Interview modality: Effects on costs and data quality in a sample of olderwomen

Objectives: To contrast the costs and quality of data collected in person, by telephone, and by self-administered questionnaires in a sample of older persons. Methods: Data collected via interviews with 586 women are contrast ed on four dimensions. Results: The cost of data collection was 25% to 30% lower for the self-administered mode than for other modes. Response rates w ere highest for the in-person interview and lowest for the self-administere d questionnaire. The rate of missing data was highest for the self-administ ered questionnaire. Significant differences in response effects were found across mode, with respondants to self-administered questionnaires profiling the poorest. Discussion: The costs and quality of data collected by variou s modes are discussed in terms of their implications for understanding the physical and mental well-being of older people.