A comparative approach to studying the spread of policy innovations has recently yielded new and interesting results, as well as theoretical advancements, for policy diffusion research. Specifically, punctuated equilibrium theory has been offered as an explanation for why some policies spread quickly, while others do so normally, and still others are adopted very slowly. Studies of adoption speed, however, currently rely on careful case selection or a dichotomous categorization of adoptions as fast or slow to test why policies diffuse at different speeds. Building on this foundational work, I propose a method for measuring adoption speed as a continuous concept, so that it can be modeled directly as an important outcome of interest to diffusion scholars. I then use the new measure to evaluate how adoption speed varies across time and policy domain. I further demonstrate the utility of the measure as a dependent variable by replicating past results, including the interactive effect of complexity and salience on adoption speed and positive effect of federal incentives, as well as finding preliminary evidence that policy clusters spread more rapidly than the average stand-alone policy.

Pooled event history analysis (PEHA) allows researchers to study the effects of variables across multiple policies by stacking the data and estimating the parameters in a single model. Yet this approach to modeling policy diffusion implies assumptions about homogeneity that are often violated in reality, such that the effect of a given variable is constant across policies. We relax this assumption and use Monte Carlo simulations to compare common strategies for modeling heterogeneity, testing these strategies with increasing levels of variance. We find that multilevel models with random coefficients produce the best estimates and are a significant improvement over other models. In addition, we show how modeling similar policies as multilevel structures allows researchers to more precisely explore the theoretical implications of heterogeneity across policies. We provide an empirical example of these modeling approaches with a unique data set of 29 antiabortion policies.