Guide to survey_poll
- 1. A citizens’ guide to survey research and public opinion polls (2017)
- 2. One definition of public opinion (Erikson and Tedin. American Public Opinion) “We define public opinions as the preferences of the adult population on matters of relevance to government.”
- 3. Glynn et al (2004). Chapter 1. The Meanings of Public Opinion “The study of public opinion does concern the formation, communication, and measurement of citizens’ attitudes toward public affairs.” (p.19)
- 4. Questions scholars have asked (1930s-) What do citizens know about issues, office holders, how the government works? Are opinions “constrained”? What structures one’s opinions? How opinion forms and changes. What group differences are there in opinions?
- 5. Attitude: the anchor of opinion Dominant theory today Attitude: “learned predisposition to respond in a consistent manner with respect to a given object.” Attitude has cognitive (“belief”), affective (“emotion”), and conative (“behavior”) components. People acquire attitude from parents, etc. Socialization process
- 6. Competing views of citizens Super Citizen“Know nothing”
- 7. Two approaches to research Observational (or correlational) methods Survey, poll (ask people questions) Focus group (let people talk) Content analysis Experimental methods Lab experiments Field experiments (e.g. Green and Gerber@ Yale University)
- 8. An alternative to survey/poll Focus Group research Open-ended responses (vs. survey) Give-and-take, follow up Group dynamics Fluid and dynamic nature of opinion formation and expression Layers of ideas and thoughts Focus group
- 9. Conover et al. (1991) “The Nature of Citizenship in the United States and Great Britain” Uses focus group discussions Planned group conversations Tape recorded and sometimes videotaped Carefully planned questions Discussions not closely controlled Participants encouraged to use the language they actually use Focus group
- 10. Conover et al. (1991) “The Nature of Citizenship in the United States and Great Britain” Four groups in the U.S., four in U.K. 7-11 participants including the interviewer Two-hour discussions (U.S.) Focus group
- 11. Conover et al. (1991) “The Nature of Citizenship in the United States and Great Britain” What are your rights as citizens? Focus group
- 12. Public Opinion Polls Polls may not be the best way to capture opinion Not all “polls” are legitimate High degree of skepticism and hostility toward polls What is the proper role of opinion polls?
- 13. Legitimate Public Opinion Polls The rule of thumb Use scientific methods Methods clearly stated Sponsored by research organizations Commissioned by companies or government offiices Purpose: learn what the public thinks and why
- 14. NOT Legitimate Public Opinion Polls “Push” polls in campaigns (smearing opponent under the guise of research) “FRUGging” (fundraising under the guise of surveying) “SUGging” (selling under the guise of research) Surveys to generate desired results Surveys as source of entertainment
- 15. More examples of pseudo-polls Online “flash” poll (“instant” poll, “quick” poll….) Straw poll “Street Corner” poll
- 16. Positive contributions of polls to democracy Citizens learn about others’ opinions Office holders learn about people’s opinions Input is more equal than voting More precise than voting
- 17. Negative effects of polls on democracy “Governing by polls” – politicians follow rather than lead Encourages horserace mentality False sense of being influential
- 18. Public opinion survey or poll “a research technique for measuring characteristics of a given population of individuals.” (Glynn et al, Public Opinion. p.76) Uses scientific method (more or less) accurately measures public opinion
- 19. The Literary Digest fiasco (1936) A popular monthly magazine Correctly predicted winner in 1928, 1932 presidential elections Received 2 million “ballots” back Predicted Landon (R) win Landslide for Roosevelt (D) George Gallup and others were right Beginning of scientific surveys
- 20. Public opinion survey/poll: the process Define population Define sampling frame Take a sample of the population Determine methods of data collection Gather data about the sample Draw conclusions about the population Key is maximizing the fit b/w sample and population
- 21. Random samples “Probability” sample Telephone RDD Male or Female who was born in… Cell phone
- 22. Opinion Poll Checklist Population defined Sample size clearly stated Large enough sample Random selection of sample (more or less) Method (telephone, for example) Date poll was conducted Sampling error (“Margin of error”) Confidence level (95% in most cases)
- 23. Survey Methods (example) Results for this survey are based on telephone interviews conducted under the direction of Princeton Survey Research Associates International among a nationwide sample of 2,005 adults, 18 years of age or older, from October 15-20, 2016 (1,507 respondents were interviewed on a landline telephone, and 500 were interviewed on a cell phone, including 199 who had no landline telephone). Both the landline and cell phone samples were designed by Survey Sampling, Inc……For results based on the total sample, one can say with 95% confidence that the error attributable to sampling is plus or minus 2.5 percentage points.
- 24. Examples of inappropriate “survey” questions Do you feel that stress in your relationship is damaging your health? (CNN.com) (To a Democratic member of Congress) “George Bush. Good president? Or an excellent president?” (The Colbert Report) Do you feel safe with Trump as president? Is Obama finished? Vote here now!
- 25. How (not) to write survey questions Avoid compound questions. Avoid double negatives. Avoid leading questions. Avoid argumentative questions. Avoid ambiguous questions. Avoid emotionally charged words. Present all appropriate response options. Use clear, unambiguous, neutral, objective words.