Differences in demographics and behaviors across two web-based survey platforms: Observations from a study of risk perceptions of heated tobacco products (HTPs).

With an escalating popularity of online surveys in behavioral research, it is critical to understand how different sources of participants can yield differing outcomes. While Mturk has been used for online surveys for almost two decades, a recent introduction of online panels allows researchers to choose participants from variety of pools. This study aims to contribute to existing knowledge of how participants from different online platforms differ in their characteristics and behavioral responses which might affect the outcomes. 300 participants were recruited each from Mturk and Prime panels for a 20 mins long survey assessing perceptions and intentions to use Heated tobacco products (HTPs). Participants answered demographic and tobacco-use related questions including their vaccination and masking for COVID-19. They were shown a picture and description of a recently launched HTP. Further, participants answered questions about their awareness of HTPs, risk perception of health conditions from use of different tobacco products (cigarettes, e-cigarettes and HTPs) and perceived severity of COVID-19 infection in smokers, vapers and HTP users. Results showed significant differences in Mturk and Prime panel participants' demographics and tobacco-use. Prime panels showed more racially diverse population (chi-sq = 10.07, p < 0.02) and significantly more current smokers (chi-sq = 44.74, p < 0.01) and current e-cigarette users (chi-sq = 38.04, p < 0.01) compared with Mturk. Mean perception scores for COVID risk in tobacco users were significantly different between Prime panels and Mturk. Study highlights significant differences in sample composition and responses that might be helpful in choosing one online platform over another based on specific study requirements.