Use of the Five Minute Speech Sample in parenting research has become increasingly widespread in recent years, raising important questions about how it maps onto other more established methods for assessing parent–child dynamics. Given the historical emphasis on direct observation as a gold standard in such assessment, our aim was to examine associations between the Five Minute Speech Sample and direct observational coding of parent–child interactions, and to evaluate the assumption that the Five Minute Speech Sample reflects moment-to-moment exchanges between parents and children. A systematic search of three databases conducted identified 25 relevant studies (total N = 2945 child participants). These studies focused largely on four distinct systems for coding the Five Minute Speech Sample, and reported on distinct developmental periods ranging from infancy (17 months) through to adolescence (17 years). In 20 of 25 studies, the Five Minute Speech Sample was significantly associated with observations of parent–child interactions. These associations were apparent in all age groups examined, yet findings for fathers were somewhat more mixed than mothers. Available evidence suggests that the Five Minute Speech Sample holds strong potential as a brief but richly informative tool for indexing parent–child dynamics—particularly affective dimensions of the parent–child relationship—in both research and clinical contexts.