Jon Krosnick, Stanford University
Tuesday, April 15th
12:00–1:30 PM
*Due to popular demand, we have moved the location to the Connection Space at SSRI, Gross Hall, 2nd floor
Host: Duke Initiative on Survey Methodology (DISM)
RSVP if you plan to join us
Attendance is free and lunch is provided
Are Surveys Destined for Death?
Recent headlines around the world have been blaring:
“Polls are failing us”
“Why are the polls getting it wrong in so many countries?”
“British Election’s Other Losers: Pollsters”
Is survey research on the verge of death? Will survey data collection be replaced by analysis of “big data”?
This presentation will offer a review of evidence on the accuracy of all types of surveys and explore whether accuracy has changed over time and is tied to the survey methods employed. Special attention will be devoted to the dramatic transformation in the collection of survey data that has occurred during the last 25 years: from random samples interviewed face-to-face or by telephone years ago to internet data collection today worldwide, mostly from samples of people who have not been scientifically sampled from populations. The presentation offers surprising and reassuring findings about the future of the field of survey research and practical advice about how to maximize accuracy and how to educate survey sponsors about the value of survey measurements.
Jon Krosnick is a social psychologist who does research on attitude formation, change, and effects, on the psychology of political behavior, and on survey research methods. He is the Frederic O. Glover Professor in Humanities and Social Sciences, Professor of Communication, Political Science, and (by courtesy) Psychology. At Stanford, in addition to his professorships, he directs the Political Psychology Research Group and has directed the Summer Institute in Political Psychology.