CME Segmentation Modeling: Using a Strategic Process for Aligning Educational Needs and Activities
Background: Current continuing medical education (CME) strategies are developed and delivered as “one-size-fits all” type programs. However, the “one-size-fits all” approach to CME may be inappropriate for certain subpopulations and therefore have little impact. The purpose of this study was to provide a conceptual overview of the essential steps involved in segmentation analysis and illustrate the use of 2 different analytical approaches designed to derive segments from a defined population.
Methods and Results: Data from a baseline assessment study (n = 1890) conducted between March and May 2007 using evidence-based case vignettes were used to illustrate two analytic procedures. The first method, K-Means Cluster Analysis, is primarily a descriptive, nonstatistical approach to segmentation. The second method, Latent Class Analysis, is a model-based clustering procedure used to identify mutually exclusive categorical latent (unobserved) classes (segments) of cases within a population based on the patterns of responses for a set of observed measures (indicators).
Conclusion: When conducted properly, segmentation research can provide valuable information for the design, implementation, and evaluation of CME activities. With implementation of segmentation strategies, educators can better align the educational needs and preferences of target groups to increase the likelihood of program effectiveness and optimize activity resources.
- There are currently no refbacks.