Upcoming Events

Title: Web Survey Paradata 1-day short course
Date: Tuesday October 9th, 2018
Location: BLS Conference Center: 2 Massachusetts Ave.

Register Here: https://www.eventbrite.com/e/dc-aapor-short-course-web-survey-paradata-tickets-49696248818

  • Student/Retiree Members $75.00
  • Student/Retiree Non-Member $85.00
  • DC-AAPOR Members $110.00
  • Non-members $150.00 (DC-AAPOR membership included)

Paradata are the data automatically generated when respondents answer Web surveys. There are many different kinds of Web survey paradata, including email tracking tools, user agent strings (to identify devices used), and server-side and client-side paradata (CSP) providing information on things like response times, mouse-clicks, scrolling behavior, and so on. This course will provide participants with an overview of the different types of Web paradata, and how they can be collected, managed, and analyzed to provide useful information on data quality and nonresponse in Web surveys, leading to design improvements. The course focuses on the use of paradata rather than the technical aspects of capturing paradata.

The goals of the course are to help participants:

  • Understand the varieties and uses of paradata in Web surveys
  • Know what is needed to collect, process, and analyze paradata
  • Understand how to extract meaning from paradata
  • Make use of paradata to evaluate and improve Web surveys

Given the wide variety of Web software tools and methods to capture paradata, this will not be a hands-on course on the technical aspects of paradata collection. Rather it will focus on what kinds of paradata can be collected and how they can be used to evaluate surveys and inform Web survey design. The course assumes familiarity with designing and conducting Web surveys.

Mick P. Couper is a research professor at the Survey Research Center (SRC), Institute for Social Research, University of Michigan, and in the Joint Program in Survey Methodology, University of Maryland. He is author of Designing Effective Web Surveys (Cambridge University Press, 2008) and co-author of The Science of Web Surveys (Oxford University Press, 2013). He has been conducting research and implementing online surveys for many years, and more recently has been exploring the implications of mobile device use for Web survey design and for enhanced measurement. He is credited with developing the term “paradata” and has used Web survey paradata in many different settings.

Questions: Contact douglaswilliams@westat.com

Title: How U.S. Pre-election Polls are Changing in an Era of Disruption and Scrutiny
Date/Time: Wednesday, November 14, 2018   4:00–5:30 p.m.

Informal reception to follow at approximately 5:45 p.m. at East Street Café on the mezzanine level of Union Station.

Speaker: Scott Clement, Polling Director, The Washington Post
Chair: Sareeta Carter Schmitt, AP Statistics teacher, The School Without Walls of Washington, DC

Sponsors: WSS Statistics Education Committee, WSS Methodology Section and DC-AAPOR (The Washington-Baltimore Chapter of the American Association for Public Opinion Research)

Location: Bureau of Labor Statistics Janet Norwood Conference Center, Rooms 7/8 (Please check board in case of change of room). BLS is located at 2 Massachusetts Avenue, NE. Use the Red Line to Union Station. Parking in the area of BLS is available at Union Station. For parking information see http://www.unionstationdc.com/parking. No validation is available from BLS for reduced parking rates.

RSVP: To be placed on the seminar attendance list at the Bureau of Labor Statistics, you need to e-mail your name, affiliation, date of the seminar and seminar name to wss_seminar@bls.gov (underscore after ‘wss’) by noon on Friday, November 9. Please bring a photo ID to the seminar.

Abstract: The past decade has brought a rapid proliferation in U.S. pre-election polls, ways they are conducted and approaches to combining and interpreting results. This change has also come amid heightened scrutiny of the accuracy of pre-election polls and efforts to use polls to provide precise forecasts of election results, which came to a head with Donald Trump’s surprise victory in the 2016 election. A report by the American Association for Public Opinion Research found national polls that year were particularly accurate by historical standards, but that state surveys significantly underestimated Trump’s support. This talk will discuss the driving forces behind changes in pre-election polling, the chief challenges for pollsters and the news media in conducting and interpreting polls in the coming years. The talk is non-technical and relevant for both statisticians and non-statisticians.

No Remote Access Will be Available for this Event

POC (Point of Contact) email: Carol Joyce Blumberg, cblumberg@gmail.com