The current study set out to explore which constructs were most important in predicting global self-esteem in undergraduate and postgraduate students. A number of studies have highlighted the negative consequences of low self-esteem, including depression (Sharma & Agarwala, 2013), substance abuse (Bachman, O’Malley, Freedman-Doan, Trzesniewski, & Donnellan, 2011), and antisocial behaviours (Bandura, 1982). It is therefore imperative to determine factors that are associated with self-esteem in order to develop successful interventions. Past research has highlighted a number of possible constructs and thus, the following predictor variables were chosen: peer communication, family relatedness, university satisfaction, appearance appraisal, age, and relationship satisfaction (Aryana, 2010; Cabeldue & Boswell, 2012; Kutob, Senf, Crago, & Shisslak, 2010; Ojanen & Perry, 2007; Trzesniewski, Donnellan, & Robins, 2003). Multiple regression analysis revealed university satisfaction to be the strongest predictor of self-esteem; similarly, peer communication and appearance appraisal were also of significance. Interactions between these predictor variables and self-esteem warrant further attention as a means of establishing successful, domain-specific interventions.