In most empirical studies on civil wars, causes and determinants of conflict have been hitherto explored assuming that actors involved were either unitary or stable. However, if this intra-group homogeneity assumption does not hold, empirical econometric estimates may be biased and policy prescriptions are less reliable. We use Fixed Effects Finite Mixture Model (FE-FMM) approach to address this issue. This methodology provides a natural representation of heterogeneity when data originate from different latent classes and the affiliation is unknown. It allows to identify sub-populations within a population as well as the determinants of their behaviors. By combining various data sources for the period 2000-2005, we apply this new technique to the Colombian conflict. Our results confirm a behavioral heterogeneity in guerrilla’s armed groups and their distinct economic correlates so showing that different patterns of behavior exist. By contrast paramilitaries behave as a rather homogenous group.