匹配和不匹配的耳鼻喉科住院医师申请人的语言分析
Language Analysis of Matched and Unmatched Otolaryngology Residency Applicants.
作者信息
Hamdi Osama A, Hirsch Scott D, Jamil Taylor L, Samarah Hani, Johnson Andrew P, Cabrera-Muffly Cristina
机构信息
Department of Otolaryngology-Head and Neck Surgery University of Colorado Anschutz Medical Campus Aurora Colorado USA.
Division of Pediatric Otolaryngology Texas Children's Hospital Houston Texas USA.
出版信息
Laryngoscope Investig Otolaryngol. 2025 Jun 14;10(3):e70170. doi: 10.1002/lio2.70170. eCollection 2025 Jun.
OBJECTIVES
The otolaryngology residency match process is highly competitive. This study evaluated demographic, academic, and linguistic differences in personal statements (PSs) and letters of recommendation (LORs) between matched and unmatched applicants.
METHODS
A retrospective quantitative analysis of 2130 residency applicants to one institution during the 2015-2021 match cycles was conducted. Linguistic Inquiry and Word Count 2022 (LIWC2022), a validated software tool, was used to analyze emotional, cognitive, and structural components of written text.
RESULTS
Among 2130 applicants, 1671 (78.5%) matched. Non-Hispanic White applicants had significantly higher match rates compared to non-white applicants ( = 0.001). Matched applicants demonstrated superior academic performance, including higher USMLE Step 1 scores (247 ± 12.2 vs. 239.2 ± 16.1, < 0.001), Step 2 CS pass rates (99.2% vs. 95.8%, < 0.001), and greater research productivity. Linguistic analysis revealed that matched applicants used more analytical (OR = 1.02, < 0.001), clout (OR = 1.01, = 0.007), and positive tone (OR = 1.16, = 0.003) language in PSs. Unmatched applicants used more negative tone (OR = 0.619, < 0.001) and negative emotion (OR = 0.54, = 0.001). LORs for matched applicants exhibited less negative tone (OR = 0.35, < 0.001).
CONCLUSION
Matched applicants displayed stronger academic metrics and cognitive language patterns in PSs, highlighting an emphasis on analytical writing over emotional expression. LOR language differed between groups, revealing linguistic features predictive of matching success.
LEVEL OF EVIDENCE
III.
目的
耳鼻喉科住院医师匹配过程竞争激烈。本研究评估了匹配和未匹配申请人在个人陈述（PS）和推荐信（LOR）中的人口统计学、学术和语言差异。
方法
对2015 - 2021年匹配周期内一所机构的2130名住院医师申请人进行回顾性定量分析。使用经过验证的软件工具语言查询与字数统计2022（LIWC2022）来分析书面文本的情感、认知和结构成分。
结果
在2130名申请人中，1671人（78.5%）匹配成功。非西班牙裔白人申请人的匹配率显著高于非白人申请人（P = 0.001）。匹配成功的申请人表现出更优异的学术成绩，包括更高的美国医师执照考试第一步（USMLE Step 1）分数（247 ± 12.2 vs. 239.2 ± 16.1，P < 0.001）、第二步临床技能（Step 2 CS）通过率（99.2% vs. 95.8%，P < 0.001）以及更高的研究产出。语言分析显示，匹配成功的申请人在个人陈述中使用了更多的分析性语言（OR = 1.02，P < 0.001）、影响力语言（OR = 1.01，P = 0.007）和积极语气语言（OR = 1.16，P = 0.003）。未匹配成功的申请人使用了更多的消极语气语言（OR = 0.619，P < 0.001）和消极情绪语言（OR = 0.54，P = 0.001）。匹配成功申请人的推荐信表现出较少的消极语气（OR = 0.35，P < 0.001）。
结论
匹配成功的申请人在个人陈述中展现出更强的学术指标和认知语言模式，突出了对分析性写作而非情感表达的重视。不同组之间推荐信的语言存在差异，揭示了预测匹配成功的语言特征。
证据水平
三级。
Zotero 插件