It is hard to judge the potential usefulness of a network approach to personality research because Cramer et al. (2012) mix up applications to one individual, inter-individual differences and intra-individual processes. From each perspective, the network units, their associations and causal interpretations of such associations have a completely different meaning, and it depends on the particular perspective, the level of aggregation and whether one wants to model measurement error whether latent variables have a place in network models in personality research. Copyright (c) 2012 John Wiley & Sons, Ltd.