Translating Legal Empowerment into a Randomized Controlled Trial

Abstract

After two-and-a-half years of practicing an indirect legal empowerment strategy in their assigned communities, it was time to determine the impact the CJAs were making. According to the methodological structure of the RCT-based evaluation, the end-line survey mirrored the original baseline survey; it entailed resurveying all original 176 communities (88 treatment and 88 control) and each of the original 2,100 households spread out across five counties — Maryland, Grand Gedeh, Nimba, Lofa and Bong. Consequently, organizing and implementing an RCT of this size in rural Liberia was an expensive, labor-intensive enterprise with many moving parts and unforeseen variables. The CSAE survey, led by Bilal Siddiqi and Justin Sandefur, reasoned that the end-line survey would require roughly 25–30 enumerators (data collectors) who could complete the survey in just over a month. The survey instrument was the actual digital multiple-choice questionnaire which was uploaded to the handheld computers or Personal Digital Assistants (PDAs). The PDAs display an interactive digital interface which the enumerator reads aloud to the respondent. The survey instrument prompts the enumerator to ask the respondent a question and displays the optional answers for selection. The answers are then recorded on the PDA and saved for later analysis.