Group projects are an established but debated pedagogical technique in higher education. The purpose of this study was to assess the appropriateness of combining individual and group marks in assessment. A mixed method design involving correlational and comparative elements was used. The sample included one cohort of students who completed a group project at Level one (n = 127) and Level three (n = 103) of an undergraduate occupational therapy BSc degree. Key findings included no statistically significant correlation between group and individual assessment marks; a significant proportion of students failing the individual written assignment passed the module overall; and neither academic ability nor demographic factors predicted group performance. This suggests that group assessment measures factors other than individual academic performance. The implications are that group assessment should be clearly linked to module learning outcomes; should explicitly grade relevant non-academic skills; and should be used with caution as a Level one assessment. (Contains 6 tables.)