Validity, Reliability and Adaptation of the DREEM in Colombian Spanish and an Integrated Curriculum

Abstract

Introduction: Measurement of the educational environment has become of key importance in medical curriculum evaluation. The DREEM has been developed for this purpose. However, a number of studies have identified issues with the psychometric properties of the DREEM. The internal consistency of the 5 domains is rather variable and, in some cases, the reliability is low. Since information obtained from the DREEM can be used to change the curriculum in our context, it is necessary to validate this instrument.
Objective: To validate the DREEM in Colombian Spanish considering an integrated curriculum.
Method: Quantitative, non-experimental, exploratory, cross-sectional study. 12 experts in education validated the DREEM content. 308 medical students participated. Exploratory Factorial Analysis (EFA) and Cronbach’s α determined construct validity and reliability.
Results: Syntactic and semantic adaptations were made to the items. Seven items were mobilized from one domain to another. Although the total DREEM score had an excellent reliability level (α = 0.91), domain analysis showed 2 questionable domains. The analysis indicated that 26 items may be enough for measuring the educational climate. Domain V items and item 17 were excluded.
Conclusions: The adapted DREEM to Colombian Spanish is reliable and valid. However, some of the domains were questionable. Findings suggest the original theory of the DREEM domains embodies debatable assumptions. Excluding the independent analysis of the domains may harden scores understanding, which in turn may obstruct decision making in specific areas of the curriculum. Therefore, A new theory of the educational climate domains is needed.