Competencies in Geriatric Nursing: Empirical Evidence from a Computer-Based Large-Scale Assessment Calibration Study

Abstract

Valid and reliable standardized assessment of nursing competencies is needed to monitor the quality of vocational education and training (VET) in nursing and evaluate learning outcomes for care work trainees with increasingly heterogeneous learning backgrounds. To date, however, the modeling of professional competencies has not yet evolved into procedures that would meet large-scale assessment (LSA) standards in VET. To empirically test a proposed structural model for client-directed nursing competence and to estimate psychometric properties of a newly developed video- and computer-based test (CBT) to inform subsequent LSA in nursing VET, 402 final-year nursing students from 24 German schools responded to a 77 item CBT. Multi-dimensional IRT modeling was employed to test the subdomain structure and estimate students’ competencies in geriatric nursing. The standardized CBT measures nursing students’ client-directed care competence with acceptable precision (WLE = 0.76) and does so across the whole range of observed proficiency levels. Structural validity was supported by substantive contributions of test items from all proposed process-oriented subdomains, practice field scenarios, as well as items with and without reference to emotional demands. However, it was not possible to empirically separate the diagnostic, practical or communicative subdomains, probably reflecting parallel, recursive and hierarchical care processes in complex care situations. On average, students in our sample attained 45 % of the maximum test score so it is a demanding assessment of nursing competence. An extensively piloted, valid and reliable CBT is suggested to assess nursing students’ client-directed care competencies at the end of the third year of the VET program.