Media dependency was examined as a complex construct involving the interactions of exposure to television news, exposure to newspapers, and expression of reliance on one medium or the other. A weighted sample of 2,402 respondents was used, representing the United States national population in 1976. A questionnaire assessed each subject's political activity, perceived efficacy of voting/political participation, television exposure, newspaper exposure, media reliance, age, educational level, and family income. Contrary to previous research results, the findings suggested that exposure to television news was not universally detrimental to political affect and behavior. For the majority of the sample who named television as the medium they relied on, television exposure was positively related to perceived political efficacy and to political activity. Television exposure was negatively related to efficacy and activity unless television was the relied upon medium. Newspaper exposure was positively related to the dependent variables except when television was the relied upon medium. (Author/RL)