Record numbers of women are entering science, technology, engineering, and mathematics (STEM) subjects and careers. Nevertheless, they remain underrepresented in STEM in most countries. The imbalance is greatest in engineering and computer science (Statistisches Bundesamt, 2015). When females demonstrate similar or greater talent levels or even better grades than males, they are nevertheless more likely to not choose a STEM major or career (Wang, Eccless, & Kenny, 2013). The underrepresentation of females is problematic for individuals and societies. STEM fields facilitate lucrative, high-status careers with considerable innovation potential (Kalwarski, Mosher, Paskin, & Rosato, 2007). At the societal level, an increase in the rates of females in STEM can help reduce the skilled-labor shortage in STEM and increase the overall number of talented persons working in STEM. As girls' interest in STEM decreases over the course of schooling (Frenzel, Goetz, Pekrun, & Watt, 2010), programs designed to get more females interested in STEM need to be offered to girls early. In the case of gifted education in particular, it is important that in-school and extracurricular gifted education programs be combined (Olszewski-Kubilius, 2009). One possibility for effective extracurricular education for girls in STEM is mentoring. It sustains talented girls' interest in STEM (Pfund, Pribbenow, Branchaw, Lauffer, & Handelsman, 2006), and it is a viable option within gifted education (Subotnik, Edmiston, Cook, & Ross, 2010). As will be explained later, online mentoring is particularly useful for getting talented female secondary school students excited about STEM. Despite this potential, however, little is known about whether one-on-one mentoring or group mentoring is more effective. It is conceivable that oneon-one mentoring is particularly effective on account of the very individual sort of guidance it provides. At the same time, however, it is possible that group mentoring using a 702215G CQXXX10.