A new method for group decision making and its application in medical trainee selection

Area of Science:

• Medical Education
• Health Professions Education
• Collaborative Decision-Making

Background:

• Generating collaborative ranked preference lists is a challenge in academic medicine, impacting trainee selection and resource allocation.
• Current pseudo-quantitative methods, like averaging arbitrary scores, are mathematically flawed and may not represent true group consensus.
• These issues arise in medical school admissions, postgraduate training selection, performance assessments, and financial prioritization.

Purpose of the Study:

• To introduce a novel algorithm for creating collaborative preference lists.
• To overcome the limitations of arbitrary scoring systems in group decision-making processes.
• To provide a transparent, reproducible, and equitable method for ranking applicants or choices.

Main Methods:

• Developed a novel algorithm based on pairwise comparisons, avoiding Likert scale scores.
• Implemented the algorithm to generate and sort a matrix of comparisons.
• Conducted a case study during the 2013 neonatal-perinatal fellowship match.

Main Results:

• The algorithm produced an efficient rank-order list that required no reshuffling or extensive debate.
• Faculty and fellows reported significantly higher satisfaction with the new algorithm compared to score-based systems.
• A unanimous preference for the new algorithm over traditional methods was observed.

Conclusions:

• The developed algorithm reduces arbitrariness in collaborative preference list generation.
• This method demonstrates wide applicability in medical education, training, and other group decision-making contexts.
• Use in the National Resident Matching Program match improved perceptions of fairness, ease of use, and efficiency.