These observed sex differences highlight the importance of future analyses that take a person-centered approach to investigate the overlap and distinctions among closely-related phenotypes. Person-centered analyses such as latent class and latent profile analyses, are able to identify whether individuals tend to cluster into different groups or longitudinal trajectories (Swartout & Swartout, 2012). If alcohol-related, drug-related, and non-substance-related aggression are indeed three distinct constructs, then person-centered analyses would show different clusters and trajectories of individuals based on different combinations of these three aggression phenotypes. If these three forms of aggression were all reflections of a single underlying construct, individuals might only differ based on whether they engaged in relatively higher or lower amounts of all three forms of aggression. Future person-centered research is needed in order to examine these possibilities.
