Individuals who participated in both waves did not differ from those who participated in only the first wave on any demographic, drinking, or trait variable. Accordingly, we concluded that data were missing at random. We therefore imputed missing data using the expectation maximization (EM) procedure [41] 1. In recent years, several monte carlo studies have been conducted that compare a set of data imputation procedures, including the EM method, to traditional, alternative methods of handling missing data, including deletion of missing cases and mean imputation [41]. Consistently, those studies have found that the EM procedure produces less biased estimates of full sample values. This resulted in a total sample size of 418 for the SEM procedure.
