Background/Purpose: The Fitness for Life Middle School (FFL) curriculum, grounded in social cognitive theories (Bandura, 2004), was developed in 2007 with many teachers using the test bank of items online to assess students conceptual knowledge; however, no validation study has been done on the items. The purpose of this study, also informed by the public health literature, was to validate the test bank items (N=289) in order to have items that provide psychometrically sound information on student healthy behavior knowledge.
Method: First, experts (N=14, both genders, mostly Caucasian) participated in two rounds of assessments, classifying items into content areas and evaluating appropriateness. Second, students from the Southwestern USA (N=663, grades 7-12 [mostly 9th, 86%], both genders and various ethnic backgrounds), completed four versions (with 21 common items and 52 unique items each) of the student knowledge assessment.
Analysis/Results: Responses were analyzed using the two-facet Rasch model for students and items. The model-data fit well with all Infit and Outfit statistics between 2 and -2. The items showed a good range of difficulty (3.09 to -1.74 logits) and precision (0.18 to 0.00 SE).
Conclusions: A test bank with 229 items and four forms has been created that can be used by teachers and researchers to assess student knowledge outcomes from the FFL middle school curricular model. Results showed items are now on the same measurement scale, thus creating four equivalent tests. With Rasch calibration, measurement problems in conventional testing assessments were eliminated and a valid and reliable assessment system was developed.