22. Parker, S. K. and Sprigg, C. A. (1999). ‘Minimizing strain and maximizing learning: the role of job demands, job control, and proactive personality’. Journal of Applied Psychology, 84, 925–39. Paul, R. J. (1974). ‘Role clarity as a correlate of satisfaction, job related strain and propensity to leave – male vs. female’. Journal of Management Studies, 11, 233–47. © Blackwell Publishing Ltd 2007

The Job Demands–Control Model

303

Perlow, L. A. (1999). ‘The time famine: toward a sociology of work time’. Administrative Science Quarterly, 44, 57–81. Rizzo, J. R., House, R. J. and Lirtzman, S. I. (1970). ‘Role conflict and ambiguity in complex organizations’. Administrative Science Quarterly, 15, 150–63. Rubery, J., Earnshaw, J., Marchington, M., Cooke, F. L. and Vincent, S. (2002). ‘Changing organizational forms and the employment relationship’. Journal of Management Studies, 39, 645–72. Savery, L. K. (1988). ‘Comparison of managerial and non-managerial employees’ desired and perceived motivators and job satisfaction levels’. Leadership and Organization Development Journal, 9, 17–22. Schaubboeck, J. and Merritt, D. E. (2003). ‘Divergent effects of job control on coping with work stressors: the key role of self-efficacy’. Academy of Management Journal, 40, 738–54. Schneider, B. (1987). ‘The people make the place’. Personnel Psychology, 40, 437–54. Shea, G. P. and Guzzo, R. A. (1987). ‘Group effectiveness: what really matters’. Sloan Management Review, 28, 25–31. Stamper, C. L. and Johlke, M. C. (2003). ‘The impact of perceived organizational support on the relationship between boundary spanner role stress and work outcomes’. Journal of Management, 29, 569–88. Staudenmayer, N. (1997). ‘Managing Multiple Interdependencies in Large Scale Software Development Projects’. Unpublished PhD Dissertation, The Sloan School of Management, Massachusetts Institute of Technology, Cambridge, MA. Thomas, E. J. (1957). ‘Effects of facilitative role interdependence on group functioning’. Human Relations, 10, 347–66. Thompson, J. D. (1967). Organizations in Action. New York: McGraw-Hill. Tjosvold, D. (1986). ‘The dynamics of interdependence in organizations’. Human Relations, 39, 517–40. Tjosvold, D. (1988). ‘Cooperative and competitive interdependence: collaboration between departments to serve customers’. Group & Organization Studies, 13, 274–89. Torrington, D. and Weightman, J. (1987). ‘Middle management work’. Journal of General Management, 13, 74–89. Van de Ven, A., Delbecq, A. L. and Koenig, R. (1976). ‘Determinants of coordination modes within organizations’. American Sociological Review, 41, 322–38. Van Yperen, N. W. and Hagedoorn, M. (2003). ‘Do high job demands increase intrinsic motivation or fatigue or both? The role of job control and job social support’. Academy of Management Journal, 46, 339–48. Wageman, R. (1995). ‘Interdependence and group effectiveness’. Administrative Science Quarterly, 40, 145–80. Wall, T. D., Jackson, P. R., Mullarkey, S. and Parker, S. K. (1996). ‘The demands–control model of job strain: a more specific test’. Journal of Occupational and Organizational Psychology, 69, 153–66. Wong, S. (2004). ‘Distal and local group learning: performance trade-offs and tension’. Organization Science, 15, 645–56. Xie, J. L. (1996). ‘Karasek’s model in the People’s Republic of China: effects of job demands, control, and individual differences’. Academy of Management Journal, 39, 1594–618.

© Blackwell Publishing Ltd 2007