The Rosenberg Self-Esteem Scale, a widely used self-report instrument for evaluating individual self-esteem, was investigated using item response theory. Factor analysis identified a single common factor, contrary to some previous studies that extracted separate Self-Confidence and Self-Depreciation factors. A unidimensional model for graded item responses was fit to the data. A model that constrained the 10 items to equal discrimination was contrasted with a model allowing the discriminations to be estimated freely. The test of significance indicated that the unconstrained model better fit the data-that is, the 10 items of the Rosenberg Self-Esteem Scale are not equally discriminating and are differentially related to self-esteem. The pattern of functioning of the items was examined with respect to their content, and observations are offered with implications for validating and developing future personality instruments.

The CAHPS 1.0 survey instrument appears to have excellent psychometric properties.

Item response theory (IRT) provides procedures for scoring tests including any combination of rated constructedresponse and keyed multiple-choice items, in that each response pattern is associated with some modal or expected a posteriori estimate of trait level. However, various considerations that frequently arise in large-scale testing make response-pattern scoring an undesirable solution. Methods are described based on IRT that provide scaled scores, or estimates of trait level, for each summed score for rated responses, or for combinations of rated responses and multiple-choice items. These methods may be used to combine the useful scale properties of IR'r-based scores with the practical virtues of a scale based on a summed score for each examinee. Index terms: graded response model, item response theory, ordered responses, polytomous models, scaled scores.

BackgroundFast-acting medications for the management of anxiety are important to patients and society. Measuring early onset, however, requires a sensitive and clinically responsive tool. This study evaluates the psychometric properties of a patient-reported Global Anxiety - Visual Analog Scale (GA-VAS).MethodsData from a double-blind, randomized, placebo-controlled study of lorazepam and paroxetine in patients with Generalized Anxiety Disorder were analyzed to assess the reliability, validity, responsiveness, and utility of the GA-VAS. The GA-VAS was completed at clinic visits and at home during the first week of treatment. Targeted psychometric analyses—test-retest reliabilities, validity correlations, responsiveness statistics, and minimum important differences—were conducted.ResultsThe GA-VAS correlates well with other anxiety measures, at Week 4, r = 0.60 (p < 0.0001) with the Hamilton Rating Scale for Anxiety and r = 0.74 (p < 0.0001) with the Hospital Anxiety and Depression Scale - Anxiety subscale. In terms of convergent and divergent validity, the GA-VAS correlated -0.54 (p < 0.0001), -0.48 (p < 0.0001), and -0.68 (p < 0.0001) with the SF-36 Emotional Role, Social Function, and Mental Health subscales, respectively, but correlated much lower with the SF-36 physical functioning subscales. Preliminary minimum important difference estimates cluster between 10 and 15 mm.ConclusionsThe GA-VAS is capable of validly and effectively capturing a reduction in anxiety as quickly as 24 hours post-dose.