Invariance of measures of prevention program effectiveness: A replication

Citation
Ab. Rosay et al., Invariance of measures of prevention program effectiveness: A replication, J QUANT CR, 16(3), 2000, pp. 341-367
Citations number
59
Categorie Soggetti
Social Work & Social Policy
Journal title
JOURNAL OF QUANTITATIVE CRIMINOLOGY
ISSN journal
07484518 → ACNP
Volume
16
Issue
3
Year of publication
2000
Pages
341 - 367
Database
ISI
SICI code
0748-4518(200009)16:3<341:IOMOPP>2.0.ZU;2-H
Abstract
Recent literature has suggested that measures of risk and protective factor s for delinquency and substance use are not equally reliable or valid acros s gender and ethnic groups and has recommended differentiated programming a nd culturally specific evaluation methods. Three data sets containing up to five ethnic groups were used to determine the degree to which risk and pro tective factors are equally reliable and valid predictors of drug use and d elinquency across gender and ethnic groups. Congeneric measurement models a nd structural equation models were evaluated to determine if the factor str uctures for these measures and their covariances with measures of drug use and delinquency were equivalent across gender and ethnic groups. Half of th e risk and protective factors included in this analysis were found to be eq ually reliable across gender and ethnic groups. When controlling for reliab ility differences, all of the risk and protective factors were found to pre dict both drug use and delinquncy for all gender and ethnic groups. In term s of the magnitude of these associations, no substantive differences were f ound in the validity of risk and protective factors for drug use and delinq uency. Differences in the validity of risk and protective factors were more prevalent for delinquency than for drug use. However, all differences were substantively trivial. We conclude that measures of prevention program eff ectiveness are invariant across gender and ethnic groups.