This dissertation examines the levels of and relationships between technical leadership, adaptive leadership, and defined autonomy among Denver school leaders along with their combined effects on school growth gains over time. Thirty principals provided complete responses to an online survey that included existing scales for technical leadership, adaptive leadership, and autonomy and newly created measures for defined autonomy. Five different scenarios for measuring defined autonomy were used combining managed instruction with different forms of school autonomy: autonomy over general operations (Scenario 1), autonomy over people, time, and money-combined (Scenario 2), autonomy over people (Scenario 3), autonomy over time (Scenario 4), and autonomy over money (Scenario 5). Technical challenges associated with the data (e.g., small sample size, self-reported leadership scales, accurately measuring school autonomy) severely limited the statistical power of the analyses and the reliability of the results. A series of hierarchical regression analyses found no statistically significant effects for any of the variables however, one overall model (when total leadership and defined autonomy scenario 3 were included together with three controls; Baseline score, enrollment, and ELL) did yield a statistically significant result, R[superscript 2] = 0.398, p < 0.05. No beta values were statistically significant; however, of the three primary variables in this model, total leadership explained the most variance (ß = 0.263), followed by managed instruction (ß = 0.249). Autonomy over people had a surprisingly negative association with school growth gains (ß = -0.220). Persistent union constraints, a lack of training in human capital management, poor or delayed implementation, and limited measures of teacher effectiveness available at the time, along with the data limitations may be to blame. More research is needed to examine these possibilities. Other patterns in the data revealed that managed instruction consistently presented the strongest effects of any other autonomy variable, autonomy over operational areas produced stronger results than autonomy over curriculum, and that autonomy over resources may matter more when managed instruction is not present. Finally, analysis of covariance (ANCOVA) was used to test the research hypothesis which was rejected. Although not the preferred results, this study affirms that school leadership matters and offers some new insights on and possible measures for the concept of defined autonomy in urban school districts. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]