Continuous and binary sets of responses differ in the field

Abstract

This paper conducts a pre-registered study aimed to compare binary and continuous set of responses in survey questionnaires. Binary responses consist of two possible opposing response options (Yes/No). Continuous responses are numerical, where respondents can indicate an option on a 0–10 horizontal blind line. We study whether feasible sets of binary and continuous responses yield the same outcome (distribution) and have the same cost (duration in minutes). We collect data from 360 households in Honduras that were randomly assigned to Yes/No questions or given a slider (0–10 visual scale) to mark their responses, therefore, we provide causal evidence. We find that respondents are 13% more likely to respond “Yes” and spend 2.1 min less in the binary setting. Additionally, the results suggest that the type of question matters.