Web Survey Bibliography

Title Comparing Face to Face, Telephone, Paper Self-Administered, and Web Survey Measurement
Author Tortora, R. D., Groves, R. M., Peytcheva, E., Manchin, R.
Source The American Association for Public Opinion Research (AAPOR) 59th Annual Conference, 2004
Year 2004
Access date 30.06.2004
Abstract While there are many methodological studies of mode effects in the research literature, most of them compare only two data collection methods. With the increasing number of mode choices, there is a growing need for studying the effects of mixed mode designs on measurement error. Ideally, studies would examine major data collection methods and compare mode effects on a variety of questions – open-ended, socially desirable, sensitive, etc. Such a study is conducted by Gallup Europe in conjunction with the European Social Survey, motivated by the anticipated need to change the dominant mode of interviewing, face-to-face. As the latter is either disappearing in some European countries, or becoming economically prohibitive, alternatives are needed. The study is the first of a series of methodological experiments to find optimal modes in various European countries. A quota sample of 1987 respondents representative of the Hungarian urban population by age, sex and education is drawn, randomly assigned to one of four modes - face-to-face, telephone, self-administered paper and pencil and web-based interviewing, and then reinterviewed in a different mode. Location of the second interview is either in-hall (a public location) or at the respondent’s home. In addition to inhall/ out-of-hall allocation, respondents are randomly assigned one of two versions of the survey, where some of the questions are manipulated within a mode. Mode effects are expected to become visible when responses in the first interview are compared to responses in the second interview. Pairwise mode comparisons are examined to determine which methods are most different from one another, the direction of dissimilarities and nature of the questions in which such differences are observed. Another set of analyses examines the overall effect of each mode, accounting for response homogeneity within a respondent. Causes of exhibited mode differences are discussed in the context of the existing literature.
Year of publication2004
Bibliographic typeConferences, workshops, tutorials, presentations
Print

Web Survey Bibliography - 2004 (512)

Page:
Page: