Given that an item can be included in the score only if the participant reports having experienced that effect from drinking alcohol, SRE scores (and, hence, sensitivity estimates) tend to be biased due to an inherent correlation between the number of items endorsed and number of drinks reported (i.e., lighter drinkers are less likely to experience effects associated with heavier doses). In other words, missing items are not missing at random. Although SRE scores have been shown to predict future alcohol consumption and problems after accounting for number of SRE items endorsed (Schuckit et al., 2006), this issue has also been addressed by alternative methods of scoring in previous studies (Schuckit et al., 2007). More recently, Lee et al. (2015) used data from the current study’s full sample to propose two additional methods to correct for this bias, including standardized person-mean imputation in which item responses are converted to z-scores before averaging across non-missing items, thereby achieving comparable item distributions across individuals. In the current data, SRE-Total scores were computed using both conventional scoring and standardized person-mean imputation. Both approaches
