Replication Data for “Eliciting Probabilities in the Field: Beads vs Sliders”

This dataset contains the data used in the paper “Eliciting Probabilities in the Field: Beads vs Sliders”. The study examines whether different visual aids used to elicit probabilistic expectations—namely a discrete method based on beads (or beans) and a continuous cardboard-based slider—affect the level, dispersion, accuracy, and internal consistency of reported probabilities. The data come from four field survey experiments conducted between 2018 and 2019 in three countries: Nigeria (two independent studies), Honduras, and Spain. Participants were randomly assigned to report probabilities using either the Beads or the Slider method. The surveys elicited both objective probabilities (based on known frequencies) and subjective probabilities related to everyday events, as well as information to assess compliance with basic probability properties such as nested events. The dataset includes individual-level responses to all probability elicitation questions (Q1–Q6), indicators of the visual aid used, socio-demographic characteristics (age, gender, and education), survey identifiers, and timing measures when available. All variables are anonymized, and no personally identifiable information is included. This repository also includes the do-files and scripts required to reproduce all tables and figures reported in the paper.