Exercise-induced testosterone response has been examined by exercise scientists for decades. However, the development of salivary testosterone (sal-T) analysis has meant increased ease of sampling. This has led to an exponential increase in research articles concerning exercise-induced sal-T response since the turn of the century. To the authors’ knowledge, no meta-analysis to date has concerned sal-T measurement in in response to acute exercise bouts.A comprehensive literature search was conducted using the search term ‘saliva AND testosterone AND exercise’. Randomized- and non-randomized controlled trials featuring healthy adult males conducting an exercise intervention were examined. Meta-Analysis software was used to determine standard differences in means (SDM).Three separate meta-analyses were performed as aerobic, anaerobic, and resistance exercise has appreciably different physiological demands. After the initial database search, 933 titles were identified. Once duplicates were removed, titles and abstracts were screened for inclusion and 80 studies were retrieved as full text and assessed for eligibility. 21 studies were included in the quantitative synthesis. There was a stepwise effect of exercise intervention type on SDM with aerobic, resistance, and anaerobic interventions resulting in an overall effect of 0.923 (95% CIs = 0.489-1.357), 0.906 (95% CIs = 0.398-1.453), and 0.325 (95% CIs = -0.377-1.028) respectively. For all intervention types (i.e. aerobic exercise, anaerobic exercise, and resistance exercise), there were significant differences in overall weighted SDMs for sal-T response between randomized control trials (RCTs), baseline as control perspectives, and control group perspectives.Discussion: All aerobic studies show a positive change in sal-T indicating an increased concentration except Doan et al. (2007). This may be due to the study concerning 36-holes of golf and therefore sal-T may follow the circadian rhythm and not be influenced by the intervention to such an extend as in other studies. SDM for anaerobic interventions was dependent on saliva sampling regime. Baseline samples taken on a control day produce a negative SDM post exercise (Eloumi et al., 2004; McLellan et al., 2010) whereas all other studies produced a positive SDM. SDM for resistance exercise was dependent on post-exercise sampling regimen. Crewther et al. (2008) demonstrated a reduction followed by an increase in sal-T following a power training protocol and a paradoxical effect following a maximal strength protocol.Timing of saliva sampling appears to influence results to an equal or even greater extent than an exercise intervention investigated. Therefore, it is necessary to define methodological guidelines for analysis of exercise-induced sal-T response.