Sarnacki R E
J Med Educ. 1982 Mar;57(3):163-9. doi: 10.1097/00001888-198203000-00002.
The study reported here assessed the relative utility of premedical grade-point average (GPA) in distinguishing among prospective medical students and in predicting the medical school performance of accepted applicants. Students from each of two medical classes were grouped into one of four categories depending on the "selectivity" rating of their premedical, undergraduate institution. Group differences were assessed according to premedical GPAs, Science subtest scores of the Medical College Admission Test, medical school course work, and scores on the Part I and Part II examinations of the National Board of Medical Examiners. Results indicated that although the groups differed significantly on premedical GPA (with mean scores from the less competitive undergraduate colleges exceeding those of the most selective schools), on all other dependent measures no significant differences resulted. It was concluded that "raw" premedical GPA is susceptible to a number of extraneous sources of variance that, unless identified and eliminated, negate its effectiveness in assessing individual differences in past academic achievement and in predicting future medical school performance.
本文所报告的研究评估了医学预科平均绩点（GPA）在区分未来医学生以及预测已录取申请人在医学院的表现方面的相对效用。来自两个医学班级的学生根据其医学预科本科院校的“选择性”评级被分为四类之一。根据医学预科GPA、医学院入学考试的科学子测试成绩、医学院课程作业以及国家医学考试委员会第一部分和第二部分考试的成绩来评估组间差异。结果表明，尽管这些组在医学预科GPA上存在显著差异（竞争力较弱的本科院校的平均成绩超过了最具选择性的学校），但在所有其他相关指标上均未出现显著差异。研究得出结论，“原始的”医学预科GPA容易受到许多外部方差来源的影响，除非识别并消除这些影响，否则它在评估过去学业成绩的个体差异以及预测未来医学院表现方面的有效性将被削弱。
