The influence of leader humor on relationships between leader behavior and follower outcomes

Journal of Managerial Issues, Summer, 2009 by Robert P. Vecchio, Joseph E. Justin, Craig L. Pearce

Procedure and Measures

Questionnaires were mailed directly to 491 principals. The questionnaire asked each principal to provide confidential ratings of the performance of the department lead teachers as part of a study of leadership and organization effectiveness within the state education system. The response rate for principals was 45.4 percent (223 questionnaires). After each principal responded to the survey, a separate confidential, personally-addressed letter was sent directly to the respective department lead teachers. A total of 1,060 confidential questionnaires were mailed to department lead teachers after the principal had confidentially responded. This questionnaire asked for ratings of the principal on dimensions of leadership behavior, as well as job-relevant perceptions of the teachers. A total of 342 teachers responded across all institutions. As the intended level of analysis is dyadic (i.e., focusing on the unique dyadic social exchange that existed between each principal and a respective lead teacher), and in order to ensure an equal representation of responses, 179 teachers were randomly selected for 179 principals for whom complete data were available (i.e., one per principal) via a sampling procedure based on a table of random numbers (Lindquist, 1940). One principal-teacher pair was also selected from each school to avoid potential problems associated with non-independence of descriptions of a common principal. This sampling approach was utilized as "nested" data structures (where multiple respondents are embedded in a common organizational unit) can generate parameter estimates that are incorrect (Bliese, 2000; Bliese and Hanges, 2004; Hofmann, 1997). A hierarchical linear modeling (HLM) approach was not feasible in the present instance as the target group (lead teachers) had too few cases per unit (a maximum of five) to permit the use of HLM (Raudenbush and Bryk, 2002).

The survey completed by each principal included a three-item "performance rating" measure of the lead teacher that was adapted from a scale by MacKenzie, Podsakoff, and Fetter (1991). The confidential survey completed by the teachers incorporated established scales for the constructs where possible. Specifically, their survey included: (1) a three-item measure of the principal's "use of contingent personal reward," developed by Podsakoff, Todor, Grover, and Huber (1984), (2) a four-item measure of the principal's "integrity," based on the work of Gabarro (1978), Mayer, Davis, and Schoorman (1995), Mayer and Davis (1999), and Luthans and Avolio (2003), (3) a four-item measure of the principal's "use of humor," developed by Avolio et al. (1999), (1) and (4) a three-item measure of employee degree of 'Job satisfaction," developed by Cammann, Fichman, Jenkins, and Klesh (1983). (The specific scale items are available from the authors on request.) All responses were obtained on five-point response scales (1 = strongly disagree, 5 = strongly agree).

In addition, the teachers and principals provided information on a variety of demographic variables. For the teachers, the average age and years of teaching were 48.17 (SD = 9.56) and 20.95 years (SD = 10.54), respectively, with 51.4% of the respondents being male. For the principals, the average age and years of teaching were 52.18 (SD = 6.84) and 26.35 years (SD = 7.65), respectively, with 67.8% of the principals being male. The principals had held their current administrative positions for an average of 4.29 years (SD = 3.71). Because subordinate job satisfaction has been found to be associated with supervisor use of humor (Decker, 1987), teacher/follower job satisfaction, along with years of teaching, were used as control variables in the subsequent analyses.