While nonprobability samples to some offer worthwhile future prospects in the polling industry, others recommend using them with caution.
Dr. Dutwin will describe over two years of research completed by SSRS in partnership with the Data Science team at the University of Massachusetts-Boston regarding cutting-edge methodology and weighting techniques for the best use of a) nonprobability samples, and b) hybrid probability and nonprobability samples. So if, for a given project, telephone is unaffordable, probability panels are not a great fit, and nonprobability are too risky to rely upon, what is a well-intentioned survey researcher to do? One approach is used for studies that gather all of their data via a nonprobability panel by using a machine learning technique to identify key interactive variables and utilize those interactions to re-calibrate the nonprobability sample. A second approach leverages state of the art of Big Data analytical techniques to provide the best solution possible for combining the probability and nonprobability data into a universal sample. The goal is not to get the nonprobability sample to be representative, but rather for it to “contribute” to representativeness in concert with the probability sample.
Drs. Jenkins and Koning will discuss the results of a joint experiment utilizing RDD, online probability samples, and online nonprobability samples at the state level. Because of the need to move away from telephone surveys, the Rutgers-Eagleton Poll at Rutgers-New Brunswick’s Eagleton Institute of Politics and the FDU Poll at Farleigh Dickinson University have combined their resources to conduct one of the first ever in-depth experiments testing the effects of both survey mode and type of sample on statewide public opinion polling. The extensive study involves testing an identical questionnaire on three different samples. What demographic and substantive differences did they find among these three types of sampling?
SPEAKERS:
- David Dutwin, PhD, SVP NORC and Senior Fellow at the University of Pennsylvania
- Krista Jenkins, PhD, Director, The FDU Poll
- Ashley Koning, PhD, Assistant Research Professor & Director, Eagleton Center for Public Interest Polling, Rutgers University
When:
Wednesday, December 4, 2019
Where:
- Princeton University, Wallace Hall, Room 300
- Parking available in Lot 10 or Lot 13 just to the north of Robertson Hall
SCHEDULE:
- 6 PM to 7 PM – Networking and Refreshments
- 7 PM to 8 PM – Panel Discussion
REGISTRATION AND PRICING:
To register for the event, please use the PayPal link or contact klubin@fuelcycle.com.
- Member: $20
- Non-Member: $25
- Students: Free (student registrants please email Kevin Lubin at klubin@fuelcycle.com to confirm registration; no PayPal ticket purchase required)
- Thank you to our generous sponsors for making events like these possible: