Psychometric study of the checklist to evaluate clinical learning environments

Abstract

Introduction: Clinical learning environments are educational settings focused on directly treating individuals with health problems. Through the experiences gained in these environments, resident physicians can integrate knowledge, develop competencies, and acquire the culture, values, and attitudes inherent to the profession. Understanding the influence of these environments on learning is crucial, and to do so, instruments are needed to measure their characteristics. Various instruments exist to measure clinical environments; however, they differ in scope, dimensions, and scoring methods. There is a theoretical proposal developed in México, designed with the Delphi methodology by a group of experts in medical education and psychometric assessment, focused on defining the main general dimensions of clinical learning settings.

Objective: To psychometrically validate the measurement instrument in a sample of medical residents at a tertiary-level hospital in Mexico City.

Method: The study involved 248 resident physicians enrolled in nine specialty courses at a tertiary hospital in Mexico City. 52.82% (131) were women, aged 24-51 years, with a mean age of 29.19 years (SD = 2.92). The instrument consisted of 15 dichotomous items to measure four dimensions: social climate, formal educational program, clinical work, and institutional culture. To prove criterion validity, the Clinical Learning Environment Instrument for Mexican Medical Residents was used, which consists of 30 Likert-type items with adequate reliability and evidence of construct validity on its structure. The study was approved by a Research Committee and an Ethics in Research Committee, and informed consent was obtained.  The scale was administered in a face-to-face session at the hospital. Analyses were conducted to obtain evidence for four psychometric properties: discriminative power, construct validity, criterion validity, and reliability.  

Results: a four-factor version with 15 items was obtained, with a high reliability (ordinal α = .96), evidence of construct validity (75% of explained variance), and criterion validity in the correlation with the Clinical Learning Environment Instrument for Mexican Medical Residents (r=.76, p<.01).

Conclusions: This study resulted in a reliable and valid version of the instrument, which will enable its use to investigate the role of clinical learning environments in medical education and to diagnose needs in Mexican hospitals.