Modeling Heaping in Self-Reported Longitudinal Cigarette Count Data

• Activity Number: 219
• Type: Invited
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Society for Research on Nicotine and Tobacco
• Author(s): Hao Wang*+ and Daniel F. Heitjan
• Companies: The Johns Hopkins University and University of Pennsylvania
• Keywords: Heaped data ; rounded data ; longitudinal data ; latent variable ; measurement error ; smoking cessasion

Abstract

In studies of smoking behavior, some subjects report exact cigarette counts, whereas others report rounded-off counts, particularly multiples of 20, 10 or 5. This form of data reporting error (heaping) can bias the estimation of parameters of interest and result in misleading inference. Self-reported consumptions collected by the time-line follow-up approach (TLFB) may not adequately characterize true consumption on a day-to-day basis. We present a method to describe longitudinal heaped cigarette count data. We formulate two latent variables; one representing the underlying true value, the second characterizing the misreporting behavior for measurement errors. The analysis of a smoking cessation trial that measured cigarette consumption using both TLFB and Ecological Momentary Assessment helps to understand the utility of the TLFB instrument in characterizing daily smoking patterns.