هوش عاطفی به عنوان اساس ظهور رهبری در تیم های خود مدیریتی

We present and test a theory on leader emergence in self-managing teams that highlights the emotional and cognitive skills underlying selection as an informal team leader. Existing theory and research reveals that informal leaders are selected because they display constructive task and team management behavior. We contribute to existing theory in two ways. First, by proposing that specific cognitive processes and skills precede the appropriate enactment of those behaviors by facilitating an accurate analysis of the task situation. Second, by proposing that empathy, an aspect of emotional intelligence, precedes and enables those cognitive processes and skills by providing an accurate understanding of team and member emotions and needs. We test our theory in a longitudinal study of 382 team members comprising 48 self-managing teams. Our theory is partially supported and implications are discussed.

Despite the growing use of empowered teams such as self-managing work teams (SMWTs) (Lawler, 1998), little research or theory has appeared in the past two decades to help answer persistent questions about their leadership needs Cohen et al., 1997 and Druskat & Wheeler, in press. One subject that has received some research attention is leadership emergence in SMWTs, and in particular, assessment of the behaviors that predict one's emergence as an informal team leader Smith & Foti, 1998 and Taggar et al., 1999. Emergent team leaders hold no formal authority and team willingness to follow their lead can end at any time. Thus, theory and research attempting to isolate why certain individuals are selected for the role lends insight into perceived leadership needs in SMWTs. Existing research reveals that the behaviors predicting one's emergence as an informal leader can be organized into two categories—task- and member-focused behavior—and that task-focused behaviors (e.g., task coordination) are the strongest predictors (see Bales, 1950, Lord, 1977 and Taggar et al., 1999). To date, however, little research or theory has sought to identify the knowledge, skills, and abilities (KSAs) that underlie and enable the behaviors that predict emergence. In this article, we introduce and test a model of the KSAs that predict leadership emergence in SMWTs. Our model underscores the relevance of emotional and cognitive skills to the exhibition of the behaviors that predict leader emergence. It combines three propositions drawn from previous leadership theory that serve as the basis of the model: (1) that emergent leaders are socially perceptive and skilled at recognizing and understanding the feelings and emotions in their team (see Chowdhry & Newcomb, 1952 and Steiner, 1972), (2) that this understanding augments a leader's cognitive analysis and prioritization of issues facing the team (see Salovey et al., 2000 and Zajonc, 1998), and (3) that the high quality cognitive analysis underlies and leads to the skilled behavior (see Bandura, 1982) that predicts leader emergence. We combine these propositions to introduce and test a model suggesting the specific KSAs necessary for attaining informal leadership status in a self-managing team. For the purposes of our study, informal leadership status was assessed by the percentage of teammate votes each member received for “informal leader.” We begin with our literature review and the presentation of our study hypotheses, which are presented from back to front. That is, we start by describing the skilled behaviors we propose to be most directly related to emergence as an informal team leader. We then present the cognitive skills that support those behaviors. Lastly, we present the emotional skill that we propose to be the basis of leadership emergence.

Table 2 shows correlations among the variables in our model and descriptive statistics. The hypotheses were tested using AMOS4 structural equation modeling. Fig. 1 shows the model tested and the resulting standardized path coefficients. To test the model for fit, a number of indicators were examined beginning with the χ2 statistic and degrees of freedom (χ2=37.3, df=8, p=.00). A good-fitting model has a χ2 value approaching the degrees of freedom and a nonsignificant p value (Jöreskog & Sörbom, 1993), thus this indicator does not indicate a fit. However, for large sample sizes, the χ2 statistic is overly sensitive (Tabachnick & Fidell, 2001). Also, the χ2 measure assumes that the model fits perfectly in the population; it does not take parsimony into account (Jöreskog & Sörbom, 1993).For these reasons, Tabachnick and Fidell (2001) and Jöreskog and Sörbom (1993) suggest that additional indications of model fit be examined including the root mean square error of approximation (RMSEA), the normed fit index (NFI), the relative fit index (RFI), and the comparative fit index (CFI). The RMSEA does not assume perfect fit in the population and examines the error per degree of freedom. Using RMSEA, a poor-fitting model occurs when the RMSEA is greater than .10 (Tabachnick & Fidell, 2001). By this measure, our model is marginally acceptable (RMSEA=.10). The NFI, RFI, and CFI are indices that, using different criteria, measure how much better the model fits than “no model at all.” The closer these indices are to 1.0, the better the fit of the model. When these indices are greater than .90, the model is considered acceptable (Arbuckle & Wothke, 1999). These indices provide inconsistent indications about the fit of our model. The NFI and CFI indicate a good fit (NFI=.96, CFI=.96), while the RFI is below, but close to, the cutoff (RFI=.88). As a whole, the fit statistics again suggest our model to be marginally acceptable. Four of our seven hypotheses are supported at p<.05: Hypothesis 3 (group task coordination is positively related to percentage votes for leader), Hypothesis 5 (perspective taking is positively related to supporting/developing others), Hypothesis 6 (empathy is positively related to pattern recognition), and Hypothesis 7 (empathy is positively related to perspective taking). Two of our hypotheses are partially supported at p<.10: Hypothesis 1 (supporting/developing others is positively related to group task coordination) and Hypothesis 2 (supporting/developing others is positively related to percentage votes for leader). The only hypothesis for which there was no support is Hypothesis 4 (pattern recognition is positively related to group task coordination). Given the marginal fit of the model, a series of modifications using a model-generating approach (Jöreskog & Sörbom, 1993) was made to the model. The model-generating approach uses modification indices to suggest changes to the model; it also requires that any change be theoretically grounded. The revised model is exploratory and requires further testing with an additional sample. Nevertheless, we present the revised model here because (1) it adds confidence to the findings that are consistent across both models and (2) it provides information that can guide future theory and research. The revised model (Fig. 2) differs in four ways from the hypothesized model. First, the direct path from supporting/developing others to percentage votes is deleted and replaced by a direct path from perspective taking to percentage votes. This is consistent with research discussed earlier showing that task-related behavior is more likely to be related to selection as an informal leader than relationship-oriented behavior (Lord, 1977). As we have defined our constructs, supporting/developing others is more relationship-oriented than perspective taking. Supporting/developing others is focused on working with and improving a member's performance whereas perspective taking is the ability to see and use another person's perspective to achieve better task results. Additionally, the high degree of diversity in the groups may have made the ability to value differing perspectives a particularly appreciated skill.Also, a path from pattern recognition to supporting/developing others replaces the path from pattern recognition to group task coordination. Although we originally argued that pattern recognition would be needed to analyze the group's performance and recognize group needs, upon reflection, our measure of group task coordination may have tapped more strongly into one's ability to engage members in the task by creating team spirit and collaboration than to actually managing task processes. In addition, supporting/developing others requires cognitive skill at recognizing and labeling the behavioral patterns exhibited by team members. Thus, it seems reasonable that, as our constructs are defined, pattern recognition would be more directly related to supporting/developing others than to group task coordination. The fit statistics for this model indicate that the model is a good fit for the data (χ2=10.8, df=8, p=.21, RMSEA=.03, NFI=.99, RFI=.97, CFI=.997). In this model, as in our hypothesized model, Hypothesis 3, Hypothesis 5, Hypothesis 6 and Hypothesis 7 are supported, thus, we gain additional confidence in the support for those findings. Additionally, Hypothesis 1 (supporting/developing others being directly related to group task coordination) is now supported at p<.05, thus the revised model gives greater support to this hypothesis. The marginal relationship originally found for Hypothesis 2 (supporting/developing others being directly related to percentage votes) was not supported in the revised model, thus, we do not have confirmation for this hypothesis. Hypothesis 4 was not supported in either model, thus, we cannot say pattern recognition is directly related to group task coordination, however, the revised model suggests that the relationship between pattern recognition and group task coordination is mediated by the ability to support/develop others. Overall, our study findings support our fundamental premise that empathy enables the cognitive skills of pattern recognition and perspective taking, which form the foundation for the leadership behaviors used by emergent leaders. However, Hypothesis 2 was not supported, thus, the ability to develop others does not appear to lead directly to being selected as informal team leader, rather, it does so indirectly through its role in supporting group task coordination.