Can Study Design Influence Substantive Survey Results?
Chowdhury, D. (2007, August). Can Study Design Influence Substantive Survey Results?. Presented at ASA JSM 2007, Salt Lake City, UT.
The objective of the Synar survey is to determine the buy rate for each state based on random, unannounced inspections of a probability-based sample of tobacco outlets accessible to youth. The buy rate is defined as the proportion of tobacco outlets that sell cigarettes to minors. Some researchers claim that states can change the buy rate by manipulating the age and gender distribution of the inspectors. We will explore to what extent inspectors' age and gender impacts the buy rate. Our primary hypotheses are 1) female inspectors are more likely to be able to buy cigarettes than male inspectors, and 2) older inspectors are more likely to be able to buy cigarettes than younger inspectors. We will also study interactions and the influence of covariates and will include procedures for the control of bias introduced by the study design.