Profiling students for remediation using latent class analysis.

Area of Science:

• Medical Education
• Educational Assessment
• Psychometrics

Background:

• Clinical exams using standardized patients (SPs) are crucial for medical student assessment.
• Current methods for identifying students requiring remediation lack consensus and vary widely.
• There is a need for robust methodologies to identify students for targeted academic support.

Purpose of the Study:

• To introduce Latent Class Analysis (LCA) as an alternative method for identifying medical students needing remediation.
• To evaluate the utility of LCA in analyzing performance data from the Clinical Performance Examination (CPX).
• To explore the identification of distinct student performance profiles for remediation purposes.

Main Methods:

• Latent Class Analysis (LCA) was applied to performance data from 147 third-year medical students.
• Students participated in the Clinical Performance Examination (CPX) as part of the study.
• The LCA model identified distinct subgroups of students based on their performance patterns.

Main Results:

• Three distinct clusters of students with varying performance profiles were identified using LCA.
• The analysis revealed two low-performing groups with contrasting weaknesses across different case types.
• This differentiation has implications for setting performance cut-points and designing remediation programs.

Conclusions:

• Latent Class Analysis (LCA) provides a flexible and advantageous alternative to traditional methods for identifying students for remediation.
• The ability to identify multiple low-performing groups allows for more nuanced and targeted interventions.
• This approach enhances the precision of identifying students who may benefit from additional instruction or support.