School-based mentoring programs (SBMP) have become widely used due to their presumptuously simple implementation and effective results; however, they remain to be scientifically under-developed and require integrity during implementation (Miller et al., 2013; Smith & Stormont, 2011). Undoubtedly, SBMP have been shown to improve students: attitudes, confidence, school engagement, school performance, behaviour, vocational skills, and emotional and social skills (Converse & Lignugaris/Kraft, 2009; Miller et al., 2013; Schwartz, Rhodes, & Herrera, 2012). Contrarily, these promising outcomes are equally met with many studies that show: no, or only little, improvements; only short-lived results; or only effectiveness with certain types of students (Converse & Lignugaris/Kraft, 2009; McQuillin, Smith, & Strait, 2011; Miller et al., 2013).
A major understanding that many program designers seemingly fail to comprehend is that SBMP are not ‘general-purpose’ programs. Thus, many SBMP are not designed and implemented strategically. Particularly problematic for these poorly designed programs are: the lack of defined success outcomes, measurements of success being open to biases, the ambiguous and faulty evaluations of such programs, and the individual program’s diversity of standards, budgets, objectives, and implementation (Converse & Lignugaris/Kraft, 2009; Gusic et al., 2010; Schwartz, Rhodes, & Herrera, 2012; Smith & Stormont, 2011). Evidently, the difficulty of replicating the