10.7910/DVN/K6PRPGBoas, TaylorTaylorBoasBoston UniversityChristenson, DinoDinoChristensonBoston UniversityGlick, DavidDavidGlickBoston UniversityHarvard Dataverse2018Social Sciencesonline surveyssamplingUnited StatesIndiaBoas, TaylorTaylorBoasBoston UniversityEnter your name here: LastName, FirstNameFirstNameEnter your name here: LastName2018-02-122018-05-08985564741112783846417985829245355103276202941748810831239882234893954547243467212386551449600492054875075999270401446854145291471914628566038759656933804714409184254039710323988224290811529767894564789086623357569411472806388571620616099158852992105640068181016104type/x-r-syntaxtext/plaintype/x-r-syntaxtype/x-r-syntaxtype/x-r-syntaxtype/x-r-syntaxtype/x-r-syntaxtype/x-r-syntaxtype/x-r-syntaxtype/x-r-syntaxtext/tab-separated-valuesapplication/pdftext/tab-separated-valuesapplication/pdftext/tab-separated-valuestext/tab-separated-valuestext/tab-separated-valuesapplication/x-rlang-transportapplication/x-textext/tab-separated-valuestext/tab-separated-valuesapplication/pdfapplication/x-rlang-transportapplication/vnd.openxmlformats-officedocument.wordprocessingml.documentapplication/vnd.openxmlformats-officedocument.wordprocessingml.documentapplication/pdfapplication/x-rlang-transporttext/csvapplication/vnd.openxmlformats-officedocument.wordprocessingml.documenttext/tab-separated-valuesapplication/x-rlang-transporttext/tab-separated-valuesapplication/x-rlang-transporttext/plaintext/plaintext/htmltext/tab-separated-valuesapplication/pdftext/htmlapplication/pdfapplication/x-rlang-transportapplication/vnd.openxmlformats-officedocument.wordprocessingml.documentapplication/vnd.openxmlformats-officedocument.wordprocessingml.documentapplication/vnd.openxmlformats-officedocument.wordprocessingml.documenttext/tab-separated-valuesapplication/x-rlang-transporttext/tab-separated-valuestext/tab-separated-values1.0CC0 1.0This article examines online recruitment via Facebook, Mechanical Turk, and Qualtrics panels in India and the United States. It compares over 7,300 respondents---1,000 or more from each source and country---to nationally representative benchmarks in terms of demographics, political attitudes and knowledge, cooperation, and experimental replication. In the U.S., MTurk offers the cheapest and fastest recruitment, Qualtrics is most demographically and politically representative, and Facebook facilitates targeted sampling. The India samples look much less like the population, though Facebook offers broad geographical coverage. We find online convenience samples often provide valid inferences into how partisanship moderates treatment effects. Yet they are typically unrepresentative on such political variables, which has implications for the external validity of sample average treatment effects.