HARMONIZATION OF DATA COLLECTION ACROSS 4 COUNTRIES - THE ENRICH EXPERIENCE

Background and objectives: Undernutrition remains a major contributor to maternal and childhood morbidity and mortality. Enhancing Nutrition Services to Improve Maternal and Child Health in Africa and Asia (ENRICH) is a multi-country, broad-spectrum and integrated intervention program designed to help member countries achieve their global nutrition targets. Methods: As part of ENRICH impact evaluation, a baseline assessment was conducted via household surveys (HHS) using a quasi-experimental design. In three of the four countries (Bangladesh, Kenya, Tanzania), a comparison area likely to have baseline characteristics similar to the intervention area was selected. In Pakistan, no comparison area was planned due to contextual reasons. Results: The survey questionnaires were developed by University of Toronto (UofT) in coordination with World Vision Canada (WVC), ensuring consistency in tools, variables and response sets across settings. In-country research partners (CRPs) in each of the four countries carried out local contextualization and translation of the questionnaires. Data collection at all sites was conducted via Open Data Kit (ODK) platform on android tablets. English and predominant local language ODK templates for field-based entry of HHS were designed in Excel, converted into XML format, and uploaded to country-specific secure ODK Aggregate servers. Blank ODK forms were then accessed by field teams on their android tablets to be filled in during data collection in each country. Completed forms were saved on the tablets until uploaded to the ODK Aggregate server at the end of each working day. Multiple challenges were encountered and met across settings. Conclusions: ODK is an effective and efficient platform for collecting large amounts of data across different geographical sites in a relatively short period of time under a range of field challenges. Data thus collected were of high quality, with minimal effort required for data cleaning and management. Using ODK also allowed for rapid, real-time updates to data collection tools based on feedback from enumerator training and first few days of data collection in each country. Strong training, supervision and real-time international coordination was key to maximizing the potential and flexibility of ODK platforms for complex and large scale evaluations in low-income settings.