Stage 3: Item Review and Finalization Sample Clauses

Stage 3: Item Review and Finalization. During this stage, four rounds of item review took place to assess content validity, face validity, and translation. Items for the survey were then finalized. First, draft items were sent to two peer-reviewers with experience researching women’s sanitation in India to assess content validity[33]. The peer-reviewers provided recommendations for revising the wording of items related to experience and concern for harm. Second, to further assess content validity the two research assistants (RAs) who carried out the qualitative data collection then reviewed the items and provided comments, with particular attention to alternative phrasing for existing items to be more specific (for example, they suggested asking about a concern for infection as opposed to a concern about health in general). They also suggested including an item about needing to go back and forth to a defecation location because of a lack of privacy. The two RAs then translated the items from English to Oriya independently and then compared translations to reconcile any discrepancies and create a single translation. Third, the RAs reviewed each translated item with the nine Oriya-speaking female data collectors (DCs) hired to administer the survey to assess face validity[33]. The DCs were from the region where the data collection was to take place and had experience with sanitation-related surveys from previous research. They were able to speak both from their own experience of sanitation and their perception of the experiences of their fellow community members. Specifically, we used cognitive interviewing methods to determine if the DCs understood the items as we intended them to be understood, asking them to explain, in their own words, what each item meant [34]. Modifications to the translations were made, as needed. As an additional check to face validity, the research team (the RAs and the nine female DCs) piloted the items in a community similar to those where the data collection was to take place. During the pilot, the team noted items that were confusing to participants and wording that would be better. The team met after the pilot to discuss feedback and amend the item translations one final time. The final survey included 68 items, 32 for urination and 36 for defecation, with four possible response options: never, sometimes, often and always. Items covered three hypothesized domains: the physical environment (i.e. had difficulty finding a clean place