Fee vs Fine

Description This dataset contains data from an online experiment designed to test whether economically equivalent penalties—fees (paid before taking) and fines (paid after taking)—influence prosocial behaviour differently. Participants played a modified dictator game in which they could take points from another participant. The dataset is provided in Excel format (Full-data.xlsx), along with a Stata do-file (submit.do) that reshapes, cleans, and analyses the data. Data Collection Platform: oTree Recruitment: Prolific Sample size: 201 participants Design: Each participant played 20 rounds: 10 in the control condition and 10 in one treatment condition (fee or fine). Order of blocks was randomised. Payment: 200 points = £1. One round was randomly selected for payment. Variables Identification session – Session number id – Participant ID treatment – Assigned treatment (1 = Fee, 2 = Fine) order – Order of blocks (0 = Control first, 1 = Treatment first) Decision Rounds For each round, participants made decisions in both control (c) and treatment (t) conditions. c1, t1, c2, t2, … – Tokens available and/or allocated across control and treatment rounds. takeX – Amount taken from the other participant in case X. Norm Elicitation Social norms were elicited after the taking task. Variables include empirical, normative, and responsibility measures at both extensive and intensive margins: eyX, etX – Empirical expectations (beliefs about what others do) nyX, ntX – Normative expectations (beliefs about what others think is appropriate) ryX, rtX – Responsibility measures casenormX – Case identifier for norm elicitation Demographics From survey responses: Sex – Gender Ethnicitysimplified – Simplified ethnicity category Countryofresidence – Participant’s country of residence Other order, session – Experimental setup metadata Stata Do-File (analysis.do) The .do file performs the following steps: Data Preparation Import raw Excel file Reshape from wide to long format (cases per participant) Declare panel data (xtset id) Variable Generation Rename variables for clarity (e.g., take for amount taken) Generate treatment dummies (treat) Construct demographic dummies (gender, race, nationality) Analysis Preparation Create extensive and intensive margin variables Generate expectation and norm measures Output Ready-to-analyse panel dataset for regression and statistical analysis