Measurement methodology in children's health locus of control

Health locus of control scales for children are often constructed in an agree/disagree format. It was hypothesized that the structure of the instrument may be in part responsible for the finding that young children (7-8 yr) have an external health locus of control relevant to children several years older. The original version and a revised version, using a choice of attribution format, were both administered to 444 students (98% of eligible students) attending 10 second-grade classes. Although the mean scores for the two formats were the same, 32.0 (SD = 3.3), item level analyses showed poor agreement (kappa, -.005-.41) and significant bias in disagreements on 15 of the 20 items. Changes in the wording of the questions led to different results, indicating possible limitations in either format. The observation that young children likely have a mixed health locus of control indicates that health educators may benefit from employing both internal and external sources of reinforcement to promote healthy behaviors in young children.