OBJECTIVE: To assess if the Thomas-Kilmann Conflict MODE Instrument predicts residents' performance. STUDY DESIGN: Nineteen residents were assessed on the Thomas-Kilmann conflict modes of competing, collaborating, compromising, accommodating, and avoiding. Residents were classified as contributors (n = 6) if they had administrative duties or as concerning (n = 6) if they were on remediation for academic performance and/or professionalism. Data were compared to faculty evaluations on the Accreditation Council for Graduate Medical Education (ACGME) competencies. P value of < .05 was considered significant. RESULTS: Contributors had significantly higher competing scores (58% versus 17%; P = .01), with lower accommodating (50% versus 81%; P 5 .01) and avoiding (32% versus 84%; P = .01) scores; while concerning residents had significantly lower collaborating scores (10% versus 31%; P = .01), with higher avoiding (90% versus 57%; P = .006) and accommodating (86% versus 65%; P = .03) scores. There were significant positive correlations between residents' collaborating scores with faculty ACGME competency evaluations of medical knowledge, communication skills, problem-based learning, system-based practice, and professionalism. There were also positive significant correlations between compromising scores and faculty evaluations of problem-based learning and professionalism with negative significant correlations between avoiding scores and faculty evaluations of problem-based learning, communication skills and professionalism. CONCLUSIONS: Residents who successfully execute administrative duties are likely to have a Thomas-Kilmann profile high in collaborating and competing but low in avoiding and accommodating. Residents who have problems adjusting are likely to have the opposite profile. The profile seems to predict faculty evaluation on the ACGME competencies.