Biller I R, Kerber P E
J Dent Educ. 1980 Apr;44(4):206-10.
Reliability of scaling error detection among dental hygiene instructors is a basis for grade equity. This study tested statistically (ANOVA) grading reliability and variance factors using ten students' scaling performance scores while treating 34 patients. Nine instructors were grouped according to experience, and the patients were grouped according to four difficulty levels. The interinstructor differences were significant in the inexperienced group only. Variance analysis implied that 26 to 94 percent of the grade given is due to instructor differences rather than the students' performances. Experienced graduate students graded most consistently (26 percent of variance due to instructor differences) while faculty were less consistent (64 percent) and inexperienced graduate students exhibited marked inconsistency (94 percent). Error detection appeared to be more closely related to the instructors' educational background than to experience. Results of the study suggest the need for clinical rotation of instructors as well as for instructor calibration.
牙科保健教员之间牙石清除误差检测的可靠性是评分公平性的基础。本研究使用10名学生在治疗34名患者时的牙石清除操作表现分数,对评分可靠性和方差因素进行了统计学检验(方差分析)。9名教员根据经验分组,患者根据四个难度级别分组。仅在无经验组中,教员之间的差异具有统计学意义。方差分析表明,所给分数的26%至94%归因于教员差异,而非学生的表现。经验丰富的研究生评分最为一致(26%的方差归因于教员差异),而教员的一致性较差(64%),无经验的研究生表现出明显的不一致(94%)。误差检测似乎与教员的教育背景比与经验的关系更为密切。研究结果表明,教员需要进行临床轮转以及校准。