The roles of age, educational level, work experience, gender and needs for connection in predicting workers' ideal preferences for relational and job-centered leadership behaviors within the workplace were investigated. Most importantly, the role of workers' needs for connections as a mediating variable between gender and their ideal preferences for worker-centered and job centered leadership behaviors was examined. Measures of ideal leadership behavior preferences and needs for connection along with demographic information were completed by 1009 participants from three mid-western organizations. Controlling for organizational variables, regression analyses revealed that age and educational level were negatively correlated with workers' ideal preferences for worker-centered leadership behaviors; age, educational level and work experience were positively correlated with ideal preferences for job-centered leadership behaviors. Hierarchical regression analysis indicated that after the effect of age and educational level had been controlled for, employees' gender accounted for a significant portion of the variance in explaining workers' ideal preferences for worker-centered leadership behaviors and mediated the relationship between gender and ideal preferences for work-centered relational leadership behaviors. Post hoc analyses revealed no significant gender differences in ideal preferences for job-centered leadership behaviors. The findings are discussed in light of self-in-relation theory and with regard to how leaders may tailor their leadership styles to more effectively meet workers' needs and preferences.
In the new work dynamic, "job-centered" leadership is being replaced by "workercentered" leadership which has the potential to drastically alter the role of an effective leader. However, insufficient empirical attention has been devoted to determining if workers prefer "worker-centered" leadership behaviors over the more traditional leadership style. One reason for this research gap may be that most empirical studies seeking to illuminate effective leadership strategies conceptualize leadership as predominantly a function of leader qualities (Eagly, Johnson, 1990; Eagly, Karau, & Makhigani, 1995). House & Aditya (1997) viewing the leader as the "major actor in leadership ... the center of action, influence, and power" (Hollander, 1992, p. 43). Consequently, the role of the worker as a fundamental factor in co-constructing or "legitimizing" effective leadership is often neglected (Hollander, 1993). Thus, substantive questions about workers' preferences for particular types of leadership behaviors remain unanswered.
The consistent tendency for researchers to neglect workers' preferences for leadership behaviors is surprising, considering that as early as 1950 Sanford suggested that "There is some justification for regarding the follower as the most crucial factor in any leadership event" (p. 4). Similarly, Hollander (1993) has consistently argued that leadership is not a quality possessed by a leader but "a process involving followership. Without followers, there plainly are no leaders or leadership ... Followers affect the strength of a leader's influence, the style of a leader's behavior, and the performance of the group, through processes of perception, attribution, and judgment" (p. 29). Moreover, several studies have provided evidence that the leader-worker relationship through which many leadership functions are carried out is critical to workers' productivity (Fiedler & Chemers, 1974; Neil & Kirby, 1985; Wilkinson & Wagner, 1993), rewards (Coates, Jarratt, & Mahaffie, 1990), morale (Meade, 1985), and work satisfaction (Fiedler, 1967; Hunt & Liebscher, 1973; Singer, 1985; Singer & Singer, 1990; Wilkinson & Wagner, 1993).
When asked, workers report distinct preferences for the type of ideal behaviors they would like their ideal leader to use within the workplace (e. …