Objective: This study reports the psychometric properties of the 24-item version of the Diabetes Knowledge Questionnaire (DKQ).
Research design and methods: The original 60-item DKQ was administered to 502 adult Mexican-Americans with type 2 diabetes who are part of the Starr County Diabetes Education Study. The sample was composed of 252 participants and 250 support partners. The subjects were randomly assigned to the educational and social support intervention (n = 250) or to the wait-listed control group (n = 252). A shortened 24-item version of the DKQ was derived from the original instrument after data collection was completed. Reliability was assessed by means of Cronbach's coefficient alpha. To determine validity, differentiation between the experimental and control groups was conducted at baseline and after the educational portion of the intervention.
Results: The 24-item version of the DKQ (DKQ-24) attained a reliability coefficient of 0.78, indicating internal consistency, and showed sensitivity to the intervention, suggesting construct validation.
Conclusions: The DKQ-24 is a reliable and valid measure of diabetes-related knowledge that is relatively easy to administer to either English or Spanish speakers.