Two potential obstacles stand between the observation of a statistical correlation and the design and deployment of an effective intervention, omitted variable bias and reverse causality. Whereas the former has received ample attention in management research (and social sciences in general), comparably scant focus has been devoted to the latter in the research methods literature. Many existing methods for reverse causality testing commence by postulating a structural model that may suffer from widely recognized issues such as the difficulty of properly setting temporal lags, which are critical to model validity. In this paper, we draw upon advances in machine learning, specifically the recently established link between causal direction and the effectiveness of semi-supervised learning algorithms, to develop a novel method for reverse causality testing that circumvents many of the assumptions required by traditional methods. Our proposed method is well suited to address the chicken-or-egg questions that have intrigued management scholars for many decades. We carried out mathematical analysis and simulation studies to demonstrate the effectiveness of our method. We applied this method to examine the relationship between job satisfaction and work-family conflict using an archival dataset from the SWISS Household Panel Study. By doing so, we show how our method can be used to identify causal relationships that management scholar are interested in.