The effect of self-leadership on work role performance in teams

Abstract

This research examined the effect of self-leadership strategies on individuals’ work role performance in teams. Using an experimental policy-capturing design, self-leadership, task interdependence and situational uncertainty were manipulated in two studies. Moreover, the moderating effect of psychological collectivism orientation on the self-leadership performance relation was explored. Results from multilevel analyses revealed that in Study 1, self-leadership had a positive effect on individual task and team member work role performance. Study 2 replicated and extended these results by showing positive effects of self-leadership on individuals’ team member proficiency, adaptivity and proactivity in teams. Furthermore, collectivism orientation moderates the effect of self-leadership on team member proficiency. Implications of the findings are identified, limitations are discussed and areas for future research are proposed.

Keywords

adaptivity proactivity collectivism HLM self-leadership team member job performance

Today’s organizations, due to increasingly flexible and dynamic work structures, grant employees an increasing degree of autonomy and responsibility to accomplish given tasks in their own way and in their own time (Ilgen and Pulakos, 1999). Additionally, teams have become more and more important (Kozlowski and Ilgen, 2006; Salas et al., 2000). Consequently, without minute external guidance and managers’ leadership, employees are at the same time increasingly expected to take responsibility for and guide their own performance as well as fill their role as members of organizational teams.

A concept of self-influence which may support team members in this endeavour is individual self-leadership (Manz, 1986; Neck and Houghton, 2006). As a normative theory based on self-regulation theory (Carver and Scheier, 1981, 1998), intrinsic motivation (Deci, 1975; Deci and Ryan, 1985) and social cognitive theory (Bandura, 1986, 1991), it is aimed at individual self-improvement. Self-leadership theory suggests that a group of self-leadership strategies (i.e. behaviour-focused, constructive thought and natural reward strategies) positively influence the individual’s cognitive and motivational processes and, in turn, performance. Indeed, self-leadership strategies have proven to be positively related to self-efficacy and performance (Konradt et al., 2009; Prussia et al., 1998; see Neck and Houghton, 2006, for a review).

In organizational teams, however, in addition to purely individual task performance, employees also need to direct efforts towards the team goal in order to be effective. To date, no empirical research has explicitly addressed how self-leadership affects the behaviours of team members with regard to the team. Our goal is therefore to provide a first examination of self-leadership’s effects on team-oriented behaviours of team members. We argue that individuals’ self-leadership in a team will enhance not only individual task behaviours, but increase team member behaviours oriented towards the team, as well.

As both team and individual level moderators may influence this relationship, we examine task interdependence (Shea and Guzzo, 1987) and environmental uncertainty (Wall et al., 2002) as well as collectivism orientation (Wagner and Moch, 1986) as moderators of the self-leadership–team member behaviours relationship.

Moreover, most performance measures used in previous self-leadership research have been restricted to task performance. A recent comprehensive model of work role performance by Griffin, Neal and Parker (2007), which incorporates and extends earlier frameworks (e.g. Borman and Motowidlo, 1993; Crant, 2000; Pulakos et al., 2000), stresses the importance of performance aspects that go beyond task proficiency (i.e. adaptivity and proactivity). Although empirical evidence suggests the relevance of different performance aspects (Griffin et al., 2007), research on self-leadership to date has focused on a rather narrow performance concept and has largely neglected to examine the self-leadership–performance relation more broadly and distinctively. We therefore adapt this conceptualization of work role performance in order to gain a comprehensive view of self-leadership’s effect on performance in teams.

Taken together, the purpose of this research is to integrate and extend self-leadership research by examining the effect of self-leadership on a broad set of outcome measures, including team member proficiency, adaptivity and proactivity. Moreover, the moderating effects of environmental (i.e. task interdependence and uncertainty) and individual factors (i.e. psychological collectivism orientation) on the self-leadership performance relation will be explored.

Literature review and hypotheses

Self-leadership theory

Self-leadership theory is concerned with the self-influence that employees exert on themselves to achieve personal effectiveness (Manz, 1986; Neck and Houghton, 2006). Self-leadership expands previous concepts of self-management (Manz and Sims, 1980) in that individual self-leadership comprises not only self-regulation with regard to externally set standards, but future-oriented self-direction (Manz, 1986). Three major categories of self-influence strategies are distinguished (e.g. Manz and Neck, 2004; Neck and Houghton, 2006), including behaviour-focused strategies, constructive thought pattern strategies and natural reward strategies. This theoretically assumed three-factor structure of the construct with a general self-leadership factor has proven accurate in the measurement of self-leadership (Houghton and Neck, 2002).

The first category of self-leadership strategies denoted as behaviour-focused strategies involves observing, shaping and evaluating one’s own behaviour. Self-observation, self-goal setting, self-reward, self-punishment and self-cueing represent the five specific strategies. Self-observation is thought to enhance self-awareness and foster understanding for the when and how of one’s own behaviour, thus providing a departure point for identifying and improving ineffective or unproductive behaviours (Neck and Houghton, 2006). Self-goal setting refers to the process of identifying goals for oneself which lead to an improvement of personal performance (Manz, 1986; Manz and Neck, 2004). Finally, self-reward, self-punishment and self-cueing serve to shape behaviour in the following of these goals. External cues such as lists and notes keep the goal in focus during goal pursuit, while self-reward and self-punishment represent self-administered consequences according to the outcome. For a piece of especially good work, the self-leader may reward himself with something enjoyable, e.g. a walk in the park.

The second category of self-leadership strategies describes constructive thought pattern strategies which are designed to facilitate the management of cognitive processes and influence thinking patterns through self-analysis, imagining positive outcomes and positive self-talk (Manz and Neck, 2004). Constructive thought pattern strategies can help to identify dysfunctional cognitions such as ‘over-generalization’, where, for example, a colleague’s one critical remark translates into a complete disregard for all of one’s work, or ‘all-or-nothing thinking’ and replace them with constructive, positive self-talk and beliefs, e.g. opportunity thinking (Neck and Manz, 1992; Neck et al., 1997).

The third category indicates natural reward strategies, which make use of the inherent pleasurable aspects of an activity. By either focusing on the existing pleasant aspects of the work or by introducing more enjoyable features to the task, the work itself becomes rewarding (Manz and Neck, 2004; Neck and Houghton, 2006).

Self-leadership strategies are theoretically based on theories of behaviour (Bandura, 1986) and human motivation (Deci, 1975; Deci and Ryan, 1985). While these theories describe the mechanisms of the self-regulatory process, self-leadership provides guidelines for effective self-regulation. Individual goal pursuit entails a gradual convergence of behaviour and goal through a process of deviance reduction (Carver and Scheier, 1998). Self-leadership strategies are thought to enhance self-regulatory effectiveness by supporting all aspects of the self-regulatory deviance-reduction process, which encompasses the setting of goals, the process of goal pursuit and the constructive evaluation of outcomes (Manz, 1986; Neck and Houghton, 2006). Self-leadership is aimed specifically at the increase of self-efficacy (Bandura, 1986) and intrinsic motivation, which are thought to influence behaviour (Manz, 1986).

Though there is considerable theoretical and empirical evidence for the positive effects of external leadership on followers’ motivation and behaviour, the role of self-leadership strategies has mostly been discussed theoretically (see Neck and Houghton, 2006, for a review). Empirically, however, the evidence for the positive effect of self-leadership on performance remains sparse. Konradt et al. (2009), for instance, examined employees in organizational teams and demonstrated positive relations of self-leadership with individual performance. Consistent with this finding, students’ course performance was positively related to self-leadership in a different study (Prussia et al., 1998). Moreover, both studies found support for the mediating role of self-efficacy.

While these results provide support for the self-leadership–performance relationship that is suggested by self-leadership theory, the causal effect of self-leadership on performance has been demonstrated only for constructive thought pattern strategies (Neck and Manz, 1996). As there is no research causally linking all three self-leadership categories to performance in an organizational setting, as originally proposed (Manz, 1986), our aim is to first provide evidence for this posited direct causal effect. We therefore hypothesize

Hypothesis 1: Individual self-leadership leads to higher levels of individual task behaviours.

Self-leadership and performance in teams

The focus of Hypothesis 1 is on replicating the results of previous studies and adding confidence to the theoretically proposed causal relationship between self-leadership and individual performance. The following, however, goes beyond previous research as the aim is to broaden the understanding of individual self-leadership in teams.

In today’s organizations, teams have become increasingly common (Kozlowski and Bell, 2003; Salas et al., 2000). As a result, team success in addition to pure individual performance has turned into an important consideration for managers. But what constitutes a successful team? Several perspectives are possible.

Team-level performance appraisals concern the behaviour or output of the group ‘as a whole’, which is captured in one rating or score for all of its members. This can be obtained either through an expert or by aggregating team members’ individual ratings (DeNisi, 2000). The adequacy of either approach depends on several contextual and conceptual factors (for details, see Bliese, 2000; Kozlowski and Klein, 2000). Several models of team performance take on a team-level perspective to identify effective behaviours of the team as a whole which are related to team outcomes (LePine et al., 2008; Marks et al., 2001), i.e. teamwork processes. There is little research on the impact of individuals’ self-leadership on these team-level phenomena. Self-leadership behavioural strategies have been shown to be positively related to team performance, mediated by job satisfaction (Politis, 2006). Uhl-Bien and Graen (1998) demonstrated a positive relationship between aggregate individual self-management and unit effectiveness in functional teams, suggesting that self-leading team members may have a positive impact on team performance. While this view can be important for the investigation of possible synergetic effects that can occur in the team as a whole, the conceptualization of team performance as a holistic concept at the team level, however, precludes an identification of individual contributions to team success.

The ‘bottom-up’ perspective on team performance, in contrast, considers team processes and performance at least in part as a concept that emerges as a result of team members’ individual efforts (Kozlowski and Ilgen, 2006; Kozlowski and Klein, 2000). Indeed, Chen (2005) demonstrated that individual team members’ performance can affect team performance in this way. Theoretically, important distinctions between different types of emergence (compositional vs. compilational) have been made (Kozlowski and Klein, 2000). These types of emergence specify the impact of individual-level phenomena (‘elemental content’, Kozlowski and Klein, 2000: 55) on team-level constructs. For the purpose of this study, however, this distinction is not as relevant as the basic assumption that individual, team-directed behaviours can and do contribute to team performance. We adopt this ‘bottom-up’ perspective to explore how the individual self-leading team member can contribute to enhanced team performance by analysing the influence of individual self-leadership on individual team member behaviours, i.e. individual behaviour that is assumed to be beneficial for team functioning and performance as a whole (Griffin et al., 2007).

The sparse existing research on self-leadership in teams does not allow conclusions with regard to these individual-level, team-directed team member behaviours. Although self-leaders’ performance in teams has been assessed and team-level moderators were examined (Konradt et al., 2009), the outcome measure used in this research (global supervisor ratings of performance) was too undifferentiated to draw conclusions on individual team-directed behaviours. Conceptually, attempts have been made to examine the role of self-leaders performance-enhancing contributions in teams. Bligh, Pearce and Kohles (2006) provide examples of self-leading team members who may set themselves the goal to present one new idea per team meeting or rehearse important team-related skills such as an oral presentation. The authors illustrate how self-leadership skills might be used to influence one’s own behaviour in a way that has positive effects for the team, but do not test their model empirically. Our goal is to provide empirical evidence for a relationship between self-leadership and team member behaviours.

In a team context, self-regulation is required not only with regard to individual goals, but also with regard to the team goal (DeShon et al., 2004). We propose that self-leadership’s strategies of behaviour-focused strategies, natural reward strategies and constructive thought pattern strategies can be used together to reduce discrepancy from the team goal in the same way that they improve individual performance, i.e. by supporting and enhancing goal setting, goal pursuit and constructive outcome evaluation. When an individual’s behaviour does not contribute to team goal attainment, effective self-regulation entails a change of behaviour to reduce discrepancy from this goal. By enabling effective self-regulation, self-leadership allows team members to influence their own behaviour so that it contributes to team goal attainment. Resulting behavioural changes, i.e. increased performance, will therefore be directed not only at individual tasks, but also at team member responsibilities. Based on this rationale we hypothesize

Hypothesis 2: Individual self-leadership leads to a greater occurrence of team member behaviours.

The extent, however, to which team member behaviours conceptualized as team-directed behaviours actually contribute to the attainment of team goals is highly dependent on the degree to which the team task is interdependent, i.e. the degree to which team members need to interact in order to fulfil a given task (Shea and Guzzo, 1987). Task interdependence is recognized as an important situational moderator for the influence of other variables on team effectiveness (Kozlowski and Bell, 2003; Langfred, 2000). High task interdependence enhances the effect of individual behaviours on team-level outcomes, while purely additive outcome interdependence or low task interdependence result in the team outcome being mainly dependent on team members’ individual task performance and less on teamwork. Effective self-regulation entails an appropriate distribution of limited resources. For this reason, self-leadership is expected to influence team-oriented behaviours more strongly under conditions of higher interdependence, where self-leaders judge their team-oriented behaviours to have a greater impact on team goal attainment. We therefore propose

Hypothesis 2a: High task interdependence leads to a stronger relationship between self-leadership and team member performance.

Another aspect which is sparsely reflected in the empirical self-leadership research relates to the multidimensional nature of the performance construct (Campbell et al., 1993). As concepts such as organizational citizenship behaviour (Bateman and Organ, 1983), which comprises employee behaviour that goes beyond formal job requirements, were introduced, conceptualizations of job performance were broadened to include a distinction between task and contextual performance (Borman and Motowidlo, 1993). Additionally, research has stressed the importance of adaptive (Pulakos et al., 2000) and proactive behaviours (e.g. Crant, 2000) for individual and organizational effectiveness. Adaptive and proactive behaviours are especially important in teams where variability in the environment, which makes tasks and requirements unpredictable, leads to situational uncertainty (Wall et al., 2002). People in uncertain environments are required to quickly adjust to changing tasks and demands. Beyond this need for adaptivity, there is a greater opportunity to proactively initiate changes in the environment (Griffin et al., 2007).

A comprehensive framework of job performance which integrates prior conceptualizations has been proposed by Griffin et al. (2007). Drawing on and expanding organizational role theory (Katz and Kahn, 1978), the taxonomy distinguishes three different kinds of individual work role behaviours (i.e. proficiency, adaptivity and proactivity) and three different levels at which these contribute to effectiveness (i.e. individual, team and organization). Apart from being theoretically plausible, the structure of the resulting nine-factor model is supported empirically (Griffin et al., 2007). Differential relations of the resulting nine factors to other variables as well as the predictive power of the proactivity dimensions lend support to the validity of the model, which has been recently suggested by Frese (2008) as the best current performance concept available for modern workplaces. Hence, we follow this conception to examine the effect of individual self-leadership on team member work role performance by relating self-leadership to team member proficiency, adaptivity and proactivity.

Proficiency

Proficiency refers to the degree to which individuals’ activities and behaviours that are part of the official job requirements and thus the individual’s work role are carried out. Besides individual task proficiency, proficiency as a team member involves coordination, cooperation and information exchange with other members. Recent reviews of teamwork behaviours (e.g. Kozlowski and Bell, 2003; Rousseau et al., 2006) emphasize that individuals’ collaborative efforts are central to team effectiveness, and demonstrate a positive association with team performance (Taggar and Brown, 2001). However, little research has examined the relationship between self-leadership and team member proficiency. The effectiveness measure which Uhl-Bien and Graen (1998) found to be related to self-management did contain a teamwork effectiveness dimension, but ratings were not made at the individual level. While there thus seems to be some relationship, the exact nature of individual self-leadership’s influence on teamwork remains unclear.

In a similar vein to above, we argue that individual self-leadership strategies positively influence team member proficiency. We propose that deviations from team standards are met by effective self-regulation through a change of behaviour to a more team-directed focus. Because teamwork behaviours are fundamentally necessary to reach team goals, self-leaders will recognize this and, in turn, intensify their team-directed collaborative behaviours.

Self-leadership can improve team member proficiency in several ways. The employee could, for example, set himself/herself the goal to talk to every team member once a day. Dysfunctional thoughts might lead him/her to believe that this annoys his/her colleagues, but constructive thought pattern strategies can help to alleviate this worry. A chat with team members can, at the same time, pose a natural reward. A person who does not enjoy talking to the colleagues might employ self-leadership skills to schedule a different, more pleasing, task afterwards. Employed in this way, we propose

Hypothesis 3: Team members’ individual self-leadership leads to a greater occurrence of team member proficiency.

Following the same reasoning as above, we extend Hypothesis 3 and expect that high task interdependence leads to a stronger relationship between self-leadership and team member proficiency (Hypothesis 3a).

Adaptivity

Adaptive performance refers to work-relevant accommodation behaviours in response to changes in the work (team) environment (Griffin et al., 2007). Adaptive behaviours are particularly important in new types of work such as virtual collaboration (Powell et al., 2004), but can be required in any kind of team which is submitted to change. Problem solving, learning new processes or skills, coping with stress or adapting to new structures are examples of adaptive behaviour that, when shown, enhances team effectiveness in a changing environment (Griffin et al., 2007; Pulakos et al., 2000). While task interdependence increases the contribution to team effectiveness for team-oriented behaviours in general, adaptive team member behaviours become especially important when the environment is unpredictable.

Research has neglected to explicitly investigate the relationship between self-leadership and team members’ adaptive behaviours. Constructive thought pattern strategies have been advocated as effective instruments for developing a constructive reaction to changes in the team environment (Neck, 1996) and indeed seem predestined as tools to help cope with change. Mental imagery provides a means to go through the change ahead of time and has been shown to improve performance not only in sports, where this technique is widespread (Feltz and Landers, 1983), but, more recently, also in business contexts such as employment interviews (Knudstrup et al., 2003). Self-analysis can discover initial fears, for example about the implementation of a new shared desktop system, (e.g. ‘I will never be able to use this’) which can be turned into willingness to learn (e.g. ‘I will try my very best’) through positive self-talk. The other self-leadership strategies can also support and foster a constructive reaction to change. Behaviour-focused strategies can be used to monitor and guide the actual adaptation behaviour. Focusing on the pleasurable aspects of the new experience shows use of natural reward strategies. All these self-leadership strategies should facilitate the adaptation to changes in the team environment. Drawing on this reasoning, we propose

Hypothesis 4: Team members’ individual self-leadership leads to a greater occurrence of team member adaptivity.

Additionally, building on the same reasoning as above, we expect high task interdependence to lead to a stronger relationship between self-leadership and team member adaptivity (Hypothesis 4a).

As adaptive behaviours are especially important for the team under uncertain conditions, we propose high situational uncertainty to lead to a stronger relationship between self-leadership and team member adaptivity (Hypothesis 4b).

Proactivity

Proactivity refers to self-initiated behaviours in anticipation or initiation of change. Proactivity is shown, for instance, by suggesting improvement regarding one’s own work behaviours or given work routines. Griffin et al. (2007) argue that proactive behaviour of a team member focuses on the improvement of team functioning as a whole, which manifests itself, for example, in the suggestion of new routines such as team meetings, implementation of innovative methods of coordination or cooperation, or introduction of alternative communication channels. Proactive behaviour seems particularly meaningful in relation to self-leadership because self-leadership does not only entail self-regulation within given boundaries, in other words discrepancy reduction, but explicitly encompasses the active creation of discrepancy through the questioning of existing structures and routines (Manz, 1986; Neck and Houghton, 2006). Neck, Ashcraft and Vansandt (1998) suggested that self-leaders may thus proactively shape and influence their work environment in the process of creating new standards and effective procedures. Empirically, self-leadership has been shown to be positively related to employees’ innovative behaviours (Carmeli et al., 2006). Self-efficacy, a mediator between self-leadership and performance (Konradt et al., 2009; Prussia et al., 1998) has been shown to serve as an antecedent of proactive or taking-charge behaviour at work (Griffin et al., 2007; Morrison and Phelps, 1999).

Self-leadership skills may lead to team member proactivity in several ways: both through awareness of areas of improvement as well as through motivation to change the status quo. By using behavioural strategies, for example self-observation of his or her own work within the team, the team member may notice when there are areas in which the team would benefit from an improvement. Constructive thought pattern strategies may support the employee in making new suggestions and overcoming initial doubts (Neck, 1996). Natural reward strategies could lead to the restructuring of a task so that it contains more pleasant aspects. Consequently, we propose

Hypothesis 5: Team members’ individual self-leadership leads to a greater occurrence of team member proactivity.

Following the reasoning made earlier, we expect high task interdependence to lead to a stronger relationship between self-leadership and team member proactivity (Hypothesis 5a).

Following a similar reasoning, we additionally propose that high situational uncertainty leads to a stronger relationship between self-leadership and team member proactivity (Hypothesis 5b).

Moderating individual factors

We have discussed situational moderators that we expect to influence the effect of self-leadership on team member behaviours. However, a somewhat more proximal consideration regards individual moderators: does the application of self-leadership strategies influence the behaviour of different persons in the same way? Not everyone is inclined towards teamwork to the same degree. Psychological collectivism describes an individual’s orientation towards the group which accords importance to the group goal and the well-being of the group (Wagner, 1995). High levels of psychological collectivism have been shown to be related to higher levels of group member performance (Jackson et al., 2006).

For team members, self-leadership strategies direct the focus on the team goal and support the regulation behaviour in pursuit of this goal. Team members who are not naturally inclined towards the group (low collectivism) should show a greater effect of of self-leadership strategies than those team members who direct effort towards the group anyway due to a high collectivism orientation. Therefore, we propose

Hypothesis 6: In relation to team member proficiency, adaptivity and proactivity, low collectivism orientation is expected to lead to a stronger relationship between self-leadership and performance.

Present research

To explore the causal effects of self-leadership on the dependent variables, we used an experimental method. Policy capturing is a methodology in which participants are asked to answer to a number of hypothetical scenarios containing the experimentally manipulated independent variables as cues. Responses that indicate how one would choose or would behave are then regressed on the cues to examine the individual effects (Aiman-Smith et al., 2002; Karren and Barringer, 2002). Policy-capturing methodology has been used in numerous research fields, including performance appraisal (Rotundo and Sackett, 2002), job choice (Judge and Bretz, 1992) and work–family balance (Greenhaus and Powell, 2003). Research has, in general, shown policy capturing to be an effective method (e.g. Cable and Judge, 1994; Dineen et al., 2004; Harold and Ployhart, 2008; Kristof-Brown et al., 2002).

However, there are notable challenges inherent to policy-capturing methodologies. A key issue in policy-capturing research is the manipulation of variables, creating realistic scenarios. For this reason, our scenarios were pre-screened by experts (organizational psychologists) familiar with the self-leadership construct with regard to content and operationalization and non-psychologists with regard to comprehensibility. To further foster participants’ ability to relate to the presented scenarios, we ensured that participants had experience relevant to the judgment of the scenarios (Aiman-Smith et al., 2002; Graham and Cable, 2001), i.e. prior work experience.

Another potential concern in policy-capturing methods is participant fatigue (Graham and Cable, 2001). This factor restricts the number of manipulated variables and levels as there is a limit to the number of judgments that participants can accurately make. For this reason, we restricted the manipulations to maximally three variables with maximally three levels. This limitation entails that scenarios are not necessarily reflective of naturally occurring situations, as other factors possibly influencing the dependent variables are neglected. On the other hand, the isolated examination of factors allows drawing conclusions about causality. For this reason, we chose policy capturing as a useful method for gaining initial evidence on the causal effects of self-leadership in teams.

Study 1 examines whether self-leadership has a causal effect on individual task behaviours (H1) and team member behaviours (H2), and investigates the moderating effect of task interdependence on team member behaviours (H2a). Study 2 extends Study 1 by examining team member behaviours resulting from self-leadership in more detail and investigating individual and team variables as moderators of this relationship (H3, H4, H5 and H6).

Study 1

Method

Participants

A total of 85 participants, 67.1 % women and 20.6% men with a mean work experience of 12.08 years (SD = 10.99), were approached in public places (e.g. train stations, airports, evening classes) in Germany. In order to qualify for participation, participants had to orally indicate that they had work experience before starting the questionnaire. Mean age was 34.8 years (SD = 12.22). As their highest level of education, 30.7% stated secondary school, 9.4% had taken University of Applied Sciences entrance exams, 33.0% university entrance exams, and 23.5% held a university degree. In exchange for their participation, participants could win gift vouchers for local stores.

Materials

Self-leadership (high/medium/low) and task interdependence (high/medium/low) were manipulated in several short descriptions of work situations. Participants were asked to indicate how they would perform in each situation (see Table 1). For each set of strategies, one was chosen to represent the category based on factor loadings (Houghton and Neck, 2002), taking into account the need for probable manipulations. Behavioural strategies were represented through goal setting either by self (high self-leadership), by self and supervisor (medium self-leadership), or by supervisor only (low self-leadership). Natural reward strategies were captured through focusing on pleasurable aspects of task (high self-leadership), focusing on task (medium self-leadership), or focusing on finishing (low self-leadership). Mental imagery was chosen to represent constructive thought pattern strategies. ‘Mental rehearsal of success’ represented high self-leadership; ‘mental rehearsal only’ stood for medium self-leadership; and ‘no mental rehearsal’ represented low self-leadership. In each scenario, all three individual self-leadership strategies were manipulated to represent the same level of self-leadership.

Table 1.

Manipulations of independent variables.

	Self-leadership	Task interdependence	Uncertainty
Low	Your specification is to improve performance by 10%. You start directly with your task without thinking about it for longer. While working, finishing is most important to you.	In order to finish the task, you and your teammates barely depend on each other.	Unusual problems seldom arise in your work.
Medium	You and your boss have agreed to aim at an improvement of performance by 10%. Before you begin your task, you go through it once in your mind. While working, you concentrate on your task	In order to finish the task, you and your teammates somewhat depend on each other.
High	You personally have set yourself the goal to improve your performance by 10%. Before you begin your task, you go through it once in your mind and imagine yourself succeeding. While working, you remind yourself of the pleasurable aspects of the task.	In order to finish the task, you and your teammates very much depend on each other.	Unusual problems often arise in your work.

Task interdependence was manipulated by stating, ‘To finish the task, you and your colleagues [very much/somewhat/barely] depend on each other.’ A full factorial approach resulted in nine (3 × 3) scenarios. An example of a full scenario and all cue levels are given in Table 1. Additionally, three scenarios were replicated at the end of the questionnaire to estimate re-test reliability (cf. Cable and Graham, 2000; Karren and Barringer, 2002). Sequence effects of the scenarios and the order of cues within each scenario were controlled for and revealed no significant effect on the dependent variables, all Fs < 1.

Measures

All items were responded to using five-point Likert scales ranging from ‘not at all’ (1) to ‘very much’ (5).

Individual task behaviours. Three items were adapted from Griffin et al. (2007). One item from each of the dimensions task proficiency (‘I carry out my task well’), task adaptivity (‘I learn new skills to help me adapt to changes in my tasks’) and task proactivity (‘I come up with ideas to improve the ways in which my tasks are done’) was chosen based on the highest factor loadings to represent individual task behaviours, i.e. behaviours related to the attainment of goals associated with the individual’s task. After rephrasing to form complete sentences, a three-item ‘individual task behaviour’ scale (Cronbach’s α = .81, re-test reliability = .76) was obtained.

Team member behaviour was measured with three items of the dimensions team member proficiency (‘I communicate effectively with my colleagues’), team member adaptivity (‘I learn new skills or take on new roles to cope with changes in the way my team works’) and team member proactivity (‘I improve the way my team does things’) adapted from Griffin et al. (2007). The resulting three-item ‘team member behaviours’ scale showed satisfactory reliability values (Cronbach’s α = .86, re-test reliability = .76).

Although Griffin et al. (2007) provide clear evidence for the validity of the nine-factor structure with a three-item measurement of each scale, we reduced each scale to one item and thus assumed higher-level individual task role and team member role factors. In the original model, this aggregation of the subscales is not suggested and does not receive good empirical support; however, in our study, a multilevel factor analysis taking into account the nesting of responses (level-1) within respondents (level-2) due to the repeated measures design (Hox, 2002) succeeded in separating team member behaviours from individual task behaviour ratings (χ²= 38.01, df = 20, p = .94; RMSEA = .05 for the two factors model; χ²= 556.08, df = 23, p < 0.01; RMSEA = .25 for the one factor model). This provides evidence for the construct validity of the dependent variables.

Controls

Because of the experimental nature of our study, we focus on controlling for variables that might influence the effect of the experimental manipulation. Previous work experience might affect the degree to which participants could relate to the scenarios (Aiman-Smith et al., 2002). We therefore controlled for participants’ work experience in years. Additionally, a three-item manipulation check measuring the ease of imagining the scenarios was presented at the end of the questionnaire (e.g. ‘I could easily put myself in these situations’, Cronbach’s α = .89). As this might also influence responses, we controlled for this variable.

Analyses

Because each participant responded to several scenarios, the data we obtained can be considered as nested (see Figure 1 for an illustration). Nested data violates the assumption of independence of ordinary least squares regression analyses, as characteristics of level-2 units (in this case, participants) can influence several level-1 units (scenarios). Due to the nested nature of the data, hypotheses were tested with hierarchical linear modeling (HLM) using HLM 6 (Raudenbush et al., 2004; for an excellent introduction to hierarchical models, see Hofmann, 1997). HLM takes into account both within- and between-person variance and thus offers an advantage over ordinary least squares regression analysis, where between-person variance is neglected (cf. Kristof-Brown et al., 2002). For each participant, HLM calculates unique regression coefficients based on the responses to the different scenarios (level-1). Each person thus is assigned a unique intercept, slope and error. The level-1 coefficients (intercepts and slopes) are then regressed onto level-2 predictors. Level-2 coefficients therefore represent the grand mean (intercept) and the average slopes (βs). Remaining variance is modelled at between-person level-2. Level-2 analyses allow the examination of person variables as moderators of the within-person relationship between the predictors and dependent variables (across scenarios). A check of model assumptions revealed adequate normality of residuals and homogeneity of variances on both levels (cf. Hoffman, 1997). Missing data on the dependent variables (less than 1.2%) and the control variables (3.5%) were substituted by using expectation maximization (EM) in SPSS. This procedure is recognized as appropriate for small amounts of missing data (Tabachnik and Fidell, 2001).

We then regressed self-leadership, task interdependence and the interaction term after centring the variables as predictors on level-1, running two separate models for individual task behaviours and team member behaviours. In both, we entered work experience and the manipulation check (both grand mean centred) as level-2 controls, so that the β coefficients in Table 3 represent the slopes for a participant with average work experience and an average value on the manipulation check variable (Hofmann, 1997).

Table 2.

Means, standard deviations, and correlations among study variables in Study 1.

Variable	M	SD	1	2	3	4
1 Individual task behaviours^a	4.14	0.76
2 Team member behaviours^a	3.48	0.96	.46***
3 Work experience (years)^b	12.08	10.99	.26*	.10
4 Manipulation check^b	3.34	0.93	.17	.06	.23*
5 Self-leadership^a,c	2.00	0.82	.38***	.15***	.00	.00
6 Task Interdependence^a,c	2.00	0.82	.02	.43**	.00	.00

Notes. ^aN = 765 scenarios. ^bN = 85 participants. ^cBecause the study utilized a completely crossed design, correlations among independent variables are zero by definition, and therefore are not shown.

p < .05, **p < .01, ***p < .001 (two-sided).

Table 3.

Hierarchical linear models of self-leadership, interdependence and interaction on individual and team member behaviours.

	Individual task behaviours			Team member behaviours
	β	SE	t	β	SE	t
Intercept
Intercept	4.14***	0.05	80.76	3.48***	0.07	51.99
Work experience	0.01*	0.01	2.15	0.01	0.01	0.89
Manipulation check	0.06	0.07	1.07	0.03	0.07	0.38
Self-leadership (SL)
Intercept	0.35***	0.05	7.84	0.17**	0.07	2.40
Work experience	−0.01	0.00	−0.13	0.01*	0.00	1.97
Manipulation check	0.02	0.05	0.34	0.03	0.04	0.83
Interdependence (INT)
Intercept	–	–	–	0.50***	0.07	6.81
Work experience	–	–	–	0.00	0.00	0.71
Manipulation check	–	–	–	−0.02	0.04	−0.34
SL × INT
Intercept	–	–	–	0.00	0.03	0.10

Notes. N = 85.

p < .05, **p < .01, ***p < .001 (two-sided; directional tests when hypotheses specified).

Results

Means, standard deviations and bivariate correlations are presented in Table 2. Results of HLM analyses can be found in Table 3.

Hypothesis 1 predicted that self-leadership would lead to higher individual task behaviours. The coefficient predicting the average self-leadership slope for individual task behaviours differed significantly from zero (β = .35, p < .001), supporting this hypothesis. Also, in support of Hypothesis 2, which predicted that higher levels of self-leadership would lead to increased efforts towards the team, the coefficient for the average self-leadership slope predicting team member behaviours also differed significantly from zero (β = .17, p < .05). The interaction between self-leadership and task interdependence for team member behaviours, as stated in Hypothesis 2a, failed to reach significance (β = .00, ns.). Self-leadership led to increased team member behaviours regardless of the degree of task interdependence in the team.

Discussion

The results from Study 1 are in line with previous studies indicating a positive effect of individual self-leadership strategies on individual task performance, with the experimental design adding confidence to the theoretically proposed causal relationship. More importantly, this study provides first evidence of a relationship between self-leadership and team member behaviours. This supports previous cross-sectional findings on the impact of self-leadership strategies in teams (Konradt et al., 2009; Uhl-Bien and Graen, 1998). However, this study is the first to allow conclusions on the exact nature of the effect of individual self-leadership on team members’ behaviour.

To assess the robustness of the results of Study 1, we conducted a replication study with a new sample and further investigated the effect of self-leadership on team member behaviour. More specifically, we examined the more differentiated outcome measure including team member proficiency, adaptivity and proactivity (cf. Griffin et al., 2007). Additionally, the moderating effect of uncertainty as a situational variable as well as the moderating effect of collectivism orientation, an individual variable, was investigated.

Study 2

Method

Participants and procedure

The participants were 63 participants with work experience (M = 12.35, SD = 10.85). Mean age was 34.7 years (SD = 12.0). The majority of participants were male (55.6%), 39.7% were female. As their highest level of education, 49.2% stated secondary school, 11.1% had taken University of Applied Sciences entrance exams, 17.5% university entrance exams, and 22.2% held a university degree. Participants could again win vouchers in exchange for their participation.

The study design and procedure were identical to those used in Study 1.

Material

Self-leadership (high/medium/low), task interdependence (high/low) and uncertainty (high/low) were manipulated as independent variables in a full factorial design. The self-leadership manipulation was similar to that employed in Study 1. The task interdependence manipulation was reduced to two categories, so that the scenarios read ‘To finish the task, you and your colleagues must [very much/barely] depend on each other’. Uncertainty was manipulated through the sentence ‘Unexpected problems [often/seldom] arise in your work’. Participants therefore had to respond to 12 (3 × 2 × 2) plus 3 replicated scenarios to calculate re-test reliability. As no effects of cue order were found in Study 1, all scenarios contained the cues in the order ‘self-leadership–task interdependence–uncertainty’. Two different orders of the scenarios were employed.¹

Measures

Participants responded using five-point Likert scales ranging from ‘not at all’ (1) to ‘very much’ (5).

Team member behaviours. To measure different aspects of team member behaviours, we used three dimensions from Griffin et al. (2007).

Team member proficiency. The three-item scale was adapted from Griffin et al. (2007), with the items rephrased to form complete sentences (e.g. ‘I coordinate my work with my coworkers’). Internal consistency and re-test reliability were satisfactory (α = .91, re-test reliability = .74).

Team member adaptivity. The three items of Griffin et al. (2007) were rephrased (e.g. ‘I deal effectively with changes affecting my team’) and combined to form a ‘team member adaptivity’ scale. Internal consistency and re-test reliability were satisfactory (α = .89, re-test reliability = .70).

Team member proactivity. The three-item scale of team member proactivity of Griffin et al. (2007) was used with rephrased items (e.g. ‘I suggest ways to make my work unit more effective’). Internal consistency and re-test reliability were acceptable (α = .93, re-test reliability = .73).

A multilevel confirmatory factor analysis was computed due to the nesting of responses (level-1) within (level-2). The three-factor solution (χ²= 89.66, df = 54, p < .01; RMSEA = .04) showed a better fit than the one-factor solution (χ²= 1001.56, df = 62, p < .001, RMSEA = .20), providing construct validity for the performance dimensions.

Collectivism orientation. A 15-item scale by Jackson et al. (2006) was used to measure collectivism orientation at the end of the questionnaire. Participants answered 15 statements regarding preference for working in groups, reliance on group members, concern for group members, norm acceptance and goal priority, and indicated their agreement on five-point Likert scales ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (5). The reliability (α) of the measure was .91.

Controls

The same controls as in Study 1 were used.

Analyses

HLM was again used to analyse the data. As the distributional assumptions of normality of residuals and homoscedasticity were not met for all models, robust standard errors are reported (Hofmann, 1997; Hox, 2002). Missing data on the dependent variables (less than 1.8%), collectivism items (<1.7%) and the control variables (<8.0%) were imputed through an EM algorithm in SPSS. In a first step, the relevant independent variables and the interaction terms were entered as level-1 predictors into a model for each of the three team member behaviour dimensions. All predictors were grand mean centred (cf. Hofmann, 1997). In a second step, to investigate the cross-level interaction between self-leadership and collectivism orientation, we predicted the team member behaviours outcome variables by self-leadership, task interdependence and uncertainty on level-1 (within person) and collectivism on level-2 (between persons).² Again, all predictors were centred on their grand mean to allow slope coefficients to represent the change associated with the variable per unit increase over the variable’s mean (Hofmann and Gavin, 1998).³

As the control variables did not show significant effects on the hypothesized relationships in Study 1, we present the models without controls in Study 2 in order to increase power. Analyses including controls showed no change among variables and in hypothesized relationships.

Results

Means, standard deviations and bivariate correlations are presented in Table 4. As shown in Tables 5 and 6, self-leadership has again a significant positive effect on all three categories of team member behaviours (β_proficiency = .31, β_adaptivity = .37, β_proactivity = .52; all p < .01), which provides support to Hypotheses 3 to 5. The corresponding standardized coefficients (calculated following Hox, 2002) were β_prof = .23, β_ada = .30, β_proa = .35, indicating a moderate effect size (cf. Cohen, 1988). People using self-leadership strategies would fulfil their role as team members better, adapt better to changes in the team and proactively make suggestions on improvements regarding the team. The interactions between self-leadership and task interdependence (H3a, H4a, H5a) and the interactions between self-leadership and uncertainty (H4b, H5b), respectively, were non-significant (β = −.08 to .03, ns.). Self-leadership had the same positive effect on all three dimensions of team member behaviours regardless of task interdependence and uncertainty.

Table 4.

Means, standard deviations, and correlations among study variables in Study 2.

Variable	M	SD	1	2	3	4	5
1 Team member proficiency^a	3.73	1.07
2 Team member adaptivity^a	3.59	1.03	.75***
3 Team member proactivity^a	3.47	1.15	.70***	.74***
4 Collectivism^b	3.60	0.73	.30*	.45***	.25
5 Work experience^b	12.35	10.85	.20	.19	.25*	−.03
6 Manipulation check^b	3.57	0.88	.03	.06	.10	.08	.17
7 Self-leadership^a,c	2.00	0.82	.25***	.29***	.30***	–	–
8 Task Interdependence^a,c	1.50	0.50	.38***	.25***	.31***	–	–
9 Uncertainty^a,c	1.50	0.50	.00	.07	.04	–	–

Notes. ^aN = 756 scenarios. ^bN = 63 participants. ^cBecause the study utilized a completely crossed design, correlations among independent variables are zero by definition, and therefore are not shown.

p < .05, **p < .01, ***p < .001 (two-sided).

Table 5.

Hierarchical linear models of self-leadership, interdependence, uncertainty and interactions on team member proficiency, team member adaptivity and team member proactivity.

	Team member proficiency			Team member adaptivity			Team member proactivity
Level-2	β	SE	T	β	SE	t	β	SE	t
Intercept
Intercept	3.71***	0.07	52.60	3.60***	0.08	48.23	3.48***	0.08	42.55
Self-leadership (SL)
Intercept	0.31**	0.11	2.83	0.37**	0.13	2.83	0.52***	0.15	3.41
Interdependence (INT)
Intercept	0.83***	0.17	4.93	0.46**	0.15	3.03	0.85***	0.17	5.03
Uncertainty (UNC)
Intercept	–	–	–	0.17	0.14	1.24	0.05	0.16	0.33
SL × INT
Intercept	0.02	0.06	0.29	0.03	0.08	0.35	−0.08	0.08	−0.99
SL × UNC
Intercept	–	–	–	−0.02	0.06	−0.41	0.02	0.08	0.24

Notes. N = 63. *p < .05, **p < .01, ***p < .001 (two-sided; directional tests when hypotheses specified).

Table 6.

Hierarchical linear models of self-leadership, interdependence, uncertainty and collectivism orientation on team member proficiency, team member adaptivity and team member proactivity.

	Team member proficiency			Team member adaptivity			Team member proactivity
Variable	β	SE	t	β	SE	t	β	SE	t
Intercept
Intercept	3.71***	0.07	55.12	3.60***	0.07	53.94	3.48***	0.08	53.90
Collectivism	0.23**	0.08	2.88	0.39***	0.08	5.27	0.24**	0.09	2.74
Self-leadership
Intercept	0.33***	0.04	7.55	0.38***	0.04	8.50	0.44***	0.05	8.72
Collectivism	−0.12*	0.06	−1.93	−0.03	0.07	−0.45	−0.08	0.07	−1.14
Interdependence
Intercept	0.86***	0.10	8.92	0.51***	0.07	7.29	0.70***	0.08	9.37
Collectivism	−0.24*	0.12	−1.98	−0.02	0.10	−0.21	−0.11	0.12	−0.91
Uncertainty
Intercept	–	–	–	0.12	0.06	1.96	0.09	0.06	1.51

Notes. N = 63. *p < .05, **p < .01, ***p < .001 (two-sided; directional tests when hypotheses specified).

The hypothesized interaction between self-leadership and collectivism as stated in Hypothesis 6 does not find strong support. Collectivism has a weak moderating effect only on the relationship between self-leadership and team member proficiency (β = −.12, p < .05). The negative slope coefficient indicates that self-leadership is especially effective in improving team member proficiency for participants with low collectivism orientation. As shown in Figure 2, the effect of self-leadership on team member adaptivity and team member proactivity did not, contrary to Hypothesis 6, change with varying levels of collectivism.

General discussion

The purpose of this study was to examine the effect of self-leadership on team member behaviours. The results from the experimental design provide evidence that self-leadership plays a causal role in enhancing team members’ performance behaviour.

Previous studies have indicated that self-leadership may have positive effects on team or team member performance (Konradt et al., 2009; Uhl-Bien and Graen, 1998), but have not investigated through which concrete behaviours self-leaders actually contribute to team effectiveness. Our results, however, expand the understanding of the effects of self-leadership due to the differentiated measure of performance, which allowed the investigation of self-leadership’s effect on several dimensions, especially team-oriented behaviours. Though only intended behaviour was measured, our findings hence empirically support theoretical postulations on the effects of self-leadership on a wider range of performance behaviours (Manz, 1986; Manz and Neck, 2004; Neck and Houghton, 2006) and provide insight into the processes that can explain previous studies’ findings. The fact that self-leadership influences more and different performance aspects than previously investigated, i.e. those that go beyond task performance (Borman and Motowidlo, 1993) is highlighted by the finding that stronger effects of self-leadership on those behaviours that were previously neglected in empirical investigations, team member adaptivity and proactivity, were found.

The hypothesized moderating effects of task interdependence and uncertainty examined in these studies failed to reach significance. We can surmise that power problems might have prevented the detection of effects because categorical interactions in regression-based designs generally suffer from lack of power (see Aguinis et al., 2001). Unfortunately, if this is the case, no definite statement on the influence of interdependence and uncertainty can be made. Otherwise, considering possible non-linear effects that may have prevented the moderating effects from being detected could provide an interesting research avenue.

Self-leadership was shown to be particularly effective in increasing team member proficiency for participants with a low collectivism orientation. This is comparable to the result that self-leadership training has been shown to be more effective for individuals with low conscientiousness scores (Stewart et al., 1996). Self-leadership does indeed seem to offer a means for self-improvement such that individual inclinations that are not appropriate for the situation can be overcome through strategic self-regulation. The fact that only the relationship between self-leadership and team member proficiency and not between self-leadership and team member adaptivity and proactivity was moderated by collectivism orientation may be due to a higher perceived centrality of team member proficiency. People less inclined to interact with the team may direct their effort towards the most central behaviours (proficiency) as a result of self-leadership rather than behaviours seen as less important for the attainment of team goals (adaptivity and proactivity).

Limitations and strengths

This research is possibly limited by its sample and design. The main limitation of the research is a potential limit to the external validity of the results for behaviour in actual organizational teams, as with all policy-capturing research (Aiman-Smith et al., 2002). In policy capturing, individuals are presented with complete information about the situations they are asked to evaluate while in actual work contexts, team members are forced to behave with incomplete information. Although research suggests that policy-capturing results can show a high generalizability (Rogelberg et al., 1999), the quality of the results depends on the quality of the scenarios. We took multiple steps to alleviate this concern and to assure high external validity (cf. Karren and Barringer, 2002). First, our operationalizations of the independent variables were judged on comprehensibility and accurate reflection of realistic variable levels by both experts and a pilot sample similar to the actual participants. Second, sufficient re-test reliabilities indicate that a necessary condition for validity was fulfilled. Third, the orthogonal cues do not, as often in policy-capturing research, pose a problem, as our independent variables can indeed be uncorrelated in real work settings. Despite these efforts, we acknowledge that a scenario-based survey may not fully generalize to actual experiences with behaviour in real organizational settings. Thus, future longitudinal research in real organizational settings should examine if our results are generalizable and will also provide more insights into the validity of policy-capturing results.

A further potential limitation refers to the operationalization of the first order self-leadership strategies by single second-order strategies as indicators. This strategy was necessary to reduce the potential concern of participant fatigue in larger scenarios (cf. Graham and Cable, 2001). The sparse and limited operationalization of first-order leadership strategies might cast doubt on the content validity of our results. However, results are consistent with other findings obtained in cross-sectional studies (e.g. Carmeli et al., 2006; Konradt et al., 2009;). Moreover, as more comprehensive and complete descriptions will possibly result in more vivid imagination of self-leadership and thus will induce stronger effects, we conclude that our results might be conservative. Future research should continue to explore this issue.

Despite these possible caveats, our research has several strengths. Firstly, we investigated the effects of self-leadership on a broad set of work role performance dimensions. The constancy of the results found in two independent samples confirms the stability of the effect. Secondly, we chose an experimental design in order to lay a foundation for further research examining the relationship in a field setting (Dipboye, 1990; Highhouse, 2009). Additionally, the heterogeneous sample of participants with work experience also gives reason to believe that our findings apply to a wide range of employees. Finally, we included situational and individual moderators to approach a contingency model of self-leadership (Neck and Houghton, 2006; Yun et al., 2006).

Research and managerial implications

While our results offer differentiated insights on the outcomes of self-leadership, further research could foster our understanding of self-leadership in teams by investigating 1) if and how the three first-order self-leadership strategies interact to influence behaviour 2) whether particular strategies are associated with particular kinds of behaviours and 3) how self-leading team members influence each other. These undoubtedly interesting questions were beyond the scope of this study. Furthermore, in real teams, team members' motives, which may not only be team-oriented, may play an important role in determining which kind of performance team members choose. Still, our results indicate that individual self-leadership offers potential as an employee development tool with the aim of increasing team-oriented behaviour. Managers wishing to improve individual performance as well as unit performance and hoping for adaptive, proactive team members may implement self-leadership training for their team members in the pursuit of this goal.

Conclusions

This study offers a contribution to the self-leadership literature by offering a unique perspective on self-leadership of team members. We employed a comprehensive and differentiated measure of behaviour and an experimental design allowing causal assumptions to demonstrate that self-leadership influences individual performance in teams in more ways than previously investigated. Self-leadership is apparently not only beneficial for individual performance, but fosters team members’ teamwork (proficiency), leads to better adaptation to changes in the team environment (adaptivity) and, especially, encourages participation in the improvement of the team’s procedures (proactivity). Whereas these relationships were not affected by the situational factors (task interdependence and uncertainty) in our study, first insights into moderating individual factors could be gained by showing that self-leadership has a stronger effect on team member proficiency for less collectivistic team members. These promising results highlight the value of further investigating individual self-leadership in teams.

Footnotes

Acknowledgements

This paper was partly made possible through financial support to the first author while she was part of the postgraduate research programme ‘Business Aspects of Loosely Coupled Systems and Electronic Business’ (GRK 517) at the University of Kiel, funded by the German Research Foundation (DFG).

We gratefully thank Margarete Ciuk for her help with data collection.

Notes

Kristina Hauschildt holds a diploma and a doctorate in psychology from Christian-Albrechts-University, Kiel, Germany. She is now working at Higher Education Information System (HIS) in Hanover, Germany, a higher education research institute.

Udo Konradt is full professor of Work, Organizational and Consumer Psychology at the University of Kiel, Germany. He holds a PhD in psychology from the University of Bochum, Germany. The main topics of his research include human–computer interaction, small group research, and leadership in distributed (virtual) teams.

References

Aguinis

Boik

Pierce

(2001) A generalized solution for approximating the power to detect effects of categorical moderator variables using multiple regression. Organizational Research Methods 4(4): 291–323.

Aiman-Smith

Scullen

Barr

(2002) Conducting studies of decision making in organizational contexts: A tutorial for policy-capturing and other regression-based techniques. Organizational Research Methods 5(4): 388–414.

Bandura

(1986) Social Foundations of Thought and Action: A Social Cognitive Theory. New Jersey: Prentice Hall.

Bandura

(1991) Social cognitive theory of self-regulation. Organizational Behavior and Human Decision Processes 50(2): 248–287.

Bateman

Organ

(1983) Job satisfaction and the good soldier: The relationship between affect and employee citizenship. Academy of Management Journal 26(4): 587–595.

Bliese

Klein

Kozlowski

SWJ

(2000) Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. Multilevel Theory, Methods, and Research in Organizations. San Francisco: Jossey-Bass, 349–381.

Bligh

Pearce

Kohles

(2006) The importance of self- and shared leadership in team based knowledge work. Journal of Managerial Psychology 21(4): 296–318.

Borman

Motowidlo

Schmitt

Borman

(1993) Expanding the criterion domain to include elements of contextual performance. Personnel Selection in Organizations. San Francisco: Jossey-Bass, 71–98.

Cable

Graham

(2000) The determinants of job seekers' reputation perceptions. Journal of Organizational Behavior 21(8): 929–947.

10.

Cable

Judge

(1994) Pay preferences and job search decisions: A person-organization fit perspective. Personnel Psychology 47: 317–348.

11.

Campbell

McCloy

Oppler

Sager

Schmitt

Borman

(1993) A theory of performance. Personnel Selection in Organization. San Francisco: Jossey-Bass, 35–70.

12.

Carmeli

Meitar

Weisberg

(2006) Self-leadership skills and innovative behavior at work. International Journal of Manpower 27(1): 75–90.

13.

Carver

Scheier

(1981) Attention and Self-Regulation: A Control-Theory Approach to Human Behavior. New York: Springer.

14.

Carver

Scheier

(1998) On the Self-Regulation of Behavior. New York: Cambridge University Press.

15.

Chen

(2005) Newcomer adaptation in teams: Multilevel antecedents and outcomes. Academy of Management Journal 48(1): 101–116.

16.

Cohen

(1988) Statistical Power for the Behavioral Sciences. Hillsdale, NJ: Erlbaum.

17.

Crant

(2000) Proactive behavior in organizations. Journal of Management 26(3): 435–462.

18.

Deci

(1975) Intrinsic Motivation. New York: Plenum.

19.

Deci

Ryan

(1985) Intrinsic Motivation and Self-Determination in Human Behavior. New York: Plenum.

20.

DeNisi

Klein

Kozlowski

SWJ

(2000) Performance appraisal and control systems: A multilevel approach. Multilevel Theory, Research, and Methods in Organizations. San Francisco: Jossey- Bass, 121–156.

21.

DeShon

Kozlowski

SWJ

Schmidt

Milner

Wiechmann

(2004) A multiple-goal, multilevel model of feedback effects on the regulation of individual and team performance. Journal of Applied Psychology 89(6): 1035–1056.

22.

Dineen

Noe

Wang

(2004) Perceived fairness of web-based applicant screening procedures: Weighing the rules of justice and the role of individual differences. Human Resource Management 43(2–3): 127–145.

23.

Dipboye

(1990) Laboratory vs. field research in industrial and organizational psychology. International Review of Industrial and Organizational Psychology 5: 1–34.

24.

Feltz

Landers

(1983) The effects of mental practice on motor skill learning and performance: A meta-analysis. Journal of Sport Psychology 5(1): 25–57.

25.

Frese

(2008) The word is out: We need an active performance concept for modern workplaces. Industrial and Organizational Psychology 1(1): 167–169.

26.

Graham

Cable

(2001) Consideration of the incomplete block design for policy-capturing research. Organizational Research Methods 4(1): 26–45.

27.

Greenhaus

Powell

(2003) When work and family collide: Deciding between competing role demands. Organizational Behavior and Human Decision Processes 90(2): 291–303.

28.

Griffin

Neal

Parker

(2007) A new model of work role performance: Positive behavior in uncertain and interdependent contexts. Academy of Management Journal 50(2): 327–347.

29.

Harold

Ployhart

(2008) What do applicants want? Examining changes in attribute judgments over time. Journal of Occupational and Organizational Psychology 81(2): 191–218.

30.

Highhouse

(2009) Designing experiments that generalize. Organizational Research Methods 12(3): 554–566.

31.

Hofmann

(1997) An overview of the logic and rationale of hierarchical linear models. Journal of Management 23(6): 723–744.

32.

Hofmann

Gavin

(1998) Centering decisions in hierarchical linear models: Implications for research in organizations. Journal of Management 24(6): 623–641.

33.

Houghton

Neck

(2002) The revised self-leadership questionnaire. Journal of Managerial Psychology 17(8): 672–691.

34.

Hox

(2002) Multilevel Analysis: Techniques and Applications. Mahwah, NJ: Erlbaum.

35.

Ilgen

Pulakos

(1999) The Changing Nature of Performance. San Francisco: Jossey-Bass.

36.

Jackson

Colquitt

Wesson

Zapata-Phelan

(2006) Psychological collectivism: A measurement validation and linkage to group member performance. Journal of Applied Psychology 91(4): 884–899.

37.

Judge

Bretz

(1992) Effects of work values on job choice decisions. Journal of Applied Psychology 77(3): 261–271.

38.

Karren

Barringer

(2002) A review and analysis of the policy-capturing methodology in organizational research: Guidelines for research and practice. Organizational Research Methods 5(4): 337–361.

39.

Katz

Kahn

(1978) The Social Psychology of Organizations. New York: Wiley.

40.

Knudstrup

Segrest

Hurley

(2003) The use of mental imagery in the simulated employment interview situation. Journal of Managerial Psychology 18(6): 573–591.

41.

Konradt

Andressen

Ellwart

(2009) Self-leadership in organizational teams: A multilevel analysis of moderators and mediators. European Journal of Work and Organizational Psychology 18(3): 322–346.

42.

Kozlowski

SWJ

Bell

Borman

Ilgen

Klimoski

(2003) Work groups and teams in organizations. Handbook of Psychology: Industrial and Organizational Psychology 12. London: Wiley, 333–375.

43.

Kozlowski

SWJ

Ilgen

(2006) Enhancing the effectiveness of work groups and teams. Psychological Science in the Public Interest 7(39): 77–124.

44.

Kozlowski

SWJ

Klein

Kozlowski

SWJ

(2000) A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent processes. Multilevel Theory, Methods, and Research in Organizations. San Francisco: Jossey-Bass, 3–90.

45.

Kristof-Brown

Jansen

Colbert

(2002) A policy-capturing study of the simultaneous effects of fit with jobs, groups, and organizations. Journal of Applied Psychology 87(5): 985–993.

46.

Langfred

(2000) Work-group design and autonomy: A field study of the interaction between task interdependence and group autonomy. Small Group Research 31(1): 54–70.

47.

LePine

Piccolo

Jackson

Mathieu

Saul

(2008) A meta-analysis of teamwork processes: Tests of a multidimensional model and relationships with team effectiveness criteria. Personnel Psychology 61(2): 273–307.

48.

Manz

(1986) Self-leadership: Toward an expanded theory of self-influence processes in organizations. Academy of Management Review 11(3): 585–600.

49.

Manz

Neck

(2004) Mastering Self-Leadership: Empowering Yourself for Personal Excellence, 3rd edn. Upper Saddle River, NJ: Pearson Prentice Hall.

50.

Manz

Sims

Jr (1980) Self-management as a substitute for leadership: A social learning theory perspective. Academy of Management Review 5(3): 361–367.

51.

Marks

Mathieu

Zaccaro

(2001) A temporally based framework and taxonomy of team processes. Academy of Management Review 26(3): 356–376.

52.

Morrison

Phelps

(1999) Taking charge at work: Extrarole efforts to initiate workplace change. Academy of Management Journal 42(4): 403–419.

53.

Neck

(1996) Thought self-leadership: A self-regulatory approach towards overcoming resistance to change. The International Journal of Organizational Analysis 4(2): 202–216.

54.

Neck

Ashcraft

Vansandt

(1998) Employee self-leadership: Enhancing the effectiveness of nonprofits. International Journal of Organization Theory and Behavior 1(4): 521–552.

55.

Neck

Houghton

(2006) Two decades of self-leadership theory and research. Journal of Managerial Psychology 21(4): 270–295.

56.

Neck

Manz

(1992) Thought self-leadership: The influence of self-talk and mental imagery on performance. Journal of Organizational Behavior 13(7): 681–699.

57.

Neck

Manz

(1996) Thought self-leadership: The impact of mental strategies training on employee cognition, behavior, and affect. Journal of Organizational Behavior 17(5): 445–467.

58.

Neck

Smith

Godwin

(1997) Thought self-leadership: A self-regulatory approach to diversity management. Journal of Managerial Psychology 12(3): 190–203.

59.

Politis

(2006) Self-leadership behavioural-focused strategies and team performance: The mediating role of job satisfaction. Leadership and Organizational Development Journal 27(3): 203–216.

60.

Powell

Piccoli

Ives

(2004) Virtual teams: A review of current literature and directions for future research. The DATA BASE for Advances in Information Systems 35(1): 6–36.

61.

Prussia

Anderson

Manz

(1998) Self-leadership and performance outcomes: The mediating influence of self-efficacy. Journal of Organizational Behavior 19(5): 523–538.

62.

Pulakos

Arad

Donovan

Plamondon

(2000) Adaptability in the workplace: Development of a taxonomy of adaptive performance. Journal of Applied Psychology 85(4): 612–624.

63.

Raudenbush

Bryk

Congdon

(2004) HLM (Version 6)[Computer software]. Lincolnwood, IL: Scientific Software International.

64.

Rogelberg

Ployhart

Balzer

Yonker

(1999) Using policy capturing to examine tipping decisions. Journal of Applied Social Psychology 29(12): 2567–2590.

65.

Rotundo

Sackett

(2002) The relative importance of task, citizenship, and counterproductive performance to global ratings of job performance: A policy capturing approach. Journal of Applied Psychology 87(1): 66–80.

66.

Rousseau

Aubé

Savoie

(2006) Teamwork behaviors: A review and an integration of frameworks. Small Group Research 37(5): 540–570.

67.

Salas

Burke

Cannon-Bowers

(2000) Teamwork: Emerging principles. International Journal of Management Reviews 2(4): 339–356.

68.

Shea

Guzzo

Rowland

Ferris

(1987) Groups as human resources. Research in Personnel and Human Resources Management 5. Greenwich, CT: JAI Press, 323–356.

69.

Stewart

Carson

Cardy

(1996) The joint effects of conscientiousness and self-leadership training on employee self-directed behavior in a service setting. Personnel Psychology 49(1): 143–164.

70.

Tabachnick

Fidell

(2001) Using Multivariate Statistics, 4th edn. Needham Heights, MA: Allyn and Bacon.

71.

Taggar

Brown

(2001) Problem-solving team behaviors: Development and validation of BOS and a hierarchical factor structure. Small Group Research 32(6): 698–726.

72.

Uhl-Bien

Graen

(1998) Individual self-management: Analysis of professionals' self-managing activities in functional and cross-functional work teams. Academy of Management Journal 41(3): 340–350.

73.

Wagner

(1995) Studies of individualism-collectivism: Effects on cooperation in groups. Academy of Management Journal 38(1): 152–172.

74.

Wagner

Moch

(1986) Individualism-collectivism: Concept and measure. Group & Organization Management 11(3): 280–304.

75.

Wall

Cordery

Clegg

(2002) Empowerment, performance, and operational uncertainty: A theoretical integration. Applied Psychology 51(1): 146–169.

76.

Yun

Cox

Sims

(2006) The forgotten follower: A contingency model of leadership and follower self-leadership. Journal of Managerial Psychology 21(4): 374–388.