Factor structure of the subtle screening inventory-2 (SASSI-2) /
The Substance Abuse Subtle Screening Inventory is a widely used instrument designed to identify individuals with substance abuse difficulties. Unfortunately, despite broad application, there have been very few validation studies of the SASSI. The present work therefore undertook a factor-analytic ap...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1999.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=730319101&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | The Substance Abuse Subtle Screening Inventory is a widely used instrument designed to identify individuals with substance abuse difficulties. Unfortunately, despite broad application, there have been very few validation studies of the SASSI. The present work therefore undertook a factor-analytic approach to examine the validity of the SASSI in two samples. The two samples were composed predominantly White males who had become involved in the criminal justice system because of alcohol or other drugs. The second edition (SASSI-was the version administered to both samples; the data were easily generalized to the most recent edition (SASSI-3) because the scale composition is essentially the same between the two versions. Internal consistency reliability coefficients for each sample on the 10 scales defined by the SASSI'S developers were comparable to those presented in other studies. Alpha values for the scales composed of face valid items (FVA, FOVD, and SYM) ranged upwards of .80. Those for the empirically defined scales (OAT, DEF, COR, and especially SAT, FAM, SAM, and RAP) were below .80, most often below .60. A confirmatory factor analysis revealed a very poor fit between the data from the larger sample and a model of uncorrelated factors. In addition, exploratory factor analyses yielded two factors that corresponded to the SASSI'S FVA and FVOD scales and that displayed good reliability for these sets of scores. However, the remaining eight factors did not correspond at all to the other SASSI scales, and also produced markedly lower internal consistency reliability coefficients. An alternative two-scale model was developed using face valid items from the FVA and FVOD scales together with a few items from the SYM scale. This model was replicated using the second sample, remained consistent across gender, and appeared not to be moderated by age of the respondent. It is concluded that a direct approach to identification substance abusers is more effective than an indirect approach, which is consistent with other literature pertaining to screening for such disorders. |
|---|---|
| Item Description: | Vita lacking. "Major Subject: Counseling Psychology". |
| Physical Description: | xii, 199 leaves : illustrations ; 28 cm. Issued also on microfiche from University Microfilm Inc. |
| Bibliography: | Includes bibliographical references (leaves 147-173). |