[Taylor Boas] Sampling Social Media: Using Facebook to Recruit Online Survey Respondents

Wednesdays @Hariri

3:00 PM on December 3, 2014 @MCS-180

Sampling Social Media: Using Facebook to Recruit Online Survey Respondents

Taylor Boas

Junior Faculty Fellow, Hariri Institute for Computing
Assistant Professor, Department of Political Science
Boston University

Abstract: Online surveys are attractive to many social scientists due to their relatively low cost and the ease and reliability with which one can randomize or customize question wordings and response options. Yet recruiting a diverse sample of respondents for such surveys is often a challenging task. Drawing on online surveys conducted in Brazil (2012, 2014) and Chile (2013), I discuss the advantages and disadvantages of using Facebook advertisements to recruit survey respondents. Facebook’s global presence and high penetration rates within many countries allows one to recruit opt-in samples that are highly representative of the general population on variables such as race, political attitudes, and region of residence. Yet survey sampling using a platform oriented toward commercial advertisers—who are typically more interested in narrow targeting than broad representativeness—presents certain challenges. I also discuss the representativeness of samples gathered using an alternative recruitment method, Amazon.com’s Mechanical Turk, and I discuss a proposal for directly comparing both methods using online surveys in India and the United States.

Bio: Taylor C. Boas is an Assistant Professor in the Department of Political Science at BU, which he joined after receiving his Ph.D. from the University of California, Berkeley in 2009. His research examines various aspects of electoral politics in Latin America, including campaigns, political communication, voting behavior, and religion and politics, and involves a variety of methods, such as online survey experiments and the statistical analysis of electoral data. Professor Boas has played a lead role in developing interdepartmental programs for training social science graduate students in quantitative methods. Prior to starting at BU, he was a Visiting Fellow at the Kellogg Institute for International Studies, University of Notre Dame.