PSYCHOMETRIC EVALUATION OF A SHORT FORM OF THE INVENTORY OF DRINKING SITUATIONS

The Inventory of Drinking Situations (IDS) was developed to assess the likelihood of alcoholic relapse in eight high-risk situations. Previous research questioned the factor stability of the instrument; therefore, an alternative, short form of the IDS (IDS42) was developed and subjected to preliminary examination. This project reports on the further examination of the psychometric characteristics of the IDS42. The instrument's primary factor stability and higher-order factor structure were examined, and subscale means, standard deviations, alpha coefficients, interscale correlations and independence (unique variance) were determined. Also, the relationships between the IDS42 subscale scores and alcohol dependence, social desirability and demographic characteristics were explored. Results of the principal components analysis support the instrument's primary factor stability and indicate that the scales load onto one higher order dimension. The subscale internal consistency estimates were very high. Significantly high positive correlations were found between the IDS42 subscales and alcohol dependence. The IDS42 is influenced by a socially desirable response set that may result in spuriously low IDS42 scale elevations. The subscales initially demonstrated high interscale correlations; however, estimates of unique variance show that the IDS42 scales have an acceptable level of independence. Except for a significant negative correlation with age for three subscales, demographic characteristics do not appear to influence IDS42 results. These findings support the use of the IDS42 in assessing the extent to which heavy alcohol use has occurred in five situations. With this information, it may be possible to prepare to avoid or cope with these situations in the future and, consequently, to reduce the potential for relapse. However, additional evaluations, especially with a variety of populations, need to be completed.