Lack of Theory Building and Testing Impedes Progress in The Factor and Network Literature

Abstract

The applied social science literature using factor and network models continues to grow rapidly. Most work reads like an exercise in model fitting, and falls short of theory building and testing in three ways. First, statistical and theoretical models are conflated, leading to invalid inferences such as the existence of psychological constructs based on factor models, or recommendations for clinical interventions based on network models. I demonstrate this inferential gap in a simulation: excellent model fit does little to corroborate a theory, regardless of quality or quantity of data. Second, researchers fail to explicate theories about psychological constructs, but use implicit causal beliefs to guide inferences. These latent theories have led to problematic best practices. Third, explicated theories are often weak theories: imprecise descriptions vulnerable to hidden assumptions and unknowns. Such theories do not offer precise predictions, and it is often unclear whether statistical effects actually corroborate weak theories or not. I demonstrate that these three challenges are common and harmful, and impede theory formation, failure, and reform. Matching theoretical and statistical models is necessary to bring data to bear on theories, and a renewed focus on theoretical psychology and formalizing theories offers a way forward.