Pattern reliability of narcotics addicts' self-reported data: a confirmatory assessment of construct validity and consistency.

Pattern reliability, or the invariance of relationships among variables, was investigated in this study. The consistency of theoretical constructs reflected by measures taken at two separate occasions can be tested using confirmatory factor analysis. Self-report data were obtained from 323 narcotics addicts in two face-to-face interviews conducted in 1974/75 and 1985/86. The two interviews overlapped approximately 4 years between 1970 and 1974/75. Through the testing of the invariance of measurement and structural models, pattern reliability was confirmed in one of the models developed. Explication of pattern reliability offers an alternative means of assessing validity of self-report data.