Improving the Quality and Lowering Costs of Household Survey Data Using Personal Digital Assistants (PDAs): An Application for Costa Rica

Luis Rosero-Bixby, Universidad de Costa Rica
Jeisson Hidalgo, Universidad de Costa Rica
Daniel Antich, Universidad de Costa Rica

Population data in developing countries comes mostly from household visits. Armies of interviewers collect those data using paper and pencil (the PAPI method). But that information is infested with nonrandom errors. Newly developed “computer aided interviewing” techniques (CAPI for personal interviews) are an important advance over PAPI in terms of reducing errors and shortening time lags for data availability. But CAPI has proved to be difficult to implement in developing countries because of the ergonomics, costs, and delinquency restrictions of laptops. In response, we developed a CAPI system for PDAs that may revolutionize the way data is gathered, by reducing fraud, errors and costs and by opening opportunities for collecting complex information. At the core of our PDA-CAPI is an “electronic questionnaire markup language” (EQML) we developed after analyzing thousands of existing questions. We are using a PDA-CAPI prototype in our complex longitudinal study of elderly Costa Ricans.