Several studies have reported conflicting results when assessing associations between stress and repetitive behaviors in individuals with autism spectrum disorder (ASD). Some factors that may have caused these discrepant results include the monitoring of a single broad category for repetitive behaviors, the heterogeneity of the participants, and the use of indirect measures. To address the prior issues, our study explored the relationship between salivary cortisol and direct observation measures of stereotypy in four minimally verbal children with ASD. To this end, we combined an alternating-treatment design with multiple regression analyses to examine the interaction between the two variables. The analyses indicated that the mean value of cortisol was negatively associated with global and motor stereotypy. No significant relation was found between mean value of cortisol and vocal stereotypy. These results highlight the complex relationship between stress and stereotypy and emphasize the relevance of conducting research on a larger scale, which would have a direct impact on our understanding of a core feature of ASD.