Perspectives and implications of the Improving Coverage Measurement Core Group's validation studies for household surveys.

BACKGROUND: Formal validation studies are of critical importance in determining whether or not household survey questions are providing accurate information on what they intend to measure. These studies supplement an array of methods used to evaluate survey questions. METHODS AND FINDINGS: This paper summarizes the methods used by the two major international household survey programmes - The Demographic and Health Surveys Program (DHS) and the Multiple Indicator Cluster Surveys (MICS) - to decide on possible modifications to the survey questions, nomenclature, tables, and interpretation of findings over time as additional information on the validity of the questions becomes available. CONCLUSIONS: Validation studies are most useful if they are conducted in a variety of different settings in low- and middle-income countries, preferably using representative samples and procedures that replicate DHS and MICS field conditions. Pilot tests, pre-tests in each country, feedback from interviewers and survey staff, and cognitive interviewing provide additional information about how well survey questions are understood and provide accurate information. The paper provides specific examples of changes that have been made in response to findings from validation studies and changes in international recommendations.