SURVEY RESEARCH UMBC调查的研究的项目.pptVIP

* * * * * * * * * * * * * * * * * * * * * Question Design (cont.) Closed-form option order (cf. ballot order) Which candidate did you prefer for the 2004 Democratic presidential nomination? (1) Howard Dean (2) John Edwards (3) Richard Gephardt (4) John Kerry (5) Other candidate; don’t know; no preference Which candidate did you prefer for the 2004 Democratic presidential nomination? (1) John Kerry (2) Richard Gephardt (3) John Edwards (4) Howard Dean (5) Other candidate; don’t know; no preference Question Design (cont.) Response bias Political Efficacy Index Index of respondents sense of “political efficacy,” built from responses to the following questions: Do you agree or disagree with the statement, People like me dont have any say about what the government does? Do you agree or disagree with the statement, I dont think public officials care much what people like me think? Do you agree or disagree with the statement, Sometimes politics and government seem so complicated that a person like me cant really understand whats going on? Do all the people who “agree” with all these statements really have a low sense of political efficacy, or are some of them just “agreeable” (i.e., tend to say they agree with any statement read to them)? Questionnaire Design These considerations point to the importance of pre-testing, survey experiments, and pilot studies: Advantage of using standard questions (e.g., POLI Student Surveys) How long will survey take? Question order can affect the distribution of responses to questions (issue preference and Presidential approval in 1984). Question context (i.e., the immediately preceding questions) can affect distribution of responses (e.g., Q4 in PS #1B). Early questions should reassure the respondent as to the legitimate and non-threatening nature of the questionnaire. “Controversial” questions should come last. What do you think is the most “controversial” standard question? Length of questionnaire. Survey