Activity Number:
|
354
- SPEED: Big Data, Small Area Estimation, and Methodological Innovations Under Development, Part 2
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, July 30, 2019 : 10:30 AM to 11:15 AM
|
Sponsor:
|
Survey Research Methods Section
|
Abstract #307739
|
|
Title:
|
A Computationally Efficient Method for Selecting a Split Questionnaire Design
|
Author(s):
|
Matthew Stuart* and Cindy Yu
|
Companies:
|
and Iowa State University
|
Keywords:
|
questionnaire;
SQD;
Horvitz-Thompson;
Pet;
KLD;
design
|
Abstract:
|
Split questionnaire design (SQD) is a relatively new survey tool to reduce response burden and increase the quality of responses. An SQD among a set of possible alternatives is considered as the best if it minimizes the amount of information loss quantified by the Kullback-Leibler divergence (KLD) distance. The calculation of KLD distance requires computation of likelihoods for the observed after integrating out all the missing variables in a missing SQD. For a typical survey questionnaire with a large number of categorical variables, this computation can be practically infeasible. Motivated by the Horvitz-Thompson estimator, we propose a way to approximate the likelihood of the observed data in much reduced computation time and lose little valuable information when comparing different choices of SQDs. We contrive a thorough simulation study, meant to represent the 2012 Pet Demographic Survey sponsored by the American Veterinary Medical Association and created to cover different distribution shapes of continuous variables, as well as a study on the empirical data to show that the proposed approximation method can correctly identify the best SQD among a set of alternatives.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2019 program
|