Markowitz and Grossman (2000) expanded upon this study in two important ways. First, they performed the analyses separately by gender of the parent, which is important because different patterns of drinking and violence have been observed for men and women. Second, they added data from another comparable survey conducted 10 years later, the 1985 National Family Violence Survey, which allows for a comparison of the effects of alcohol regulation over time as well as for the pooling of several years and the addition of State-level fixed effects. Such fixed effects are important in determining whether the effects of the State-level alcohol regulation variables reflect State sentiment toward regulation and violence, which cannot be measured directly, rather than true policy effects.
