Hanson Mark D, Dore Kelly L, Reiter Harold I, Eva Kevin W
HSC 3G41, McMaster University, Hamilton, ON L8N 3Z5, Canada.
Acad Med. 2007 Oct;82(10 Suppl):S8-S11. doi: 10.1097/ACM.0b013e3181400068.
Some form of candidate-written autobiographical submission (ABS) is commonly used before interviews to screen candidates to medical school on the basis of their noncognitive characteristics. However, confidence in the validity of these measures has been questioned.
In 2005, applicants to McMaster University completed an off-site ABS before being interviewed and an on-site ABS at interview. Five off-site ABS questions were completed, plus eight on-site questions. On-site ABS questions were answered in variable timing conditions. ABS ratings were compared across sites and time allowed for completion.
Off-site ABS ratings were higher than on-site ratings, and the two sets of ratings were uncorrelated with one another. On-site ABS ratings increased with increased time allowed for completion, but the reliability of the measure was unaffected by this variable.
Confidence that candidates independently answer preinterview ABS questions is weak. To improve ABS validity, modification of the current Web-based submission format warrants consideration.
在面试前，通常会采用某种形式的由候选人撰写的自传式材料（ABS），以便根据非认知特征对医学院校的候选人进行筛选。然而，人们对这些测评方式的有效性存在质疑。
2005年，麦克马斯特大学的申请者在面试前完成一份校外ABS，面试时完成一份校内ABS。校外ABS需回答5个问题，校内ABS需回答8个问题。校内ABS问题在不同的时间限制条件下回答。比较了校外和校内ABS的评分以及完成作答的时间。
校外ABS评分高于校内评分，且两组评分互不相关。校内ABS评分随着完成时间的增加而提高，但该测评方式的可靠性不受此变量影响。
对于候选人独立回答面试前ABS问题的可信度较低。为提高ABS的有效性，值得考虑对当前基于网络的提交格式进行修改。
