Design and participants. We used a 2 x 2 factorial design in which we manipulated recommendations (referral, no referral) by exposing the treatment group to a situation in which participants had to articulate a recommendation; the referral reward was one of two levels (small reward, large reward). Thus, we experimentally created four groups.
For the data collection, we recruited participants using an online panel. The invitation to participate in an academic study directed any interested respondents to a website containing the experiment. In total, 234 participants took part in the study and were randomly assigned to one of the four groups. The mean age was 40.7 years, and 47.4% of the respondents were women. We chose a scenario design, a popular method in marketing literature (Palmatier et al. 2009; Wagner, Hennig-Thurau, and Rudolph 2009), and used postpaid cellular telecommunication services as our setting. In the sample population, familiarity with the specific service category is high: 96.2% of the participants owned a cellular phone, and of these, 61.6% used a contract based on a monthly rate (postpaid), whereas 38.4% used prepaid phones.