Adults' responses to a survey of television viewing were investigated to examine the premise that respondents assume that the range of precoded response alternatives reflects the researcher's knowledge of the distribution of opinions or behaviors in the population. Responses are, therefore, a function of the response alternatives provided. Respondents assume that the average respondent is represented by the values in the middle range of the response alternatives, and that the values at the extremes of the scale also represent the extremes of the distribution. These assumptions may mediate the impact of response alternatives on respondents' reports in two ways. Respondents may either use the range of the response alternatives as a frame of reference in estimating their own behavioral frequencies, or they may be reluctant to report frequencies that appear extreme in the context of the scale. Three experiments were conducted to differentiate between the frame of reference and the social desirability hypotheses. Subjects included American undergraduate students and West German adults. The results of all studies favored the frame of reference hypothesis, i.e. that respondents use the range of the response alternatives as a frame of reference in estimating their own behavioral frequencies. (Author/MAC)