Do sports develop leadership? The impact of college varsity sports on leadership capacity and self-efficacy.

Introduction: Effective leadership in sports is associated with better team functioning, physical and psychosocial wellbeing, and reduced incidence of severe injuries. In spite of widespread anecdotal support, few studies have empirically evaluated leadership development in athletes. The purpose of this study was to investigate socially responsible leadership capacity and self-efficacy in college varsity athletes and determine the impact of competitive sports participation on leadership capacity and self-efficacy. We hypothesized that college varsity athletes would (1) exhibit greater leadership capacity and self-efficacy than nonathletes and (2) demonstrate greater growth in leadership capacity and self-efficacy than nonathletes. Methods: The Multi-Institutional Study of Leadership survey was distributed to a representative sample of United States college students in 2018. The leadership outcomes of college varsity (intercollegiate) athletes and nonathletes were compared using the Socially Responsible Leadership Scale and Leader Self-Efficacy Scale. The Socially Responsible Leadership Scale has an omnibus score and six subscales: self-awareness, personal responsibility, integrity, collaboration, open-mindedness, and civic engagement. Student's t-tests were used to compare scores on the Socially Responsible Leadership Scale and Leader Self-Efficacy Scale between athletes and nonathletes. Hierarchical multiple linear regression analyses were conducted for each leadership outcome, both with and without leadership high-impact practices. Results: The study contained 50,653 participants, including 8,447 college varsity athletes (age 20.3 ± 1.5 years, 56.2% women) and 42,206 nonathletes (age 20.2 ± 1.5 years, 65.9% women). While athletes reported statistically higher levels of self-awareness and leader self-efficacy but lower integrity and open-mindedness compared with their nonathlete peers, all comparisons had trivial effect sizes (all Cohen's d < 0.02; range: -0.067 to 0.159). There were no differences between athletes and nonathletes in terms of personal responsibility, collaboration, civic engagement, and the omnibus score of the Socially Responsible Leadership Scale. In regression models with and without leadership high-impact practices, athlete status explained only 0.1% or less of variance in each leadership outcome. Discussion: The findings from this large sample of college athletes and nonathletes (