Chapter 6 concentrates on the outcomes computed by the hypothesis tests. Results are outlined for each hypothesis consequently. Supporting a comprehensive structure, hypotheses are clustered in two blocks. The first block concerns direct effects of perceived leadership behavior on follower self-leadership and performance, whereas the second block concerns moderation and mediation effects of distance dimensions.
For testing hypotheses, differentiated statistical procedures were applied. Hypotheses 1.1 to 1.3 assess direct effects of perceived leadership behavior on followers’ self-leadership and performance. Results are calculated with multiple linear regression modelling. This statistical procedure has been applied successfully in prior studies investigating the field of leadership (Davis & Bryant, 2010; Luo & Cheng, 2014). Hypotheses of the second sequence (2.1, 2.2, 2.5) are tested using moderation analysis. Hypothesis 2.3 is again tested with linear regression modeling and finally, hypothesis 2.4 is investigated using mediation procedures according to Baron and Kenny (1986), Hayes (2009), and Sobel (1982). Bootstrapping was performed throughout the entire analysis as it assumes the sample to be representative for the population. Bootstrapping can be seen as a robust method to test for significant relations that are relatively vigorous to violations of assumptions as it estimates the characteristics of the sampling distribution from the actual sample (Field, 2013, pp. 198ff, 871).