Date of Graduation
Doctor of Education (Ed.D.)
Learning and Instruction
The MTMM approach was used to evaluate construct validity of three nursing leadership traits (prioritization, delegation, and patient care management) across four assessment methods (multiple-choice tests--MCT, oral questioning-- OQ, high fidelity--HFS, and low fidelity simulation--LFS). Using a correlational descriptive design a 21-item MCT exam, a 21-item oral question instrument, a patient care HFS, and three LFS stations were embedded into a two hour objectively structured clinical examination (OSCE) assessment environment whose aim was to compare traditional assessment methods (MCT and OQ) in nursing education to burgeoning assessment methods (HFS and LFS). Generated scores from 137 senior-level baccalaureate nursing students at a private university located in Northern California were correlated with scores from standardized instruments measuring cognitive abilities (TEAS®) and scores from another outside instrument measuring dimensions of nursing leadership (Kaplan® RN Predictor Exam) to these OSCE scores. Further, a cost comparison and analysis for designing and implementing an OSCE assessment including high- and low-fidelity simulation was compared to the projected costs of similar OSCEs found in the literature.
Results concluded that all four criteria for construct validity were not uniformly met. Yet, the method reliability estimates were high across all four measurement methods. Correlation comparison of items from the TEAS® Exam and items from the Kaplan® RN Predictor Exam indicated that one subtest from each external exam correlated highly with trait items across all methods. Lastly, the estimated budget for the OSCE assessment study was considerably less than two estimates for similar OSCEs in the literature with the actual OSCE cost ($28,022.04) being significantly less expensive than the estimates found in the literature though higher than the estimated budget.
Embry, Toby, "An Evaluation of a Nursing Leadership Simulation Experience Using Multitrait Multimethod Matrix" (2013). Doctoral Dissertations. 65.