According to the weak local independence approach to defining dimensionality, the fundamental quantities for determining a test's dimensional structure are the covariances of item-pair responses conditioned on examinee trait level. This paper describes three dimensionality assessment procedures-HCA/CCPROX, DIMTEST, and DETECT—that use estimates of these conditional covariances. All three procedures are nonparametric ; that is, they do not depend on the functional form of the item response functions. These procedures are applied to a dimensionality study of the LSAT, which illustrates the capacity of the approaches to assess the lack of unidimensionality, identify groups of items manifesting approximate simple structure, determine the number of dominant dimensions, and measure the amount of multidimensionality. Index terms: approximate simple structure, conditional covariance, DETECT , dimensionality, DIMTEST, HCA/CCPROX, hierarchical cluster analysis, IRT, LSAT, local independence, multidimensionality, simple structure.

ABSTRACT. Objective: This study evaluated the psychometric properties of the 15-item alcohol Short Index of Problems (SIP) instrument and those of a newly constructed 15-item drug Short Index of Problems (SIP-D) instrument in 277 newly entered substance-abuse patients. Method: The SIP is derived from the longer, 50-item Drinker Inventory of Consequences (DrInC), which was designed to assess adverse consequences of alcohol use. The SIP-D was constructed by substituting the term "drug use" for the term "drinking" in each SIP item. A 3-month recall interval was employed. Results: Factor analyses of each of the instruments revealed similar solutions, with only one main factor accounting for the majority of variance. Nonparametric item response theory methods produced the same fi nding. Internal consistency reliability estimates for the SIP and SIP-D total scores were .98 and .97, respectively. Concurrent validity was demonstrated by examining the correlations of the total scores for each of the instruments with the recent summary indexes of the newly revised Addiction Severity Index (ASI-Version 6): alcohol, drug, medical, economic, legal, family/social, and psychiatric problems. (Blanchard et al., 2003). Indeed, research suggests that consumption levels alone are not necessarily good predictors of substance use-related impairment (Bender et al., 2007). Within this context, Miller and colleagues (1995) developed the Drinking Inventory of Consequences (DrInC) for alcohol-dependent patients to serve as a relatively brief, self-report measure of the severity of the consequences of alcohol use. Constructed by a panel of experts, the DrInC consists of 45 primary items in fi ve domains: Physical, Intrapersonal, Social Responsibility, Interpersonal, and Impulse Control consequences. The instrument can be administered using either a lifetime or a 3-month recall timeframe. The DrInC also includes an additional fi ve "control" items designed to detect careless responding or dishonesty. These items are not included in psychometric evaluations of the instrument. Although factor analysis has not supported fi ve independent domains (Anderson et al., 2001;Miller et al. 1995), these domains have been used in research in addition to the total score. The generally excellent internal consistency reliability and test-retest reliability of each of the subscales appear to have been accepted as indications of their independent integrity. Miller et al. (1995) also developed a short version of the DrInC labeled the Short Index of Problems (SIP). The SIP was derived by including the three items in each DrInC subscale that were most highly correlated with the total subscale score. Internal consistency of this new instrument was lower than for the full DrInC subscales, ranging from .57 to .66; that for the entire instrument was .81. Six-month test reliability was good for both the subscales and the total score. In another study that used the full DrInC (Feinn et al., 2003), the internal consistency of the embedded 3-month SIP subscales rang...

Aims To describe the derivation of Recent Status Scores (RSSs) for Version 6 of the Addiction Severity Index (ASI-6). Design 118 ASI-6 recent status items were subjected to nonparametric item response theory (NIRT) analyses followed by confirmatory factor analysis (CFA). Generalizability and concurrent validity of the derived scores were determined. Setting and Participants 607 recent admissions to variety of substance abuse treatment programs constituted the derivation sample; a subset (N = 254) comprised the validity sample. Measurements The ASI-6 interview and a validity battery of primarily self-report questionnaires that included at least one measure corresponding to each of the seven ASI domains were administered. Findings Nine summary scales describing recent status that achieved or approached both high scalability and reliability were derived; one scale for each of six areas (medical, employment/finances, alcohol, drug, legal, psychiatric), and three scales for the family/social area. Intercorrelations among the RSSs also supported the multidimensionality of the ASI-6. Concurrent validity analyses yielded strong evidence supporting the validity of the six of the RSSs (Medical, Alcohol, Drug, Employment, Family/Social Problems, Psychiatric). Evidence was weaker for the Legal, Family/Social Support and Child Problems RSSs. Generalizability analyses of the scales to males versus females and whites versus blacks supported the comparability of the findings with slight exceptions. Conclusions The psychometric analyses to derive Addiction Severity Index-6 Recent Status Scores (RSSs) support the multidimensionality of the ASI-6 (i.e., the relative independence of different life functioning areas), consistent with research on earlier editions of the instrument. In general, the ASI-6 scales demonstrate acceptable scalability, reliability and concurrent validity. While questions remain about the generalizability of some scales to population subgroups, the overall findings coupled with updated and more extensive content in the ASI-6 support its use in clinical practice and research.

This exploration is the first step in identifying how types and categories of features impact outcomes. While the findings are inconclusive due to lack of homogeneity, this provides a foundation for future feature analysis.

This Monte Carlo study compares the ability of the parametric bootstrap version of DIMTEST with three goodness-of-fit tests calculated from a fitted NOHARM model to detect violations of the assumption of unidimensionality in testing data. The effectiveness of the procedures was evaluated for different numbers of items, numbers of examinees, correlations between underlying ability dimensions, skewness of underlying ability distributions, and the presence or absence of a guessing parameter. In the absence of guessing, DIMTEST and the NOHARM-based statistics had similar power, with the χ2 statistic having a very low Type I error rate. In the presence of guessing, however, two of the NOHARM-based statistics had unacceptably high Type I error rates, while the third performed similarly to DIMTEST. Given this inflated error rate, the study compares the empirical powers after adjusting for the discrepancy in Type I error rates.