Principal leadership has been widely regarded as a powerful catalyst for school improvement and student learning. This article presents a multivariate meta-analysis of 42 empirical studies, published between 2000 and 2020, that examined the effects of principal leadership on student achievement in the United States. The focus is on the conceptual models and methodological approaches used to examine these effects. Our study reveals three major findings: (a) principal leadership had a positive impact on student achievement, when the effects were conceptualized as direct effects without controls (0.25 SD), or as indirect effects (0.22 SD); (b) the observed effects of principal leadership are contingent upon how impacts were conceptualized and analyzed; and (c) substantial heterogeneity existed in observed principal leadership effects, with student outcome measure serving as a significant moderator. We offer directions for future studies, including conducting more replication studies using high-quality approaches across different contexts; developing common typologies, sets of variables and their definitions and measurements; identifying significant mediators in the school process; applying multiple conceptual models and methodological approaches on a more comprehensive dataset; and moving beyond the effectiveness perspective to adopting an equity lens.