Steps towards validation of the Dental Education Clinical Learning Instrument (DECLEI) in American dental schools (DECLEI-USA).

Abstract

Few tools assess the dental school clinical learning environment from students' perspectives. Considering previous efforts to validate the Dental Education Clinical Learning Instrument (DECLEI) in the United States, the goal of this study was to look for the fewest number of items that accounted for the most amount of variability in responses and/or had the highest correlation to the total DECLEI score using a larger, multi-center sample. The DECLEI was distributed to 286 students in two dental schools (University of Iowa and University of Texas at Houston) during the 2017-2018 academic year. Two alternative methods were applied. In the first approach, all 24-items were regressed on the total composite score using a forward conditional method. In the second approach, the item-total correlation for the full scale was calculated and then items with relatively poor coefficients were eliminated. A cutoff of 0.30 or less was used. The first approach, Total R2 by Regression Model, produced a 9-item scale accounting for 90% of the variance in total score and a Cronbach's α coefficient of 0.79. The second approach, Item-Total Correlation, produced a larger scale (20 items), as well as a higher Cronbach's α coefficient of 0.89. The instrument also presented appropriate sensitivity to measure differences between race groups and school of origin. DECLEI may have the potential to be used as an instrument to measure clinical learning environments for U.S. dental students using either a smaller, concise scale (Mini DECLEI-USA) or a larger (DECLEI-USA), more thorough scale.