Authentic Leadership’s Impact on Follower Psychological Capital and Performance Through Organizational Identification and Role Clarity

ABSTRACTScholars have criticized positive leadership styles, such as authentic leadership, as being limited to influencing follower performance through relations-oriented behaviors without necessarily providing more task-oriented direction. Applying this behavioral leadership theory dichotomy, we extend authentic leadership theory and research by proposing and testing how authentic leadership influences followers' psychological capital (PsyCap) and subsequent performance through both relations- (organizational identification) and task-oriented (role clarity) pathways. The results of a three-wave field study, multiple experiments, and a time-lagged, multi-source field study support that authentic leadership influences follower psychological resources and performance through both organizational identification and role clarity. Moreover, our results hold when controlling for other leadership constructs (ethical and transformational leadership) and other potential mediators that have been the focus of prior authentic leadership research (leader trustworthiness, leader identification, and LMX). We discuss the implications of our findings for expanding work on authentic leadership by examining a more task-oriented focus in future research, including how this research sheds light on several recent critiques of authentic leadership theory. Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1 Scholars have used different terms or labels for these two perspectives including concern for people and production (Blake & Mouton, Citation1964), consideration and initiating structure (Stogdill, Citation1963), and employee versus production orientation (Likert, Citation1961).2 Due to 329 leaders having one follower, we also ran the analysis as a single-level analysis, and results are consistent with the results of the multi-level analysis.3 The 23 observed variables at Level 1 (6 items measuring role clarity, 6 items measuring organizational identification, 4 parcels measuring PsyCap, and 7 items measuring LMX) provided 276 observations as a result of the variances and covariances among the 23 observed variables (calculated as (p*(p + 1))/2) and 23 means. The 8 observed variables at Level 2 (4 parcels measuring authentic leadership and 4 parcels measuring transformational leadership) provided 36 observations as a result of variance and covariances among the 8 observed variables and 8 means. This totals 343. We loaded these items on each of their respective latent factors, so six latent factors were modeled. One factor loading in each of the six latent factors was fixed at 1. Thus, the number of parameters we estimated in this 6-factor CFA model was 25 factor loadings (31 observed variables minus the 6 fixed factor loadings) + 31 variances (because we have 31 observed variables) + 6 variances for each of the latent variables + 7 covariances (each pair of the 4 latent variables at Level 1 and the pair of the 2 latent variable at level 2) + 31 intercepts = 100 parameters. Thus, our degrees of freedom are 343–100 = 243, which are the degrees of freedom for the 6-factor model.Additional informationFundingThe work was supported by the Wake Forest Faculty Development Fund.