The Berkeley authoritarianism study recruited almost exclusively from formal organizational members

Evidence

While the authors of the Berkeley study guessed that their findings could be generalized to the population of white, non-Jewish, native-born, middle-class Americans they recognized that the sample of persons actually studied was not a representative or random sample of this population or of any other specifiable population. To mention only one restriction, the subjects were almost all members of at least one formal organization since the major method of recruiting subjects was through such organizations. It is known that people who belong to at least one formal organization are in very many respects different from people who belong to no organizations (Christie, 1954). The authors of the Berkeley study took the position that sampling considerations were not vital to their work because they were not interested in estimating the incidence of certain attitudes but rather in establishing relationships among attitudes. — John T. Jost and Jim Sidanius, Political Psychology, 576

Categorization

• XREF: Relates to bias in sampling and selection effects — connects to any literature on non-representative research samples, e.g. social capital work on organizational membership differences (Coleman). — connects to: Organizational screening, not just disposition, filters out mentally ill recruits — John T. Jost and Jim Sidanius, Political Psychology: Key Readings, 4090