# Survey Experiments # ## April 8 ## --- name: outline ## Outline - Guest presentation - Overview of survey experiments - Discussion of readings - Synopsis presentations --- ## Guest Presentation - Rune Slothuus, Professor in IFSK - Discussing survey experiment by Druckman, Peterson, and Slothuus (2013) in *American Political Science Review* --- ## Uses of survey experiments - Substantive experiments - Procedural experiments --- ## Substantive experiments - Only work well when units are individuals - "Lab in field" - Focus on sample representativeness --- ## Readings - Druckman et al. - Hainmueller and Hiscox --- ## Procedural experiments - How do we best measure some construct? - How do we recruit participants? - How do we prevent nonresponse or break-off? - How do we test respondent engagement? --- ## Procedural experiments - Question wording and ordering - Labelling and number of response scales - List experiment - Etc. --- ## Readings - Holbrook and Krosnick --- ## Protocol Considerations - Survey mode - Sampling - Control - Questionnaire Design --- ## Survey mode - Face-to-face - Computer Assisted Self-Interviewing - Computer Assisted Telephone Interviewing - Paper-and-pencil Self-Interviewing - Web-based (or mobile) Self-Interviewing ??? Interviewers? Costs? Training? Impact on measurement? Does mode affect responses? Can/should you implement multiple modes? --- ## Sampling - What is your target population? - How do you sample and contact them? - Do you need a representative sample? ??? Nonresponse affects representativeness Can you use a panel? What effect does that have? --- ## Control - Surveys are less controlled than lab - Setting - Broader context - Engagement - Surveys may be more private than lab - Respondents decide when to complete study --- ## Questionnaire Design - Also relevant to lab and field experiments - How do we measure constructs? - How do respondents understand those measures? - How do we hold respondents' attention? - Ethical considerations (esp. deception) --- ## Synopsis Presentations - Everyone present for 1-2 minutes - Discuss --- ## Preview Next week - Last seminar - Talk about uses of experimental findings Rest of semester - Work independently on exams - Setup a meeting with me (via email) in Late April to check on the status