客观结构化临床考试的多元概化模型研究

探讨客观结构化临床考试(OSCE)的可靠性,为改进其设计和提高考试可靠性提供依据.根据概化理论,拟合两因子多元概化模型,探讨OSCE及其分测验的可靠性,同时考察各分测验样本含量的变化对各自测验信度和全域合成总分测量可靠性的影响.结果OSCE合成全域总分的概化系数为0.44217,SP考站的概化系数为0.65864,非SP考站作为常模参照测验(概化系数0.21999)不理想.研究表明今后OSCE设置8个SP考站与14个非SP考站即可达到测量的可靠性要求.

An application of multivariate generalizability theory in the study on the reliability of objective structured clinical examination

To further investigate the overall reliability of an OSCE and the effect of differences in the number of items on the reliability,a total of 219 final-year undergraduates of West China Medical School Sichuan University took a 12-station OSCE. Multivariate Generalizability Theory was used to estimate the overall reliability of the examination,the reliability of the two subtests and their respective contributions to overall reliability. The composite reliability of the examination was 0.4422. The loss of reliability due to the use of fewer stations in the existing OSCE can be compensated by lengthening the SP-based subtests to eight and Non-SP stations to 14. The OSCE achieved an acceptable overall reliability sufficient for evaluating the clinical competence. However,examination structure must be carefully planned and results combined with caution. The components testing different aspects of knowledge and clinical skills must be carefully balanced to ensure reliability.