Capturing Voices: Comparing Audio and Typed Open-Ended Survey Formats

Typed open-ended responses remain the standard approach for probing how and why people arrive at their survey answers, yet they are not without drawbacks. They can deter participation and pose challenges for respondents with lower literacy or from harder-to-reach groups (Shaghaghi, Bhopal, & Sheikh, 2011; Stoop, 2005; Duvoisin et al., 2024). Collecting audio-recorded responses offers a potential alternative. In contexts where voice messaging is widespread, such as in Latin American countries, speaking may feel easier and more natural than typing, lowering cognitive and literacy demands while encouraging more spontaneous, detailed, and affectively rich answers. This raises an important methodological question: which format—typed or audio—better captures the reasoning behind responses while also supporting data quality and validity?