Invitation choice structure has no impact on attendance in a female business training program in Kenya

Business training programs are a common form of support to small businesses, but organizations providing this training often struggle to get business owners to attend. We evaluate the role of invitation choice structure in determining agreement to participate and actual attendance. A field experiment randomly assigned female small business owners in Kenya (N = 1172) to one of three invitation types: a standard opt-in invitation; an active choice invitation where business owners had to explicitly say yes or no to the invitation; and an enhanced active choice invitation which highlighted the costs of saying no. We find no statistically significant effect of these alternative choice structures on willingness to participate in training, attending at least one day, and completing the course. The 95 percent confidence interval for the active treatment effect on attendance is [-1.9%, +9.5%], while for the enhanced active choice treatment it is [-4.1%, +7.7%]. The effect sizes consistent with our data are smaller than impacts measured in health and retirement savings studies in the United States. We examine several potential explanations for the lack of effect in a developing country setting. We find evidence consistent with two potential reasons being limited decision-making power amongst some women, and lower levels of cognition making the enhanced active choice wording less effective.