|
Activity Number:
|
352
- Small Area Estimation, Analysis of Complex Sample Survey Data, and New Advances for Health Surveys
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Survey Research Methods Section
|
|
Abstract #317908
|
|
|
Title:
|
A Bayesian Model for Inference on Multiple Panel Public Opinion Surveys
|
|
Author(s):
|
Brittany Alexander*
|
|
Companies:
|
Texas A&M University
|
|
Keywords:
|
political polling;
bayesian modelling;
hierarchical models;
mrp;
panel survey
|
|
Abstract:
|
Panel surveys, which track individuals opinion over time using multiple administrations of the same survey, are common. Since they involve repeated measurements, the results from one survey date, or wave, are correlated with the next wave. Combining information from two panel surveys with different numbers of waves or dates administered is non-trivial. We present a case study in which we use Bayesian inference to combine two panels about terrorism policy from 2016. The first panel was a large 1730 individual two wave probability-based panel with dropouts taken six months apart in May 2016 and November 2016. The second panel was a non-probability panel that had six waves taken every month from June 2016 to November 2016 and had 779 respondents including dropouts and includes a 110 person replenishment sample. The model uses multilevel regression with poststratification and partial pooling across time to create a regression model to understand public perceptions of terrorism and support for counterterrorism policies.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2021 program
|
