Agnew introduced a new integrated theory; the General Theory of Crime and Delinquency, in which he attempted to corral the most influential predictors of criminal behavior into more parsimonious propositions of multiple life domains—self, family, peer, school, and work—as well as constraints against crime and motivations for it. This study presents a partial test of the theory using longitudinal data of 2,351 Korean adolescents. A group-based modeling approach (latent class growth analysis) was run to examine direct effects of life domains on peer delinquency as well as mediating effects of constraints and motivation on their relationships. The study identified three subgroups: early onset/decreasing (3.2%), moderate (12.4%), and low/none (84.4%). The findings revealed that the self and peer domains exhibited a positive impact on the early onset/decreasing trajectory group compared to the low/none group with the constraint exhibiting a negative impact. The moderate trajectory group demonstrated that the self-domain was significant but was not rendered insignificant after controlling for constraints and motivations. The study provided moderate support for life domains within Agnew’s new theory for peer delinquency in nonwestern countries.