Sample size (or number of clusters) by treatment arms
Individuals in our experiment will be assigned to a video message arm with 60 percent probability. Remaining 40 percent will receive no video messages. Those who receive a video message will be assigned to a health, success or a (placebo) control message arm with equal probabilities. Therefore, each of these subgroups will constitute 20% of the total participant pool.
We hereby, pre-specify the total number of participants and the duration of the experiment.
Expected Participants: We aim to engage about 15,000 individuals in our study. Sixty percent of these will be randomly assigned to one of the three video message groups as outlined above. Based on our preliminary data, we estimate that we will be able to monitor approximately 35% of these participants during the three-week experimental period.
Experimental Duration: The study will run for three weeks, during which time the video messages will be sent out. Following the last video message, we intend to collect data for at least an additional three weeks to assess the potentially persistent effects of the messages. If funding allows, we may extend the data collection period beyond this period to measure medium to long-term outcomes.
Please note that our participant pool consists of adults who: i) have been observed at least once in our partner grocery stores, and ii) use WhatsApp. Our aim is to monitor individuals across a network of stores within the same community, which is why we have installed our devices in these locations. Unfortunately, due to high setup costs, we were unable to conduct a pilot study in a different community. Additionally, inviting individuals from our current network of stores to participate in a pilot study would directly reduce the sample size available for the main study.