Reproduction Files for 'Should We Worry About Sponsorship-Induced Bias in Online Political Science Surveys?'

This repository contains file to reproduce the results for: > Leeper, Thomas J., and Emily Thorson. "Should We Worry About Sponsorship-Induced Bias in Online Political Science Surveys?" Journal of Experimental Political Science. - `analysis.do`: the main Stata data analysis file producing all results, including tables to a subdirectory called `/tables` which must be created by the user - `analysis_coding.do`: a file called by `analysis.do` that recodes the source data file - `sponsordata.dta`: a Stata data file (.dta) containing the original data - `log.txt`: the output of running `analysis.do` in Stata 14 - `tables.zip`: a zip directory containing the separate tabular output of `analysis.do` as latex (.tex) tables - Qualtrics QSF files to reproduce all experimental conditions: - `control.qsf` - `marketingheavy.qsf` - `marketinglight.qsf` - `universityheavy.qsf` - `universitylight.qsf` - Word documents containing full questionnaires (as exported from Qualtrics) - `control.docx` - `marketingheavy.docx` - `marketinglight.docx` - `universityheavy.docx` - `universitylight.docx` To reproduce results, the `analysis.do` file needs to be run in Stata. It uses mainly commands included in Stata 14 (or lower) with the addition of the `estout` package available via ssc.