Paternalistic leadership styles and follower performance : examining mediating variables in a multi-level model

Abstract

This study developed a conceptual-based paternalistic leadership model (Farh & Cheng, 2000) to examine how the three components of paternalistic leadership behaviors, authoritarian, benevolent, and moral are transmitted to followers' performance (in-role and extra-role performance) by intermediate variables like dependence, perceived supervisory support (PSS), and self-efficacy. Aggregation is a common procedure used in recent leadership research to determine whether leadership behaviors should be interpreted as group-level or individual-level variables for multi-level analysis (Bono & Judge, 2003; Dvir & Sharmir, 2003; Kark, Shamir & Chen, 2003). Using a sample of 178 scout leader-member dyads collected in a voluntary organization in Hong Kong, and a sample of 556 leader-follower dyads from a manufacturing firm in China, HLM analysis indicated that authoritarian leadership was negatively related to followers' performance. Benevolent leadership and moral leadership were positively related to followers' performance. Consistent with previous research studies (e.g. Farh & Cheng, 2000), the direct effects of the three components of paternalistic leadership behaviors on followers' performance were predicted. In the voluntary organization, , perceived supervisory support primarily mediated the relationship between benevolent leadership and followers' performance, while self-efficacy primarily mediated the relationship between moral leadership and followers' performance. In the manufacturing firm, results indicated that dependence primarily mediated the relationship between authoritarian leadership and followers' performance, and perceived supervisory support mediated the relationship between benevolent leadership and followers' performance. Implications for the theory and practice of leadership are also discussed.