More on the origins and methodology of the F-Scale studies

Levinson & Sanford, two American researchers at the University of California, Berkeley, conducted research which aimed to understand the roots of anti-Semitism- prejudice against Jewish people, which had became particularly prominent due to the policies of Nazi Germany in the 1930s. They pioneered an anti-Semitism scale (the ‘A-S scale’), based on the psychoanalytic theory that prejudice was based on repression of one’s own undesirable characteristics, and projection of these onto an external target.

They found a strong relationship between attitudes towards minorities including Jews and Blacks, as well as ‘superpatriotism’). It was concluded that all of these findings stem from a common syndrome of ethnocentrism. However, a limitation is that the research assumed that participants were white and non-Jewish, as attitudes towards whites were not assessed.
After five revisions, the researchers produced the ethnocentrism scale (the ‘E scale’), containing statements about Jews, blacks, other minorities, and superpatriotism.

With funding from the American Jewish Committee, the study was broadened in the late 1940s. Two Austrian researchers, psychologist Frenkel-Brunswick & sociologist Adorno, joined Levinson and Sanford to study authoritarianism among American workers, and together they wrote the 1950 book, ‘The Authoritarian Personality’.

The theory suggested that the strict, repressive parenting of the Austrian middle-classes should result in high levels of authoritarianism. However, Frenkel-Brunswick conducted extensive interview-based research and found that high F-scores were common among lower social classes too, weakening this aspect of the theory.

Cronbach (1946) pointed out the problem of response set with the F-Scale, where it is hard to distinguish between respondents who agree with the content of the statements, and those who would agree to almost any item. He argued that this is most likely to occur when items are ambiguous, and some items in the F-scale are deliberately written to allow for projection (e.g. “The wild sex life of Romans…”).

To tackle this problem, Bass (1955) attempted a reversal of some items (so for example ‘familiarity breeds contempt’ was changed to ‘familiarity does not breed contempt’) and concluded that three fourths of the reliable variance on the F-scale is due to acquiescence.

Bass, B.M. (1955). Authoritarianism or acquiescence? Journal of Abnormal and Social Psychology, 51, 616-623.

Cronbach,L.J. (1946). Response sets and test validity. Educational and Psychological Measurement, 6, 475-494.