Objectives: Salivary testosterone (T) and cortisol (C) concentrations were monitored across a sports competition. Data were compared using two enzyme-immunoassay (EIA) methods and two sample preparations to determine their influence on hormone concentrations. Design and methods: A group of male athletes (n=19) provided a saliva sample the morning before and one day after (24h post) an international rugby union match. Following an extraction procedure, the samples were analysed for T and C concentrations using a commercial kit (CME) and an in-house method (IHE). Raw samples (no extraction procedure) were also tested using the commercial kit (CMR). Results: There were no significant changes in T and C levels from pre to post competition with each EIA method and sample preparation, but significant differences in T (IHE>CME>CMR) and C (CMR>IHE and CME) concentrations were seen when both samples were pooled. Bland-Altman analyses confirmed the presence of fixed and proportional bias. Strong and significant correlations were demonstrated between the IHE and CME measures of salivary T (r=0.93-0.97) and C (r=0.95-0.97). The T and C values from the raw and extracted samples were also strongly correlated (r=0.93-0.96). Conclusions: The measurement of salivary T and C concentrations across an international sports event was influenced by different EIA methods and sample preparations, but all measures were strongly correlated with some bias. Both T and C were unresponsive to the sports event, but within the group results large individual variation was seen. © 2012 The Canadian Society of Clinical Chemists.