Abstract Researchers in the behavioral and social sciences, particularly psychology, are beginning to conduct laboratory experiments on the World Wide Web. The Web venue offers conveniences that are apparent to market researchers and academic investigators, who have used the Web extensively for survey research. Like traditional experimental researchers. Web experimenters must assure the reliability and inferential validity of their experiments to determine whether a causal relationship exists between independent and dependent variables. In research design and implementation, therefore, Web experimenters must avoid the many artifacts of the experimental situation that threaten internal validity, construct validity, and external validity. This article discusses the application of methodological techniques to social science Web experiments and suggests topics in library and information science most conducive to research on the Web.