Words That Match: Linguistic Analysis of Plastic Surgery Residency Personal Statements to Predict Match and Identify Bias

Background: Plastic surgery residency selection remains among the most competitive in medicine. With a trend away from objective measures such as Step scores and grades, there is an increasing emphasis on holistic review and subjective assessments such as the personal statement. Prior analyses across other specialties suggest that linguistic features—reflecting tone, reasoning, and self-presentation—may influence applicant success and reveal systemic bias. Whether similar linguistic patterns exist in plastic surgery remains unknown. Methods: Retrospective analysis of 298 integrated plastic surgery applicants (24-25 cycle) to one institution was performed. De-identified PS were analyzed using the Linguistic Inquiry and Word Count (LIWC-22) software to quantify psychological and linguistic domains. Applicant demographic and academic metrics were extracted. Independent-sample t tests and multivariable regression controlled for gender, Step 2 CK, AOA, and first-author publications to identify linguistic features associated with match outcome and racial differences. Significance was set at p<0.05. Results: Of 298 applicants, 185 (62.1%) matched and 113 (37.9%) did not. Matched applicants had higher Step 2 scores (p<0.001), more first-author and total number of publications and presentations (all p<0.001), and greater AOA membership (p=0.010). Matched applicants showed higher analytical thinking (β=2.84, p=0.005), greater vocabulary complexity (β=1.11, p=0.04), and used second-person pronouns 51% less (β=-0.065, p=0.006). Word count itself showed no significant association with match outcome (p=0.85). Controlling for academic factors and match status, Asian applicants used more analytical and complex language than White peers (β=3.53, p=0.004; β=1.58, p=0.017). Black and Hispanic applicants expressed lower positive emotion (β=-0.157, p=0.034) and certitude (β=-0.081, p=0.038). Interaction analyses confirmed linguistic features predict match equivalently across races—criteria are applied uniformly, but baseline patterns differ systematically. Conclusion: Linguistic style predicts plastic surgery match outcomes independent of traditional metrics. Successful applicants wrote with structured reasoning, complex vocabulary, and a self-referential rather than audience-directed tone—linguistic markers of belonging within the field. However, underrepresented applicants exhibited distinct expressive styles, possibly reflecting cultural norms or outsider perspectives. Even if evaluators apply uniform criteria, these standards may privilege communication patterns more common among majority applicants. Addressing this requires structural change such as standardized rubrics, blinded applicant review, faculty bias training, and greater representation in academic leadership.