van der Vleuten C P, van Luyk S J, Swanson D B
University of Limburg, The Netherlands.
Res Med Educ. 1988;27:228-33.
This study investigated the reproducibility of scores on the University of Limburg Skills Test, using data drawn from tests administered to medical students in all years of training during 1984-85, 1985-86, and 1986-87. Generalizability analyses indicated that, while inter-rater reliability is good, a minimum of three to four hours of testing time is required to obtain reproducible scores due to content-specificity of examinee performance.
本研究利用1984 - 1985年、1985 - 1986年和1986 - 87年期间对各年级医学生进行测试所获得的数据，调查了林堡大学技能测试分数的可重复性。概化分析表明，虽然评分者间信度良好，但由于考生表现的内容特异性，需要至少三到四个小时的测试时间才能获得可重复的分数。
