Project management and burnout: Implications of the Demand–Control–Support model on project-based work

Available online at www.sciencedirect.com

ScienceDirect International Journal of Project Management 32 (2014) 578 – 589 www.elsevier.com/locate/ijproman

Project management and burnout: Implications of the Demand–Control–Support model on project-based work Jeffrey K. Pinto a,⁎, Shariffah Dawood b , Mary Beth Pinto a a b

Black School of Business, Penn State — Erie, Erie, PA 16563, United States Department of Psychology, Penn State — Erie, Erie, PA 16563, United States

Received 2 July 2013; received in revised form 3 September 2013; accepted 5 September 2013 Available online 22 October 2013

Abstract Project-based work has long been characterized as frenetic, fast-paced, and dynamic. The often competing constraints imposed by schedules, stakeholders, and budgetary restrictions make project activities conflict-laden and highly conducive to work-related stress. Stress is not an end unto itself but instead, is often a precursor for burnout. Burnout is a psychological syndrome of emotional exhaustion, cynicism, and reduced personal accomplishment. This paper reports on the results of a study of burnout among project management personnel. Using the Demand–Control– Support model as our conceptual framework, we analyzed a sample of respondents from four project-intensive organizations. Our findings demonstrated that women tend to experience emotional exhaustion to a greater extent than their male counterparts. Further, control and social support do serve as moderators for the burnout dimensions of emotional exhaustion and cynicism, suggesting limited support for the Demand– Control–Support model. Implications of this study for project management and workplace burnout are discussed. © 2013 Elsevier Ltd. APM and IPMA. All rights reserved. Keywords: Burnout; Work-related health; Stress

1. Introduction Project managers and their teams face complex, highly demanding, and often stressful work environments. As a model for organizational activities, project management continues to grow in popularity as project-based work becomes a favored means for promoting organizational output, initiating critical change, and penetrating into industries that had hitherto operated using more formalized and bureaucratic processes, e.g. health care, insurance, banking and financial services (Aitken and Crawford, 2007). Typically fast-paced and dynamic, projects require constant alignment with organizational strategies while also balancing competing concerns for schedules, budgets, stakeholder satisfaction, and quality. “The project manager experiences a significant level of stress because of an endless list of demands, deadlines, and problems throughout the project's life cycle.” (Verma, 1996, p. 176). As such, it is little wonder that project settings are highly ⁎ Corresponding author at: Burke Building, Penn State Erie, Erie PA 16563. E-mail address: [email protected] (J.K. Pinto). 0263-7863/$36.00 © 2013 Elsevier Ltd. APM and IPMA. All rights reserved. http://dx.doi.org/10.1016/j.ijproman.2013.09.003

conducive to work-related stress (Verma, 1996; Richmond and Skitmore, 2006; Haynes and Love, 2004). The ever-present nature of conflict and stress in the professional roles of project managers and team members is heavily discussed in the project management literature. The project manager's job is characterized by “role overload, frenetic activity, and superficiality,” resulting from the wide scope of their responsibilities coupled with limited resources and authority (Slevin and Pinto, 1987; p. 33). For example, over two decades of research has led to important insights into this long-assumed but under-tested phenomena; namely, the propensity for project managers and their teams to experience significant work-related stress (Asquin et al., 2010; Chiocchio et al., 2010; Djebarni, 1996; Gallstedt, 2003; Love and Edwards, 2005; Richmond and Skitmore, 2006; Sommerville and Langford, 1994). Stress, as a psychological state, is not perceived as an end unto itself, but rather it is understood to be the cause of other psychological pathologies, none more significant than workplace burnout. Burnout is defined as a psychological syndrome of emotional exhaustion, depersonalization and reduced

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

personal accomplishment (Maslach, 1993). Emotional exhaustion refers to the depletion of one's emotional resources and is linked to anxiety, physical fatigue, and tension. The depersonalization (cynicism) component represents the interpersonal context dimension of burnout. It suggests a negative and detached response to clients or other organizational stakeholders. Finally, reduced personal accomplishment or efficacy represents the self-evaluation dimension of burnout, implying a low level of perceived competence and inability to successfully complete work assignments (Maslach, 1993). Maslach (1982) viewed burnout as a natural consequence of forces acting on the individual over time. It results from a continuous imbalance between resources and demands that promote emotional exhaustion, result in depersonalization, and finally, reduced personal accomplishment (Maslach and Leiter, 1997). Another feature of burnout is its perception as a sequential process; that is, work stressors can lead to emotional exhaustion, which in turn can cause a sense of depersonalization or cynicism, ultimately resulting in a sense of loss of workplace efficacy (Leiter, 1993). Thus, burnout offers a number of significant negative consequences for individuals and the performance of workplace duties. Interestingly, while recent work has moved to address the linkage between project-based work and stress, little research has pursued this subsequent cause/effect relationship; namely, the propensity for project managers and team members to develop burnout from their responsibilities. One notable exception is the work of Emelander (2011), who investigated the impact of burnout and intrinsic needs fulfillment among project managers. His field study reported moderate levels of burnout among project managers and significant correlations with needs fulfillment and self-determination. Project-based organizations need to recognize the likelihood of their project management staff encountering burnout. The purpose of this paper is to examine the relationship between project management roles and duties and the potential for burnout. Research suggests that certain starting conditions encourage a negative experience that can result in burnout. Two antecedents, heavy workloads and time pressure, are strongly and consistently related to burnout, particularly the exhaustion dimension (Maslach et al., 2001). Studies of qualitative job demands have focused primarily on role conflict and role ambiguity, both of which consistently show a moderate to high correlation with burnout (Maslach et al., 2001). Research also suggests that lack of control in decision-making is generally related to the inefficacy or reduced personal accomplishment aspect of burnout (Karasek, 2008). Mismatches in control most often indicate that individuals have insufficient control over the resources needed to do their work or have insufficient authority to pursue the work in what they believe is the most effective manner. In addition to studying the presence of job demands and lack of control, burnout researchers have investigated the absence of job resources. The resource that has been studied most extensively has been social support, and there is now a consistent and strong body of evidence that a lack of social support is linked to burnout (Maslach et al., 2001). As project organizations increase in number and size, project-based work continues to grow in popularity through

579

its acceptance and use in diverse industries, and the demand for project management professionals expands to fill this need, it is appropriate to consider the implications of enhanced work demands on project managers and their teams. The following research questions are proposed: What is the likelihood of the incidence of burnout in project management? How can we better understand the factors that can lead to burnout? What are the psychological or work-related issues that can moderate the feelings (and negative consequences) of projects managers' and their teams' burnout? 2. Understanding the nature of burnout: the Job Demand–Control–Support model For nearly 40 years, the concept of burnout has received a great deal of attention, especially in the psychological literature, where it has been applied to a variety of professionals including social workers, educators, medical and mental health workers, police officers, child care workers, lawyers, and customer service representatives (Maslach et al., 1996). Burnout has been shown to have a variety of dysfunctional consequences, including turnover, absenteeism, and reduced performance on the job, all resulting in significant costs to the individual and organization (Bernin and Theorell, 2001; Jackson and Maslach, 1982; Leiter and Maslach, 1988; Shirom, 1989). For example, studies have shown that high burnout in the nursing industry has negative consequences for not only nurses' job performance but also their home life and personal relationships (Proost et al., 2004). Burnout in the health care industry has also been shown to lead to lower levels of organizational commitment, decreased job satisfaction, higher health care costs, and decreases in creativity, problem solving and innovation (Halbesleben and Buckley, 2004; Shirom, 1989). Finally, burnout has also been examined for gender differences; that is, the implied differential predilection toward burnout and its resulting impact on men versus women. Research suggestions that both men and women experience burnout but differently; that is, burnout effects vary by gender. Interestingly, there were also reported larger gender differences in burnout in USA organizations relative to those in the EU (Purvanova and Muros, 2010). Much of the research on burnout has focused on its antecedents. One of the more useful models of burnout is the Job Demand–Support–Control (JDCS) model, originally operationalized as the Job Demand–Control model (Karasek, 1979) but more recently modified to include a social support dimension (Johnson and Hall, 1988). The JDCS model identifies three major elements in the work setting that impact an individual's level of well being: job demands, job control, and social support (Sargent and Terry, 2000). Job demands refer to the workload that individual faces and are often associated with time pressure, role ambiguity and role conflict (see also Karasek, 1985; De Bruin and Taylor, 2006). Job control or decision latitude refers to the extent to which an individual feels they can exert influence over tasks they face and is most often operationalized through the constructs of skill discretion and decision authority. The skill discretion

580

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

component mostly addresses issues on task variety, while the decision authority component refers to the opportunity to make independent decisions and to have a say in what happens in the workplace. Finally, social support refers to both supervisor support as well as the support of colleagues and co-workers (Van der Doef and Maes, 1999). Applying the first two dimensions (job demand and control), researchers and theorists have classified jobs into four categories (see Fig. 1). A job that contains high demands but also offers concomitantly high control is referred to as an “active job,” observed to lead to high motivation and a sense of personal growth. On the other hand, jobs that offer low demands and low control are called “passive jobs.” Jobs with high control and low demands are “low strain” jobs while those with high demands but low control are referred to as “high strain” jobs. High strain jobs are characterized as more conducive to stress due to the negative implications of carrying high demands but a perceived inability to control one's work environment. Later, when the social support dimension was added, it was possible to further identify the worst job situations as being “iso-strain jobs;” that is, those offering no social support (suggesting personal isolation). On the other hand, workplace stress theory tends to support the idea that high levels of social support can offset and ease the pressure of high strain jobs. Thus, the most threatening circumstance faced by managers are those of high demands, perceived low control, and a sense of isolation through lack of social support. Research on the JDCS model offers some important implications for project team members and the potential for burnout. First, the “strain” hypothesis, suggests that employees working in a “high-strain” situation (high demands–low control) experience the highest level of burnout. Second, support for the “iso-strain” hypothesis suggests that the lowest level of psychological well-being is experienced by employees working in an “iso-strain” situation (high demands–low control–low support). Third, research further analyzed the cause–effect relationship between

High

Active Job

Job Demands

High Strain Job

Passive Job

Low Strain Job

Low Low

Control

Fig. 1. Job Demand–Control–Support model.

High

demand and burnout, suggesting that burnout is often seen as the difference (interaction) between work-related demands and a project manager's perceived control; that is, control is argued to moderate (a “buffer” hypothesis) the relationship between demands and burnout. Fourth, when social support is factored into the model, it is argued to serve as a further moderator of this relationship; that is, it is assumed that one's method for coping with high strain jobs is through a supportive social network that helps moderate (a second “buffer” hypothesis) the negative effects of the project manager's job. Generally speaking, the strain and iso-strain hypotheses yielded more consistent support than the buffer hypotheses of the JDC model and the JDCS model (Van der Doef and Maes, 1999). Research on the JDCS model and burnout suggests that there is only minimal evidence to support the contention that control and/or support serve as effective moderators of the negative effects of high strain jobs (Landsbergis, 1988; Melamed et al., 1991; Van der Doef and Maes, 1999). Several studies have applied the JDCS model to the individual sub-dimensions of burnout (emotional exhaustion, depersonalization, and reduced personal accomplishment) but continue to find only limited support for the moderator effects of control and social support on the impact of high demand on burnout; however, they do suggest that job demands and control may differentially affect the degree of burnout (De Jonge et al., 1996; Schreurs and Taris, 1998). Critics emphasize that the JDCS model conceptualizes demand and control in a very general way, and argue that the test of specific types of demands and control would be more valuable. Specifically, Van der Doef and Maes (1999) conclude that the buffer hypothesis has been mainly supported for specific occupational subgroups (in terms of person characteristics or position in the organization) and when a specific (e.g. time pressure) instead of a broader (e.g. quality concern) type of demand interacts with a specific type of control (e.g. authority over pace). There is a considerable body of research that has examined the construct of burnout and its relationship to the JDCS model and suggested some underlying conclusions. First, support for the iso-strain hypothesis is fairly consistent, indicating a relationship between high iso-strain work and burnout (Van der Doef and Maes, 1999). Second, it does not appear that the negative effects of high job demands on burnout can be moderated by higher levels of control (de Jonge and Kompier, 1997). In reaction to these findings, the current study will specify the job demands and control that particularly characterize the occupation of project managers and team members. That is, it may be the case that demand and control constructs have been too broadly specified in past research. Third, job demands appear to be a stronger predictor of emotional exhaustion and a weaker predictor of depersonalization and reduced personal accomplishment (Taris et al., 1999). Fourth, level of control appears to be associated with each burnout dimension, although the relationship between control and burnout is not as clear as the linkage between job demands and burnout. Finally, the additional construct of social support does not appear to offer any significant amelioration between that high strain jobs and burnout. Although several studies have

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

suggested that social support can positively influence the three dimensions of burnout, the construct has not demonstrated a moderator effect among the three principal dimensions of job demands, control, and social support (Rafferty et al., 2001). Thus, we hypothesize: H1. Greater levels of perceived iso-strain (high demand, low control, low social support) will lead to greater levels of burnout in project team members, as evidenced by emotional exhaustion, cynicism, and reduced personal efficacy. Previous studies have suggested “buffering” effects in the JDCS model; that is, there are expected to be significant interaction effects among job demands, job control, and social support from both colleagues and superiors (see, for example, Proost et al., 2004). Following this logic, we suggest that job control can buffer for the negative effects of high demand. Further, social support in both its forms can buffer for the negative effects of high demand, low control jobs. Finally, job autonomy can buffer for the negative effects of a high iso-strain job, identified as high demand, low control, and low social support (see Fig. 1). Therefore, we expect: H2. Job control and social support (colleague and supervisor) will have significant moderating effects on the relationship between job demands and burnout. Looking at the present knowledge of determinants of burnout and some conflicting results regarding this issue, it is very important to further investigate how project team members experience their work. Because of the unique demands and high potential for stress so commonly found among project managers and their teams when the nature of the work environment is examined (i.e., fast-paced work, high demands, oftentimes minimal direct authority over resources, the potential for social isolation, and so forth), it is appropriate to study the impact of the project management work environment on burnout rates.

581

large manufacturer of medical device technology. The liaison sent out an email to employees requesting their participation in the study, with a follow-up email sent approximately ten days later. Data were collected using a web-based, self-report questionnaire employing Survey Monkey. Participants completed an implied consent form when they agreed to complete the questionnaire. All data reported are aggregated and complete confidentiality was guaranteed to participants. No inducements were offered to motivate participation. A total of 353 people responded to the survey, representing a response rate of approximately 42% across all four organizations (estimated based on size of electronic mailing lists). 3.2. Participants The sample was composed of project managers, executives, team members, and resource/support individuals currently involved in projects. The final sample was 75.5% male, and 24.5% female. The average age of respondents was 44.2 years. Fifty-two percent of the sample was college graduates, with an additional 34.5% having either a graduate degree or some graduate education. Of those who gave their title, a total of 98 respondents listed their job as project manager, 21 were project administrators, 36 were project engineers, seven were senior executives, and 90 were project team members (see Table 1). The “other” category had a sample size of 72 and included a variety of professional affiliations, including designers, discipline leads, and QC personnel. Respondents were asked to indicate the types of projects they typically were involved with. Table 2 lists their primary project responsibilities by project type. In some cases, respondents indicated their participation on multiple types of projects (e.g., construction, consulting (services), and process improvement). Finally, Table 3 offers some additional descriptive statistics of the respondents in the sample and the characteristics of projects to which they are typically assigned. 3.3. Measures

3. Method 3.1. Procedure All procedures were carried out at the individual level of study. We contacted key organizational members at four major North American corporations who were part of their firms' “project management groups” and agreed to serve as liaison within their organization for the purposes of encouraging employees to respond to the questionnaire. For confidentiality reasons, we were prevented from more directly or completely identifying the specific groups within the organizations. All had primary responsibility for planning and executing major projects within their respective firms, either through new product development initiatives, information systems solutions development and implementation, or industrial construction projects. These organizations included: one large IT-based organization, two construction/EPCM (Engineering, Procurement, and Construction Management) organizations, and one

The current study used the following scales: A short demographic questionnaire; Maslach's Burnout Inventory General Survey (MBI-GS); Job Content Questionnaire (JCQ); De Jonge's (1995) job autonomy scale, and De Jonge et al. (1993) psychological demand scale. The demographic questionnaire was designed to ascertain variables including job title, Table 1 Job title. Job title

Frequency

Percent

Project manager Project administrator Project engineer Senior executive Project team member Other No title given Total

98 11 36 7 89 74 26 341

28.7 3.2 10.6 2.1 26.1 21.7 7.6 100.0

582

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

Table 2 Sample breakdown by project type. Type of projects

Sample size

Percentage of sample

Information technology Construction/EPCM Manufacturing Research & development Financial Services Other or none listed

85 141 28 49 12 45 52

21% 33% 7% 12% 3% 11% 13%

age, gender, average number of team members supervised, average number of projects typically worked on at the same time, size of the project budget, duration for a typical project, duration in current position, duration with current company, and types of projects typically undertaken. The MBI-GS is designed as a diagnostic tool to label individuals as “burned out” (Maslach et al., 1996). The MBI-GS can be used in any occupational context and is comprised of 16 statements with three subscales (i.e. exhaustion, cynicism and professional efficacy) that are parallel to those of the original MBI, except that items do not explicitly refer to working with people. Exhaustion is measured with 5 items, including “I feel burned out from my work” and “I feel tired when I get up in the morning and have to face another day on the job.” Cynicism is also measured with five items. Example items are: “I have become less enthusiastic about my work” and “I have become more cynical about whether my work contributes anything.” Finally, professional efficacy is measured with six items, including “I feel I am making an effective contribution to what this organization does” and “In my opinion I am good at my job.” We employed a modified form of the MBI-GS in which participants rated items on a five-point Likert scale with end points: 1 = “Strongly Disagree” to 5 = “Strongly Agree.” Leiter and Schaufeli (1996) have shown that internal consistency of each of these scales is satisfactory. They found Cronbach's alpha coefficients ranging from 0.84 to 0.90 for exhaustion, 0.74 to 0.84 for cynicism, and from 0.70 to 0.78 for professional efficacy. In this study, the alpha for exhaustion was .92, alpha for cynicism was .82, and the alpha for professional efficacy was .78, suggesting acceptably high reliabilities for these scales (Nunnally, 1978). The Job Content Questionnaire (JCQ) is a self-administered instrument designed to measure social and psychological characteristics of jobs (Karasek, 1985). The best-known scales – (a)

Table 3 Descriptive statistics on projects and respondents. Sample Mean Standard reporting deviation If manager on team, average number of subordinates Average number of projects worked on simultaneously Length of time in current position (in months) Length of time with current company (in months)

135 239 278 280

25.4 50.2 2.86 2.72 35.84 2.29 82.23 86.26

psychological demands and (b) social support – are used to measure the demand–support model of job strain development. The JCQ questionnaire was applied in this study, consisting of two scales. They are “psychological job demands” (5 items) and “social support,” including coworker support (4 items) and supervisor support (4 items). Items in the scales were scored using a 5-point Likert scale in which 1 indicated that respondents strongly disagree and 5 indicating that they strongly agree. In this study, Cronbach's alpha for psychological job demands was .77, alpha for coworker support was .81, and for supervisor support, the Cronbach's alpha was .93. In order to include specific types of job control that are more relevant to project managers, an additional 9 items were adopted from De Jonge's (1995) job autonomy scale. It measures the worker's opportunity to determine multiple facets of task elements such as method of working, pace of work, work goals, order in which work is carried out, amount of work, working hours, kind of work, and work evaluation. Scale reliability, as assessed through Cronbach's alpha, was .81. Similarly, job demand was further assessed using the eight-item scale developed by De Jonge et al. (1993), which consists of a relatively wide range of both quantitative and qualitative demanding aspects, like working under pressure of time, working hard, strenuous work, and task complexity. The scales for psychological job demands (Karasek, 1985) and the De Jonge et al. (1993) scale were combined into a single measure of job demands. The Cronbach's alpha measure of scale reliability was .85. Confirmatory factor analysis showed unidimensionality for this scale. In summary, the constructs and scales are summarized in Table 4.

3.4. Analysis To examine the individual and interaction effects of job demands, job control and social support for the JDCS (Job Demand–Control–Support) model, we used moderated multiple regression, similar to the procedure that is commonly followed in testing this model (Van der Doef and Maes, 1999). We executed three moderated hierarchical regression analyses,

Table 4 Scales used in study. Constructs and scales

Source

Burnout – Exhaustion — 5 items – Cynicism — 5 items – Efficacy — 6 items Job demands – Psychological demands — 5 items – Job demands — 8 items Job control – Job autonomy — 9 items Support – Supervisor support — 4 items – Coworker support — 4 items

Maslach MBI-GS (Maslach et al., 1996)

Karasek (1985) and De Jonge et al. (1993)

De Jonge (1995) Karasek (1985)

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

583

Table 6 Moderated hierarchical regression analysis on emotional exhaustion dimension.

for each of the three dimensions of burnout: exhaustion, cynicism, and personal efficacy. There is some evidence to suggest that burnout effects have been found for gender (Maslach and Jackson, 1981; Vlerick, 1996), age (Anderson and Iwanicki, 1984; Maslach and Jackson, 1981) and work experience (Anderson and Iwanicki, 1984). We also wished to determine if, for project management professionals, the size of the project (as measured by budget and duration) would demonstrate effects on burnout. In other words, would larger projects produce higher levels of stress resulting in job burnout? As a result, we first entered these demographic variables into the regression analysis to control for their potential effects. In the second step, the main effect terms of job demands, job control, and social support of the supervisor and colleagues were entered. It has been argued that a proper test of the three-way interaction in the Demand– Control–Support model using moderated multiple regression would also include all possible two-way interactions, so we entered these interaction terms in the third step of the regression (Van der Doef and Maes, 1999). Finally, in the fourth step, we entered the three-way interaction effect.

Predictor

Step 1

Step 2

Step 3

Step 4

Gender − .36 + − .47 ⁎ − .48 ⁎⁎ − .52 ⁎⁎ Age .01 − .006 − .004 − .007 Project budget .001 .001 .001 .001 Project duration .001 − .002 − .002 − .003 Job demands .78 ⁎⁎ .80 ⁎⁎ .96 ⁎⁎ Job control − .33 ⁎ − .38 ⁎ − .46 ⁎⁎ Super. support − .04 − .02 − .007 Coworker supp. − .28 ⁎ − .29 ⁎ − .21 + Job control × job demands .35 + .28 Job control × super. support − .06 .05 Job control × coworker supp. .33 − .15 Job demands × super. support .04 − .14 Job demands × coworker supp. − .39 + − .35 Job control × job demands × .29 super. support Job control × job demands × − 1.05 ⁎⁎ coworker supp. .05 .34 .37 .42 R2 cumulative Adjusted R2 .02 .30 .31 .36 R2 change .05 .29 .03 .05 F change 1.75 15.28 1.36 5.86 DF1 4 8 13 15 DF2 143 139 134 133

4. Results

All entries are unstandardized regression (β) coefficients. Listwise deletion of missing variables. + p b .10. ⁎ p b .05. ⁎⁎ p b .01.

Table 5 shows the means, standard deviations, and Pearson correlations of the variables in this study. As can be seen in the table, all the independent variables in the model were significantly related to the three dimensions of burnout (emotional exhaustion, cynicism, and personal efficacy). The results of the hierarchical regression analyses are presented in Tables 5 through 7 to account for each of the three dimensions of burnout. No variance inflation factor (VIF) exceeded the value of 10, indicating that multicolinearity was not a problem (Hair et al., 2009). For our control variables, age, project budget, and project duration had no significant effects on any of the dimensions of burnout. Interestingly, for the dimension of personal exhaustion, gender showed strong significant effects (β = − .52, p b .01, for three-way interaction), suggesting that women tend to experience the burnout effect of exhaustion more often than men in projects.

In line with the first hypothesis, there is evidence for the three-way interaction effects of the model; that is, the expectation that higher levels of perceived iso-strain (high demand, low control, low social support) will lead to greater levels of burnout (see, for example, Figs. 2 and 3). This effect was further evidenced by the tests for the second hypothesis and the expected moderating effects of control and social support on the relationship between demands and burnout. Interactive effects were tested by means of moderated hierarchical regression with main effects (demand, job control, and social support) entered first and each of the interaction terms entered second (job control × job demands, job control ×

Table 5 Correlation matrix. Variable

Mean

S.D.

1

2

3

4

5

6

7

8

9

1. Job demands 2. Job control 3. Social support of coworkers 4. Social support of supervisor 5. Emotional exhaustion 6. Cynicism 7. Personal efficacy 8. Age 9. Project budget ($) 10. Project duration (months)

3.47 3.64 3.58 3.72 2.79 2.27 1.95 44.1 642,035,631 21.5

0.61 0.61 0.70 0.97 1.02 0.86 0.58 12.24 1,118,341,343 16.1

.25 ⁎⁎ − .14 ⁎⁎ − .13 ⁎⁎ .47 ⁎⁎ .12 ⁎ − .13 ⁎ .21 ⁎⁎ − .09 − .04

.31 ⁎⁎ .31 ⁎⁎ − .15 ⁎⁎ − .51 ⁎⁎ − .55 ⁎⁎ .23 ⁎⁎ − .14 ⁎ − .08

.61 ⁎⁎ − .33 ⁎⁎ − .44 ⁎⁎ − .35 ⁎⁎ − .10 − .05 − .11

− .28 ⁎⁎ − .41 ⁎⁎ − .29 ⁎⁎ − .17 ⁎⁎ − .001 − .08

.55 ⁎⁎ .27 ⁎⁎ .06 .10 .05

.54 ⁎⁎ − .08 .03 .01

− .07 .14 ⁎ .11

.05 .04

.33 ⁎⁎

Correlations were two-tailed. ⁎ p b .05. ⁎⁎ p b .01.

10

584

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

Table 7 Moderated hierarchical regression analysis on cynicism dimension. Predictor

Step 1

Step 2

Step 3

Step 4

Gender − .16 − .09 − .11 − .10 Age .001 − .003 − .003 − .002 Project budget .001 .001 .001 − .001 Project duration .001 − .003 − .002 − .003 Job demands .21 ⁎ .23 ⁎ .35 ⁎⁎ Job control − .66 ⁎⁎ − .64 ⁎⁎ − .71 ⁎⁎ Super. support − .12 − .07 − .09 Coworker supp. − .23 ⁎ − .26 ⁎ − .23 ⁎ Job control × job demands − .05 − .40 ⁎ Job control × super. support − .03 .10 Job control × coworker supp. .34 .37 + Job demands × super. support − .12 .11 Job demands × coworker supp. .02 − .30 + Job control × job demands × − .74 ⁎⁎ super. support Job control × job demands × .57 ⁎⁎ coworker supp. .009 .34 .37 .43 R2 cumulative Adjusted R2 − .02 .30 .30 .36 R2 change .009 .33 .03 .06 F change .30 16.82 1.2 6.57 DF1 4 8 13 15 DF2 139 135 130 128 All entries are unstandardized regression (β) coefficients. Listwise deletion of missing variables. + p b .10. ⁎ p b .05. ⁎⁎ p b .01.

supervisor support, job control × coworker support, job demands × supervisor support, job demands × coworker support) and finally, three-way interactions were entered (job control × job demands × supervisor support, job control × job demands × coworker support). Separate analyses were performed using each of the burnout dimensions. For the emotional exhaustion dimension, the main effects for job demands, job control and coworker support were statistically significant, with

job demands showing the highest incremental contribution to the predictive power of the model (R-square). The total variance explained by these variables was 34%, and was statistically significant at F = 15.28, p b .05. For the cynicism dimension, the main effects for job demands, job control and coworker support were statistically significant, with job control contributing the most. The total variance explained by these variables was 34%, and was statistically significant at F = 16.82, p b .05. As for the personal efficacy dimension, only the main effect for job control was significant and explained 36% of the variance at F = 18.37, p b .05 (Table 8). It should be noted that supervisor support was the only main effect variable which was consistently non-significant across all three dimensions of burnout. Interestingly, when tested for two-way interaction effects, results revealed that supervisor support acted as a moderator between job control and personal efficacy. This finding suggests that the back-up or reassurance of one's supervisor is crucial when making decisions, and this in turn increases one's sense of personal accomplishment (being put down or second-guessed by one's supervisor is not likely to help in developing a sense of personal accomplishment). Two-way interaction effects for the remaining IVs (job control × job demands, job control × coworker support, job demands × supervisor support, job demands × coworker support) did not show significant effects in predicting the outcomes of emotional exhaustion and cynicism dimensions. The change in variance explained by the three-way interaction effects in predicting emotional exhaustion (ΔR2 =5%) was statistically significant at F = 5.86, p b .05. However, the three-way interaction effect of (job control × job demands × coworker support) was the only statistically significant combination, (β = − 1.05, p b .01) in predicting emotional exhaustion (see Fig. 2). Fig. 2 shows the interaction effect slopes for the various combinations and demonstrates that high demands coupled with low perceived control and low coworker support leads to significantly higher levels of emotional exhaustion. On

Fig. 2. Three-way interaction effects between job demands, control, and coworker support on emotional exhaustion.

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

585

Fig. 3. Three-way interaction effects between job demands, control, and coworker support on cynicism.

the other hand, high control and high coworker support can effectively mitigate the effects of high job demands on the emotional exhaustion dimension. The change in variance explained by the three-way interaction effects in predicting cynicism (ΔR2 = 6%) was statistically significant at F = 6.57, p b .05. The three-way interaction effect of (job control × job demands × supervisor support) was statistically significant (β = − .74, p b .01). Project managers

who are working in demanding situations with low control and high supervisor support were rated high in the cynicism dimension, suggesting the lack of buffering effect from supervisor support (Fig. 4). In addition, the three-way interaction effect of (job control × job demands × coworker support) was also statistically significant (β = .57, p b .01). Social support from coworkers seems to be especially important for project managers working in highly demanding projects with a perception of low control.

Table 8 Moderated hierarchical regression analysis on personal efficacy dimension. Predictor

Step 1

Step 2

Step 3

Step 4

Gender − .008 .07 .04 .04 Age − .002 .001 − .001 − .002 Project budget .001 .001 .001 .001 Project duration .003 .001 .002 .001 Job demands − .02 − .05 .02 Job control − .57 ⁎⁎ − .52 ⁎⁎ − .55 ⁎⁎ Super. support − .009 − .06 − .06 Coworker supp. − .13 + − .09 − .07 Job control × job demands .07 − .009 Job control × super. support .20 ⁎ .24 ⁎ Job control × coworker supp. − .12 − .11 Job demands × super. support .04 .05 Job demands × coworker supp. .11 .05 Job control × job demands − .08 × super. support Job control × job demands − .11 × coworker supp. R2 cumulative .02 .36 .40 .42 Adjusted R2 − .007 .32 .35 .36 R2 change .02 .34 .04 .02 F change .74 18.37 1.89 1.99 DF1 4 8 13 15 DF2 142 138 133 131 All entries are standardized regression (β) coefficients. Listwise deletion of missing variables. + p b .10. ⁎ p b .05. ⁎⁎ p b .01.

5. Discussion The goal of this study was to examine the impact of burnout on project team members. We noted that to date, while much research has begun to explore the issues of work-related stress and pressure to perform on organizational actors engaged in project-based work (e.g., Asquin et al., 2010; Gallstedt, 2003; Love and Edwards, 2005; Richmond and Skitmore, 2006), stress, of itself, is not to be viewed as an end-stage effect. Rather, the psychological literature suggests that stress is a state that often leads to a more insidious and debilitating state: burnout, as evidenced by the dimensions of emotional exhaustion, cynicism, and perceived loss of efficacy. There were a number of intriguing findings from this study. First, high levels of job demands in project settings were clearly evidenced as independent variables (predictors) in the cases of two dimensions of burnout (emotional exhaustion and cynicism). For the third dimension – personal efficacy – job demands was not a significant predictor. These findings support previous research that has shown project-based work to be characterized by high pressure in the form of job demands (Verma, 1996; Richmond and Skitmore, 2006). Second, when testing only for main effects (step 2 of our moderated hierarchical regression analyses), higher levels of job control and co-worker support are significant “negative” predictors

586

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

Fig. 4. Three-way interaction effects between job demands, control, and supervisor support on cynicism.

of project burnout, suggesting that they can play a role in mitigating the onset of burnout for project-based work. Another interesting finding was the propensity for female project team members to experience burnout, in the form of emotional exhaustion, at higher levels than their male counterparts. This finding is in line with previous work that studied burnout in nurses (Proost et al., 2004; Vlerick, 1996) as well as general organizational settings (Maslach and Jackson, 1981; Pretty et al., 1992) and reported gender effects for dimensions of burnout. A meta-analysis conducted by Purvanova and Muros (2010) examined 183 previous studies of burnout and found that women experience emotional exhaustion to a greater degree than men, though men are more prone to the burnout effect of cynicism. Thus, our study offers limited support for the Purvanova and Muros (2010) results by demonstrating the significance of this same burnout effect (emotional exhaustion) as it applies to the women in our sample. Other studies had employed a number of demographics as control variables, just as we did in this study. We found no evidence of any significant effects for the control variables of age or project size (as measured by budget and duration). Gender, however, appears to continue to be a significant predictor of at least the emotional exhaustion dimension of burnout. The differential impact of burnout on men versus women in project-based work is an interesting and potentially, highly significant area for future exploration. As noted, it would be an error to suggest that project work remains the domain of “alpha males;” that is, that women simply find project work more taxing and are prone to higher levels of stress and resultant burnout. In fact, long and well-established streams of research are pointing to the fact that men and women are equally subject to burnout symptoms, though they typically manifest themselves in different ways, with women prone to emotional exhaustion and men inclined toward cynicism. It is worth considering the question of whether or not there are potential burnout symptoms that many project managers are simply not

attuned to identifying. For example, though emotional exhaustion may manifest itself through more noticeable employee reactions like sluggish responses or apparent fatigue and withdrawal, the cynicism response (particularly as it is most prone in men) may be masked more easily or misattributed by supervisors to simple flippancy or exaggerated sarcasm. In short, burnout in project-based work is no respecter of gender; without adequate training, however, project managers and senior executives may fail to recognize its impact on both their male and female employees. Our study showed some evidence in support of the iso-strain conceptualization of workplace burnout through the interaction of higher levels of control and social support (either co-worker or supervisor) on the effects of job demands; that is, there was partial support for the iso-strain model of workplace burnout. On two of the burnout dimensions – emotional exhaustion and cynicism – higher levels of control and social support did moderate the direct impact of job demands on burnout. This finding was particularly intriguing because to date, the iso-strain model has not consistently yielded empirical evidence in its support; that is, while many studies have identified the existence of two-way interaction effects (demands × control) on burnout, the more complex iso-strain model, which incorporates a social support dimension, has been less well-supported. Although we did not find clear evidence of iso-strain moderating effects across all forms of burnout in our study (i.e., the burnout dimension of personal efficacy showed no support for iso-strain), we did find support on two of the dimensions for at least one form of social support (co-worker support). Thus, in combatting the effects of emotional exhaustion, project team members working in high strain jobs benefitted most from receiving social support from co-workers. Put another way, true iso-strain work can strongly influence the burnout effect of exhaustion in project team members. Equally interesting, although the three-way moderating effects of demand/control/support were evidenced for the

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

burnout dimension of cynicism, the effects were in different directions. That is, high supervisor support, coupled with low control and high demands (most likely originating from the same supervisor) appears to increase employee cynicism (see Fig. 3). On the other hand, coworker support appears to offer a genuine buffer against cynicism, both in cases where perceived control is high or low. This “negative effect” of supervisor support is intriguing, as it suggests evidence of additional interpersonal effects between leaders and members of project organizations. In their study of transformational leadership in project-based environments, Keegan and Den Hartog (2004) found that the nature of the leader–member dyad – defining the impacting role of leadership on team member goal commitment and motivation – is a complex phenomenon. This raises the question of whether it is possible that leader–member exchange in project-based settings can actually weaken the effectiveness of supervisor support. Certainly, our findings suggest that this may be a fruitful avenue for future exploration. Our findings also suggest partial support for the control dimension and its impact on burnout. As we noted earlier in this paper, research to date (see, for example, de Jonge and Kompier, 1997) has not shown consistent evidence that the negative effects of high job demands on burnout can be moderated by higher levels of control. One reason that we posited was that it may be the case that “control” was too broadly specified to be useful, particularly within the specific context of project-based work. Using de Jonge's (1995) Job Autonomy scale as a measure of control, this two-way interaction effect was found to be significant for the cynicism dimension (β = − .40, p b .05). Our findings also confirm Taris et al. (1999) work that determined that job demands are a stronger predictor of emotional exhaustion than cynicism or reduced personal efficacy. We found no empirical support for job demand's impact on efficacy, though it was strongly significant for both exhaustion and cynicism. The results are interesting as well in light of recent work on the idea of “project involvement,” as developed by Chiocchio et al. (2010). In their study, they observed that high levels of project involvement within the firm tend to offset distress symptoms. That is, mental fatigue and stress symptoms were somewhat mitigated by the level of project involvement that an organization manifests. Thus, more highly “projectized” firms tended to provide more support for their human resources involved in project-based work and had lower levels of work-related distress. Further, people involved in projects from a non-projectized organization show less mental health than those from a projectized organization. Our sample for this study consisted of four organizations that would be classified as “project involved,” according to the Chiocchio et al. (2010) conceptualization and might, as a result, show less pathology in relation to the JDCS model. Nevertheless, our research still demonstrates strong evidence of the potential for distress and workplace burnout. 6. Limitations and directions for future research This paper addressed a number of factors shown in the literature to be associated with the construct of burnout. It also

587

offered partial validation of the JDSC model and the construct of iso-strain jobs, particularly set within the project management context. Project-based work has been shown for many years as being particularly prone for creating an environment in which stress can occur (Verma, 1996; Slevin and Pinto, 1987). A logical extension of this work – our present study – has been to address the common effect of stress; namely, workplace burnout. While this study offers a number of intriguing findings, there are some limitations that should be considered. First, the data are cross-sectional, preventing us from drawing conclusions about the temporal nature of these constructs and how they impact on burnout through following the same sample over time. We had noted earlier that burnout is often considered a result of pressures affecting workplace actors; that is, burnout is proposed to have a number of antecedent conditions that must first be observed. For example, past research suggests that the JDSC model be viewed as a set of antecedent conditions (and moderators) of burnout. So, too, is the issue of workplace stress. Research has likewise pointed to the mediational role that behavioral factors such as role ambiguity (Rizzo et al., 1970), job autonomy (De Jonge, 1995), or “work–home interference” (Geurts et al., 1999) can play on the link between job demands and feelings of well-being. Future research, in which data are collected at various points across a project's life cycle, could shed additional light on the assumed temporal nature of these antecedents of project burnout. A second limitation concerns the sample itself. The majority of respondents came from two large construction/EPCM organizations that were currently engaged in very large capital projects. In fact, the average project budget reported in the study was in excess of $650 million. While the study did include a sample of respondents from two organizations in which project budgets and durations were smaller, it is clear that overall, the sample is reporting on psychological effects while engaged in large (nearly “mega”) projects. These results are intriguing of themselves but they beg the follow-on question: what about burnout effects from personnel who work on smaller projects of shorter duration? Are the effects found in equal measure within these lesser project settings? Future research should develop samples of projects from a variety of settings, including different project classes (IT, construction, manufacturing, pharmaceutical, and so forth) as well as projects of different time and budget parameters. Further, it would be useful to break down data samples by job classification. Much work in the project management field has focused on the stress that project managers face in the work environment, facing the myriad demands and constraints that their job provides (c.f., El-Sabaa, 2001; Love and Edwards, 2005). Thus, contrasting burnout levels of project managers versus team members and other significant project stakeholders could lend credence to the notion that project managers face a higher-than-average degree of burnout potential from their work. A third limitation calls into question the potential for burnout to be a cultural phenomenon or at least one that differentially affects project team personnel from different

588

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589

countries and cultures. This study investigated the JDCS model and other moderators on project burnout within North American organizations. It would be interesting to expand the sample population with data from project teams working in different parts of the world to see if these same effects are prevalent across cultural boundaries or are, in fact, specific to project team personnel in North American organizations. To date, significant work has focused attention of the unique characteristics of project-based work and the demands that it places on project managers and their teams. The constraints of time pressures, coupled with limited resources, myriad (and often conflicting) stakeholder demands, make managing projects in modern organizations a challenging and demanding undertaking (Zika-Viktorsson et al., 2006). In such a context, it is not surprising that research strongly suggests the potential for stress and psychological pathology to develop. Stress, as the literature notes, is not an end unto itself, but instead signals the potential onset of a more insidious outcome: the development of workplace burnout. This study has confirmed the potential for burnout to exist due to the unique combination of high job demands, coupled with perceptions of control and social support. Although demonstrating only partial support for the JDCS conceptualization of burnout, our study does support the perception that for thousands of project workers, the threat of iso-strain jobs is real and has significant debilitating potential. Finding and supporting effective methods for enhancing job control and social support (particularly co-worker support) are critical steps in helping ameliorate these negative effects and have been shown to offset the negative impact of the high job demands so common in undertaking projects. References Aitken, A., Crawford, L., 2007. Coping with stress: dispositional coping strategies of project managers. Int. J. Proj. Manag. 25, 666–673. Anderson, M.B.G., Iwanicki, E.F., 1984. Teacher motivation and its relationship to burnout. Educ. Adm. Q. 20, 109–132. Asquin, A., Garel, G., Picq, T., 2010. When project-based management causes distress at work. Int. J. Proj. Manag. 28, 166–172. Bernin, P., Theorell, T., 2001. Demand–control–support among female and male managers in eight Swedish companies. Stress. Heal. 17, 231–243. Chiocchio, F., Beaulieu, G., Boudrias, J.-S., Rousseau, V., Aube, C., Morin, E., 2010. The project involvement index, psychological distress, and psychological well-being: comparing workers from projectized and non-projectized organizations. Int. J. Proj. Manag. 28, 201–211. De Bruin, G.P., Taylor, N., 2006. Sources of Work Stress Inventory: Technical Manual. Jopie van Rooyen & Partners, Johannesburg. De Jonge, J., 1995. Job Autonomy, Well-being, and Health: A study among Dutch health care workers. Unpublished doctoral thesis, Datawyse, Maastricht. De Jonge, J., Kompier, M.A.J., 1997. A critical examination of the Demand– Control–Support model from a work psychological perspective. Int. J. Stress. Manag. 4, 235–258. De Jonge, J., Landeweerd, J.A., Nijhuis, F.J.N., 1993. Constructie en validering van de vragenlijstten behoeve van het project ‘autonomie in het werk’ [“Construction and validation of the questionnaire for the ‘job autonomy project’”], Studies bedrijfsgezondheidszorg, nummer 9. Rijksuniversiteit Limburg, Maastricht. De Jonge, J., Janssen, P.P.M., van Breukelen, G.J.P., 1996. Testing the demand–control–support model among health-care professionals: a structural equation model. Work. Stress. 10, 209–224. Djebarni, R., 1996. The impact of stress on site management effectiveness. Constr. Manag. Econ. 14, 281–293.

El-Sabaa, S., 2001. The skills and career path of an effective project manager. Int. J. Proj. Manag. 19, 1–7. Emelander, S.J., 2011. A study of burnout and intrinsic needs fulfillment among project managers. Unpublished doctoral dissertation, Capella University. Gallstedt, M., 2003. Working conditions in projects: perceptions of stress and motivation among project team members and project managers. Int. J. Proj. Manag. 21, 449–455. Geurts, S.A.E., Rutte, C., Peeters, M., 1999. Antecedents and consequences of work-home interference among medical residents. Soc. Sci. Med. 48, 1135–1148. Hair, J.F., Black, W.C., Babin, B., Anderson, R.E., 2009. Multivariate Data Analysis, 7th ed. Pearson, Upper Saddle River, NJ. Halbesleben, J.R.B., Buckley, M.R., 2004. Burnout in organizational life. J. Manag. 30, 859–879. Haynes, N.S., Love, P.E.D., 2004. Psychological adjustment and coping among construction project managers. Constr. Manag. Econ. 22, 129–140. Jackson, S.E., Maslach, C., 1982. After-effects of job-related stress: families as victims. J. Occup. Behav. 3, 63–77. Johnson, J.V., Hall, E.M., 1988. Job strain, work place social support and cardiovascular disease: a cross-sectional study of a random sample of the Swedish working population. Am. J. Public Health 78, 1336–1342. Karasek, R.A., 1979. Job demands, job decision latitude, and mental strain: implications for job design. Adm. Sci. Q. 24, 285–308. Karasek, R.A., 1985. Job Content Instrument: Questionnaire and User's Guide. Department of Industrial and Systems Engineering, Los Angeles, CA. Karasek, R.A., 2008. Low social control and physiological deregulation — the stress-disequilibrium theory, toward the new demand–control model. Scand. J. Work Environ. Health Suppl. 6, 117–135. Keegan, A.E., Den Hartog, D.N., 2004. Transformational leadership in a project-based environment: a comparative study of the leadership styles of project managers and line managers. Int. J. Proj. Manag. 22, 609–617. Landsbergis, P.A., 1988. Occupational stress among health care workers: a test of the job demands–control model. J. Organ. Behav. 9, 217–239. Leiter, M.P., 1993. Burnout as a developmental process: consideration of models. In: Schaufeli, W.B., Maslach, C., Marek, T. (Eds.), Professional Burnout: Recent Developments in Theory and Research (Series in Applied Psychology: Social Issues and Questions). Taylor & Francis, Philadelphia, PA, pp. 237–250. Leiter, M.P., Maslach, C., 1988. The impact of interpersonal environment on burnout and organizational commitment. J. Organ. Behav. 9, 297–308. Leiter, M.P., Schaufeli, W.B., 1996. Consistency of the burnout construct across occupations. Anxiety Stress Coping 9, 229–243. Love, P.E.D., Edwards, D.J., 2005. Taking the pulse of UK construction project managers' health. Eng. Constr. Archit. Manag. 12, 88–101. Maslach, C., 1982. Burnout: The Cost of Caring. Prentice-Hall, New York. Maslach, C., 1993. Burnout: a multidimensional perspective. In: Schaufeli, W.B., Maslach, C., Marek, T. (Eds.), Professional Burnout: Recent Developments in Theory and Research. Taylor & Francis, Philadelphia, PA, pp. 237–250. Maslach, C., Jackson, S.E., 1981. The measurement of experienced burnout. J. Occup. Behav. 2, 99–113. Maslach, C., Leiter, M.P., 1997. The Truth About Burnout: How Organizations Cause Personal Stress and What to Do About It. Jossey-Bass, San Francisco, CA. Maslach, C., Jackson, S.E., Leiter, M.P., 1996. The Maslach Burnout Inventory, 3rd ed. Consulting Psychologists Press, Palo Alto, CA. Maslach, C., Schaufeli, W.B., Leiter, M.P., 2001. Job burnout. In: Fiske, S.T., Schacter, D.L., Zahn-Waxler, C. (Eds.), Annual Review of Psychology, 52, pp. 397–422. Melamed, S., Kushnir, T., Meir, E.I., 1991. Attenuating the impact of job demands: additive and interactive effects of perceived control and social support. J. Vocat. Behav. 39, 40–53. Nunnally, J.C., 1978. Psychometric Theory. McGraw-Hill, New York. Pretty, G.M.H., McCarthy, M.E., Catano, V.M., 1992. Psychological environments and burnout: gender considerations within the corporation. J. Organ. Behav. 13, 701–711. Proost, K., de Witte, H., de Witte, K., Evers, G., 2004. Burnout among nurses: extending the Job Demand–Control–Support model with work–home interference. Psychol. Belg. 44–4, 269–288.

J.K. Pinto et al. / International Journal of Project Management 32 (2014) 578–589 Purvanova, R.K., Muros, J.P., 2010. Gender differences in burnout: a metaanalysis. J. Vocat. Behav. 77 (2), 168–185. Rafferty, Y., Friend, R., Landsbergis, P.A., 2001. The association between job skill discretion, decision authority and burnout. Work. Stress. 15, 73–85. Richmond, A., Skitmore, M., 2006. Stress and coping: a study of project managers in a large ICT organization. Proj. Manag. J. 37 (5), 5–16. Rizzo, J.R., House, R.J., Lirtzman, S.I., 1970. Role conflict and ambiguity in complex organizations. Adm. Sci. Q. 15, 150–163. Sargent, L.D., Terry, D.J., 2000. The moderating role of social support in Karasek's job strain model. Work. Stress. 14, 245–261. Schreurs, P.J.G., Taris, T.W., 1998. Construct validity of the demand– control model: a double cross-validation approach. Work. Stress. 12, 64–84. Shirom, A., 1989. Burnout in work organizations. In: Cooper, C.L., Robertson, I. (Eds.), International Review of Industrial and Organizational Psychology. Wiley, New York, pp. 26–48.

589

Slevin, D.P., Pinto, J.K., 1987. Balancing strategy and tactics in project implementation. Sloan Manag. Rev. 29 (1), 33–41. Sommerville, J., Langford, V., 1994. Multivariate influences on the people side of projects: stress and conflict. Int. J. Proj. Manag. 12, 234–243. Taris, T.W., Schreurs, P.J.G., Schaufeli, W.B., 1999. Construct validity of the Maslach Burnout Inventory-General Survey: a two-sample examination of its factor structure and correlates. Work. Stress. 13, 223–237. Van der Doef, M., Maes, S., 1999. The Job Demand–Control (–Support) model and psychological well-being: a review of 20 years of empirical research. Work. Stress. 13, 87–114. Verma, V.K., 1996. Human Resource Skills for the Project Manager. Project Management Institute, Upper Darby, PA. Vlerick, P., 1996. Burnout and work organization in hospital wards: a crossvalidation study. Work. Stress. 13, 87–114. Zika-Viktorsson, A., Sundstrom, P., Engwall, M., 2006. Project overload: an exploratory study of work and management in multi-project settings. Int. J. Proj. Manag. 24, 385–394.