Survey Practice in Non-Survey-Literate Populations: Lessons Learned from a Cognitive Interview Study in Brazil.

Abstract

Survey research relies on cooperation and coordination between researchers and respondents. Survey-literate respondents possess a level of understanding of the survey process that facilitates their participation. Non-survey-literate respondents, on the other hand, are less equipped to provide accurate responses, which can lead to increased survey error. In this article, we present findings from a cognitive interview project conducted in Brazil to illustrate potential barriers to respondent participation and demonstrate how these barriers contribute to response error. To test questions on inclusive education, researchers from the Collaborating Center for Questionnaire Design and Evaluation Research at the National Center for Health Statistics facilitated the collection of data through 80 cognitive interviews conducted in various neighborhoods in Rio de Janeiro, Brazil. Recruitment targeted caregivers of children with and without disabilities, and respondents were predominantly female with low literacy levels. The results indicated that respondents with limited familiarity with the survey process (non-survey-literate) struggled to orient themselves to the survey task. They faced challenges in choosing response options, understanding scale relationships, and interpreting vocabulary. Additionally, many respondents expressed a need to share salient details of their lives, which the survey was not designed to capture. Understanding these barriers to participation and identifying ways to mitigate them can help reduce survey error, particularly in vulnerable populations.