Deception and Shopping Behavior Among Current Cigarette Smokers: A Web-Based, Randomized Virtual Shopping Experiment

BACKGROUND: Virtual stores can be used to identify influences on consumer shopping behavior. Deception is one technique that may be used to attempt to increase the realism of virtual stores. OBJECTIVE: The objective of the experiment was to test whether the purchasing behavior of participants in a virtual shopping task varied based on whether they were told that they would receive the products they selected in a virtual convenience store (a form of deception) or not. METHODS: We recruited a US national sample of 402 adult current smokers by email from an online panel of survey participants. They completed a fully automated randomized virtual shopping experiment with a US $15 or US $20 budget in a Web-based virtual convenience store. We told a random half of participants that they would receive the products they chose in the virtual store or the cash equivalent (intervention condition), and the other random half simply to conduct a shopping task (control condition). We tested for differences in demographics, tobacco use behaviors, and in-store purchases (outcome variable, assessed by questionnaire) by experimental condition. RESULTS: The characteristics of the participants (398/402, 99.0% with complete data) were comparable across conditions except that the intervention group contained slightly more female participants (103/197, 52.3%) than the control group (84/201, 41.8%; P=.04). We did not find any other significant differences in any other demographic variables or tobacco use, or in virtual store shopping behaviors, including purchasing any tobacco (P=.44); purchasing cigarettes (P=.16), e-cigarettes (P=.54), cigars (P=.98), or smokeless tobacco (P=.72); amount spent overall (P=.63) or on tobacco (P=.66); percentage of budget spent overall (P=.84) or on tobacco (P=.74); number of total items (P=.64) and tobacco items purchased (P=.54); or total time spent in the store (P=.07). CONCLUSIONS: We found that telling participants that they will receive the products they select in a virtual store did not influence their purchases. This finding suggests that deception may not affect consumer behavior and, as a result, may not be necessary in virtual shopping experiments.