Generalizability analyses of NBDE Part II

This research applied generalizability theory to assess the effect of varying the number of cases and items nested within cases on generalizability of scores on Part II of the National Board Dental Examinations (NBDE Part II). In this research, sources of error were defined. Measurement conditions were classified. Error variances and generalizability coefficients for different conditions were computed. The data analyzed were the item responses of 1,535 candidates enrolled in accredited dental education programs who all took the same test form in 2007. Results showed that using more cases of fewer items might lead to a greater increase in generalizability than using more items per case. Other practical considerations such as time and cost constraints must be taken into account when applying the results of this research in other testing situations.

The reliability of the student version of the Jenkins Activity Survey

Two studies assessed two types of reliability of the student Jenkins Activity Survey (JAS; a questionnaire measure of Type A behavior). In Study 1, 810 undergraduates completed the JAS. Analyses revealed that estimates of the internal consistency (Chronbach's alpha) of the JAS and its subscales (hard-driving/competitive and speed impatience) ranged from moderately low to moderately high. In Study 2, independent samples of undergraduates completed the JAS at two testings, separated by 2 weeks (N = 137) or 3 months (N = 124). Analyses revealed that estimates of the temporal stability (test-retest reliability) of the JAS total score ranged from moderately to extremely high. The results of these studies suggest that the student JAS has satisfactory internal consistency and excellent test-retest reliability and support its continued use in research on these psychometric grounds. Additional potential difficulties with the JAS are discussed in the context of recommendations for future research.

Measurement of negative symptoms

The issue of the dimensionality of negative symptoms is addressed. In reference to data reported by Lewine, Fogg, and Meltzer (1983), it is suggested that the nonsignificant correlation between a SADS-C negative symptom scale and a NOSIE negative symptom scale, both independently developed using the Rasch model, is evidence for the multidimensionality of negative symptoms. Additional data supporting the multidimensionality of negative symptoms using confirmatory factor analysis is also discussed (Gibbons et al. 1985). A new theory for the structure of negative symptoms is needed, which specifies the number of dimensions, their classification, and their pattern of intercorrelation.