Related Experiment Videos
Pattern reliability of narcotics addicts' self-reported data: a confirmatory assessment of construct validity and
C P Chou1, Y I Hser, M D Anglin
1Department of Preventive Medicine, University of Southern California, Los Angeles 90033, USA.
Substance Use & Misuse
|July 1, 1996
Summary
This study assessed pattern reliability, the consistency of relationships between variables, using self-report data from 323 narcotics addicts. Pattern reliability was confirmed in one model, offering a new way to evaluate self-report data validity.
Area of Science:
- Psychology
- Psychometrics
Background:
- Assessing the consistency of relationships among variables (pattern reliability) is crucial for research validity.
- Self-report data are widely used but require rigorous validation methods.
Purpose of the Study:
- To investigate pattern reliability in the context of self-report data from a longitudinal study of narcotics addicts.
- To explore the utility of confirmatory factor analysis in testing the invariance of theoretical constructs over time.
Main Methods:
- Longitudinal self-report data were collected from 323 narcotics addicts across two interviews (1974/75 and 1985/86).
- Confirmatory factor analysis was employed to test the measurement and structural invariance of proposed models.
Main Results:
- Pattern reliability was confirmed for one of the measurement models tested.
- The findings suggest that pattern reliability can be a viable indicator of construct consistency over time.
Conclusions:
- Pattern reliability, assessed through model invariance, provides an alternative method for validating longitudinal self-report data.
- This approach enhances confidence in the stability of relationships among variables in psychological research.