An empirical analysis of a clinical assessment tool.
A tool to measure clinical performance was required by a college-based programme for undergraduate nursing students. Multiple tools currently exist, but after extensive literature review the researchers decided that no tool exactly fitted the requirements of this nursing course. The tool developed by Bondy (1983) was most aligned to our purposes and was used as the basis for the ongoing development of the Clinical Performance Assessment instrument. The CPA measures four dimensions of clinical performance against a 5-point rating scale. Each Likert point has specific criteria for the level relating to the dimensions. The clinical facilitator compares student behaviours against the performance levels of the dimensions for a variety of nursing procedures which are determined by the stage of the program the student has reached. Scores are given for the four dimensions and immediate student feedback is available. Measures were taken on six separate clinical occasions across 2 years on over 100 students in an attempt to determine validity and reliability indices. Relations between variables that appeared to have logical/theoretical connections with the clinical scores were assessed. The findings indicated that scores lacked reliability across the six occasions of measurement and hence validity could not be established. The need to improve the specificity of critical dimensions in clinical performance for the purposes of providing reliable measurement and accurate student feedback was acknowledged by academic staff and the researchers.