AAPOR
The leading association
of public opinion and
survey research professionals
American Association for Public Opinion Research

Webinar Details

Identifying Likely Voters in Pre-Election Surveys

Ruth Igielnik and Scott Keeter

Wednesday, March 23, 2016
12:00 - 1:30 PM CDT/ 1:00 - 2:30 PM EDT/ 10:00 - 11:30 AM PDT

ASA-SRMS members will receive AAPOR member pricing on webinars when registering for live webinars or purchasing recordings of webinars.
If you are an ASA member, click here to purchase.

Purchase a recording of this webinar nowPreview webinar recording

About This Course:

 
Election polls face a unique problem in survey research: They are asked to identify a population that does not yet exist at the time the poll is conducted, the future electorate. For instance, in a typical telephone survey before a midterm election, something on the order of 80-90% of registered voters say they will vote when in fact we know it’s often closer to 40%. This is because many people who are eligible to vote and who tell pollsters they intend to cast a ballot will not actually do so. Similarly, some people who express uncertainty about voting or little interest in the election will nevertheless turn out to vote. Likely voter models are used by pollsters to identify exactly that: people who are actually likely to vote in the upcoming election. This webinar examines the importance of such models and a variety of approaches used by pollsters to capture this population. Among the topics to be covered are the types of samples used for pre-election polls; the survey questions used to measure past behavior and gauge voting intentions and engagement in the election; and the modeling approaches used to combine the survey questions and records of past turnout to yield a forecast.

Learning Objectives:

 
  • Understand why likely voter models are necessary in election polling.
  • Identify typical questions included in likely voter models.
  • Understand the types of likely voter models typically used, including probabilistic and deterministic (cut-off) methods.
  • List a variety of voter file options and why they can be useful for election research.

About the Instructors:

 
SCOTT KEETER
is senior survey advisor for the Pew Research Center in Washington, DC. He is a past president of the American Association for Public Opinion Research, and has been an election night analyst of exit polls for NBC News since 1980. Keeter’s published work includes books and articles on public opinion, political participation and civic engagement, religion and politics, American elections, and survey methodology. A native of North Carolina, he attended Davidson College as an undergraduate and received a Ph.D. in political science from the University of North Carolina at Chapel Hill.
 
 
 
 
 
RUTH IGIELNIK is a Research Associate for the Pew Research Center in Washington, DC. In her two years at the Pew Research Center she has focused on understanding and identifying likely voters as well as incorporating big data and new methodologies into the Center’s work. Ruth received a bachelors in political science from University of Maryland and a master’s in Public Policy and Analytics from Carnegie Mellon University.