The IMO framework has had a critical impact on our understanding of why some teams are more effective than others. However, despite the multitude of studies drawing from and testing elements of the IMO framework, as well as its centrality to conceptual work in teams research, very few empirical studies have conducted true longitudinal tests of the various phases of the model (i.e., I?M; M?O), and even fewer have provided longitudinal tests of the entire model (i.e., I?M?O) and its recursive paths. Accordingly, we use cross-lagged panel (CLP) meta-analytic methods to empirically evaluate the extent to which the IMO framework provides a substantiated view of team dynamics. In contrast to the much larger and significant synchronous path effects, we find that when inputs, mediators, and outputs are modeled in a causal fashion using CLP data, the magnitude of direct effects proposed by the model becomes smaller in the case of the path from mediators to outputs, and nonsignificant for the path from inputs and mediators. We also find that the indirect effect implied by the model is small and nonsignificant. Further, the magnitude of effects for the recursive relationships are comparable to those of the direct effects, although we find some differences across analytic methods. We also found some support for alternative orderings of constructs in the IMO model. Our results support the notion that team constructs are dynamically related over time while supporting calls to reconsider the use of the IMO framework to model team dynamics. We situate our findings in the broader teams literature and elucidate our contributions to theory and research on team dynamics.
