A methodological review was conducted to critically analyse the strengths and weaknesses of distinct research designs employed with a variety of populations to investigate the functions of self-injury. Identified designs included: retrospective self-report; retrospective informant report; qualitative phenomenological interviews; direct observation; ecological momentary assessment; experimental functional analysis; and lab-based self-injury proxy studies. The inclusion of multiple functional assessment methods within future studies was strongly supported. The empirical and discussion papers focused on the functions of suicidal and nonsuicidal self-injury within clinical populations of adults and adolescents. Participants completed the Suicide Attempt Self-Injury Inventory (Linehan, Comtois, Brown, Wagner, & Heard, 2006) to assess the reasons, antecedents, and consequences associated with different methods of self-injury. Multiple methods of self-injury, serving multiple functions, were reported by all participants. Within-person analyses found that individuals’ nonsuicidal acts were intended to relieve negative emotions and punish themselves, more than their suicidal acts, and resulted in greater reductions in feeling numb/dead. Suicidal acts were intended to benefit others, preceded by intense feelings of burdensomeness, and resulted in receiving help, more than nonsuicidal acts. Within-person comparisons for individual methods of self-injury found that cutting was intended to relieve negative emotions, occurred following an argument, and resulted in the generation of pain and/or normal feelings, more than self-injury involving hitting the body. The experience of conducting the research, and the motivation behind it, were reflected on. When considering the contribution of the findings to future theory and research, two areas were focused on: evidence for conceptualising suicidal and nonsuicidal self-injury along a continuum of experiential avoidance behaviours; and evidence regarding the stability of self-injury functions across different methods, situations, and time. Specific implications were highlighted for clinical interventions aimed at addressing underlying vulnerabilities, and the multiple triggers and reinforcing consequences, of these life-threatening behaviours.