Objective: Legislators, researchers, and clinical practitioners are increasingly aware of coercive and controlling behaviors and attitudes, in relation to intimate partner violence (IPV). There is limited empirical research into how specific behaviors and attitudes might aid understanding and measurement of coercive control, and into the role of coercive control in predicting physical IPV, including the severity of future violence. Method: In a secondary analysis of 1,421 police reports of IPV by men against their female intimate partner, we applied structural equation modeling to test whether a construct of coercive control could be defined by variables measuring nonphysical abuse and attitudes. We tested whether coercive control was associated with physically violent IPV and future IPV recidivism in a retrospective longitudinal follow-up. Results: Analyses suggested two main factors, one we called “Psychological Control” (jealousy, psychological abuse, stalking, and suicide threats) and the other, “Controlling Attitudes” (controlling behavior, IPV attitudes, and IPV denial). Both latent factors defined a second-order “Coercive Control” factor. Coercive Control was significantly associated with physical IPV at index, as well as the occurrence and severity of subsequent physical IPV. Conclusions: Findings are consistent with the view that coercive control is a distinct concept comprising a variety of behaviors and attitudes that can be documented during police domestic dispute investigations. Results also indicate the importance of coercive control in the understanding and assessment of risk of physical assault in IPV. Research is needed with more comprehensive measures of coercive control and IPV, using multiple data sources and gender-inclusive samples.