Comparing Twitter and Online Panels for Survey Recruitment of E-Cigarette Users and Smokers

Introduction: E-cigarettes have exploded in popularity in recent years, driven, at least in part, by marketing and word-of-mouth discussion on Twitter. However, there is scarce research on marketing and use of e-cigarettes. While focus groups and interviews offer insights, these methods are time-intensive and produce only qualitative data. The rapid proliferation of e-cigarettes leaves researchers in need of timely quantitative data from a hard-to-reach user population, given that only 13% of US adults are e-cigarette users. Online panels offer a method of accessing this population, but have been criticized for recruiting too few young adults (Heeren et al., 2008), among whom e-cigarette use rates are highest. Recruiting hard-to-reach populations via social media offers a promising supplemental resource for online panels. This study compares the effectiveness of recruiting e-cigarette users via online panel to Twitter, and explores how users recruited differ by demographics and e-cigarette-use. We also discuss how results compare to those from smokers to understand how insights apply to other subgroups.

Methods: Participants were adult ever e-cigarette users (65.9% male; Mean Age= 32.92, SD= 10.76) who reported having public Twitter profiles. Participants were recruited via online panel (N= 420), or promoted Tweets using keyword targeting for e-cigarette users (N= 411).  Independent variables were demographics (age, sex, education, race/ethnicity), e-cigarette use (eg, past 30 day e-cigarette use (M=13.83; SD=10.99), e-cigarette puffs per day (M=31.36; SD=89.96)) and social media behaviors (eg, Twitter use frequency).  Recruitment method (Twitter, panel) was the dependent variable.  Multivariate regression analyses were run in STATA 13.1.

Results: We found that adult e-cigarette users aged 25-34 years (O.R.=.09, p<.01) and 35-44 years (O.R.=.05, p<.01) were less likely to be recruited via Twitter (than panel) compared to young adult e-cigarette users. People who used e-cigarettes on a larger number of days (of the past 30) and who took more puffs per day were more likely to be recruited via Twitter than panel (O.R.=1.05, p<.05; O.R.=1.04, p<.001). People who reported using Twitter more frequently were more likely to be recruited via Twitter than panel (O.R.=3.25, p<.001).

Conclusions: Our findings demonstrate that while equally effective in terms of recruitment rates, Twitter and online panels provide access to different subgroups of a hard-to-reach population (e-cigarette users). Twitter provided access to younger adults, who were heavier users of e-cigarettes and Twitter. Recruiting via social media along with online panel offered access to a broader population from which to understand e-cigarette use and online marketing than a single source.