Subjects were randomly assigned to the four cells of the balanced placebo design, with 10 males and 10 females per cell. Following told-alcohol or told-no alcohol beverage instruction manipulations, participants consumed either a vodka-tonic beverage containing a dose of vodka sufficient to induce a peak blood alcohol level of 0.05% or a beverage containing only tonic water. Subjects' self-report ratings of beverage alcohol content indicated that alcohol overrode the effects of beverage instructions in the told-tonic/given-alcohol condition. It was concluded that the design cannot independently evaluate effects of both alcohol and beverage instructions when behaviorally significant alcohol doses are administered.