Informing and Mobilizing Voters by Texting: Evidence from a Large-Scale Experiment in Kenya

This project explores the role of voter’s information in a new institutional context. The setting of the 2013 Kenyan elections differed from the widely contested 2007 elections on several dimensions, including a new electoral commission and a new set of six offices to elect. Researchers report the results of a countrywide randomized field experiment: 2,461,099 phone holders registered in 6,074 polling stations were randomly chosen out of a total of 4,908,975 phone holders and 12,160 polling stations to receive text messages in the last week before the elections. One third of the phone holders received general encouragements to vote. The second third received specific information about and encouragements to vote for each of the six elected positions. The last third received specific information about the work conducted by the electoral commission to ensure free and fair elections. The project will compare voter turnout and vote shares across groups to isolate the impact of information from the impact of mobilization.