The investigators used both qualita tive and quantitative analytic tech niques in their genetic analyses. Quali tative techniques are used to assess traits that can only be either present or absent (e.g., alcohol dependence). One approach chosen in the COGA study was the affected sibling pair method, which examines the degree to which sibling pairs, both of whom exhibit the phenotype under investigation, share genetic markers. This method looks for deviation from the assumption that siblings on average share 50 percent of their genes. Thus, two affected siblings would share genetic factors that contribute to a given phenotype more than 50 percent of the time. Phenotypes studied using this method included alcohol dependence, low level of response to alcohol, the presence of alcoholism or depression, and being unaffected by alcoholism (all of these phenotypes are described in the follow ing sections). The statistical power of these analyses was limited because they involved only study participants who exhibited the phenotype under investi gation and thus were unable to take advantage of the detailed assessment available for all study participants.
