Doctor of Philosophy (PhD)


Human Resource Education and Workforce Development

Document Type



Many organizations are committing to education and training that deepens skills, perspectives, and competencies of their leaders. This research located 103 studies from 1982-2001 with a full range of managerial leadership development interventions including feedback, developmental relationships, on-the-job experiences, and formal training. It integrated results of 83 of these studies with formal training interventions via meta-analytic techniques to determine the effectiveness of interventions, in their enhancement of performance, knowledge, and expertise at the individual, team or group, or organizational level. The studies were viewed through a "macro-lens," that used a full range of managerial leadership development interventions (McCauley, Moxley, & Van Velsor, 1998), a high-performance leadership competency model (Holton & Naquin, 2000), and the Results Assessment System (Swanson & Holton, 1999). The studies were separated into four separate data sets by the research design used in individual studies (posttest only with control group, pretest-posttest with control group, single group pretest-posttest, and correlation) with the unit of analysis being the outcome measure of the study. Effect sizes, derived by using Carlson and Schmidt's (1999) formulas, were adjusted for artifacts sampling error and error of measurement. in this meta-analysis study Hunter and Schmidt's (1990) method of partitioning observed effect size variability into portions attributable to subject-level sampling error and between-study differences was used to determine the presence of seven possible moderating variables. This research found formal training programs with knowledge outcomes highly effective. The average effect size for knowledge outcomes ranged from .96 (control group, knowledge-objective) to 1.37 (pretest-posttest, knowledge-objective). The average effect size for expertise outcomes ranged from .30 (control group, expertise-subjective) to 1.01 (pretest-posttest, expertise-objective). System outcomes had an average effect size of .39 (control group). Two primary methodological issues were raised regarding Burke and Day's (1986) meta-analysis on the effectiveness of managerial training: 1) independence of outcomes measured (effect sizes), and 2) over weighting of studies with multiple effect sizes. Implications were provided for future research opportunities and for practical use of the findings. This meta-analysis synthesized existing studies from a broad range of settings, researchers and circumstances and integrated conflicting findings to establish a general knowledge base about managerial leadership development.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Elwood F. Holton, III