The validity of Niederhoffer's cynicism scale

Arthur Niederhoffer's (1967) Behind the Shield is widely regarded as a classic in the policing literature, yet problems associated with measurement of the key latent trait, cynicism, have limited the extent to which conclusions may be drawn from Niederhoffer's work, as well as some subsequent police cynicism research. In this article, Niederhoffer's research is revisited using survey data recently collected from a random sample of 499 Philadelphia police officers. The analysis begins by examining the validity of Regoli's (1976) modified cynicism scale from the perspective of Item Response Theory (IRT), using Rasch modeling techniques in an effort to more fully understand the scale's measurement properties. Then, Niederhoffer's primary research hypothesis is revisited. Three main findings are drawn: (1) the Likert response categories are being used by respondents as intended; (2) some of the scale items exhibit gender and race bias; and (3) the scale can be improved by dropping several items. Once the scale is adjusted, the findings indicate that the relationship between officer cynicism and years of service is slightly stronger than when the scale is used in its original form. Further, regression analyses yield theoretically consistent findings for the relationship between cynicism and one job-related measure (departmental disciplinary charges).