Tuesday, May 11 at 12:15pm to 1:15pmVirtual Event
Our upcoming candidate has a Ph.D. in Statistics.
Research Talk, "Two National Surveys of Tobacco Use: Statistical Considerations and Beyond."
The objectives of my talk are: (1) to describe two national surveys of tobacco use behaviors in terms of survey sampling characteristics, statistical analysis considerations, the process for getting data access, and study limitations associated with the survey, and (2) to provide a summary of pros and cons of the surveys. I will describe two national surveys, Tobacco Use Supplement to the Current Population Survey (TUS-CPS) and American College Health Association-National College Health Assessment (ACHA-NCHA). The TUS-CPS is a key source of tobacco use among adult population in the U.S. It uses a multi-stage (also termed complex) sample of about 240,000 respondents per survey period. To account for nonresponse, oversampling some groups and design specifics, statistical analyses should include survey weights, e.g., variance of an estimate should be computed using Balanced Repeated Replications with 160 replicates. Utilization of the TUS-CPS data is illustrated using a study conducted to estimate longitudinal trends of current smoking prevalence in the U.S. and assess whether the trends differ by survey mode and respondent type. The ACHA-NCHA is a nation-wide assessment of college students’ health behaviors including tobacco use behaviors. I illustrate how the data were used in a study of ever-use of tobacco products among sexual and gender minority students at University of Central Florida. In addition, I summarize the key pros and cons of each survey and emphasize importance of a data source choice when designing a study.