Key points are not available for this paper at this time.
BACKGROUND: With improvements in technology, electronic data capture (EDC) for large surveys is feasible. EDC offers benefits over traditional paper-based data collection, including more accurate data, greater completeness of data, and decreased data cleaning burden. METHODS: The second Kenya AIDS Indicator Survey (KAIS 2012) was a population-based survey of persons aged 18 months to 64 years. A software application was designed to capture the interview, specimen collection, and home-based testing and counseling data. The application included: interview translations for local languages; options for single, multiple, and fill-in responses; and automated participant eligibility determination. Data quality checks were programmed to automate skip patterns and prohibit outlier responses. A data sharing architecture was developed to transmit the data in real-time from the field to a central server over a virtual private network. RESULTS: KAIS 2012 was conducted between October 2012 and February 2013. Overall, 68,202 records for the interviews, specimen collection, and home-based testing and counseling were entered into the application. Challenges arose during implementation, including poor connectivity and a systems malfunction that created duplicate records, which prevented timely data transmission to the central server. Data cleaning was minimal given the data quality control measures. CONCLUSIONS: KAIS 2012 demonstrated the feasibility of using EDC in a population-based survey. The benefits of EDC were apparent in data quality and minimal time needed for data cleaning. Several important lessons were learned, such as the time and monetary investment required before survey implementation, the importance of continuous application testing, and contingency plans for data transmission due to connectivity challenges.
Ojwang' et al. (Mon,) studied this question.