The Revised Danish Learning Styles Inventory (R-D-LSI) (Nielsen 2005), which is an adaptation of Sternberg-Wagner Thinking Styles Inventory (Sternberg, 1997), comprises 14 subscales, each measuring a separate learning style. Of these 14 subscales, 9 are eight items long and 5 are seven items long. For self-assessment, self-scoring and…

Developed a method for conditional pairwise estimation in the Rasch model for ordered response categories using principal components analysis and evaluated the procedure through a simulation study. Results show that the quality of the estimates is excellent. (SLD)

Proposes formulas for expected true-score measures and reliability of binary items as a function of their Rasch difficulty when the trait (ability) distribution is normal or logistic. Provides an illustrative example for using the proposed formulas. (SLD)

Developed a procedure for evaluating item-level nonresponse bias in questionnaire items using logistic regression to determine whether nonresponses are random or systematic in nature for one question from the National Education Longitudinal Study of 1994 concerning drug use behaviors. Identified systematic nonresponses and the magnitude of…

Examined the sensitivity to change over time of the Amytrophic Lateral Sclerosis Assessment Questionnaire (ALSAQ-40; C. Jenkinson and others, 1999). Findings for 764 individuals show that the ALSAQ-40 is a valid and responsive disease-specific health-related quality of life instrument for studies of patients with motor neuron diseases. (SLD)

Developed school versus benchmark comparison reports based on groups of easily comparable school types for Queensland, Australia schools. The state benchmark for each set of comparable schools estimated how easy it was, on average, for members of the benchmark group (parents or students) to endorse each of the School Opinion Survey Likert-scale…

Attempted to illuminate practical limitations on the Rasch model by focusing on the recovery of the density scale through five simulation trials. Results show that when error distributions are insufficient, the results may be ordinal at best, and when error distributions are nonsymmetrical, the positions of items may be biased with respect to the…

Discusses substantive scale construction and the need for Rasch measurement practitioners to give careful consideration to designing variables based on theory, item construction, and models for the analysis of data. (SLD)

Studied four different approaches for handling missing data for their capacity to mitigate against the effect of omitted responses on person location estimation. Results from a Monte Carlo study show that the hot-decking procedure performed best of the methods examined. (SLD)

Compared the observation of children's growth in reading using grade equivalents with that using the Rasch simultaneous vertical equating procedure. Results from equating 25 reading tests show that grade equivalents differ more between two different test forms than do ability measures obtained from the Rasch simultaneous vertical equating. (SLD)

Examined the effectiveness of the Sympson-Hetter technique and rotated content balancing relative to no exposure control and no content rotation conditions in a computerized adaptive testing system based on the partial credit model. Simulation results show the Sympson-Hetter technique can be used with minimal impact on measurement precision,…

Developed an instrument to measure the aerobic exercise equipment preference of a frail older population and applied many-facet Rasch analysis to study construct validity and equipment preferences. Results for 16 participants show the usefulness of many-facet Rasch analysis in guiding instrument revision. (SLD)

Analyzed the Frequency and Intensity subscales of the Clinician Administered Posttraumatic Stress Disorder Scale (PTSD) separately using a Rasch rating scale model. The Frequency subscale produced measures that encompassed the level of severity found in the sample of 201 veterans. (SLD)

Compared the performance of four polytomous item response theory (IRT) and three linear models for constructing adjusted grade-point-average (GPA) measures. Studies involving cohorts of 1,255 and 1,796 college students and an additional 1,823 and 1,879 college students. Discusses implications of findings for correlation of grade-based measures and…

Studied the usefulness of the Multidimensional Random Coefficient Multinomial Logit model using a collection of embedded assessments administered to 541 students in middle school science classes. Findings illustrate the usefulness and good fit of the multidimensional approach in contrast to a unidimensional approach. (SLD)