Development and Validation of Adolescent Media Health Literacy Scales: A Rasch Measurement Model Approach (Preprint)
BACKGROUND High media use is implicated in negative social and health outcomes in adolescents. Therefore, it is critical that adolescents develop skills to healthily engage with media-content. Yet, media health literacy (MHL), skills for assessing and responding to health-related media content and a potentially targetable moderator for the relationship between media use and health-related outcomes is understudied in adolescents. Lack of MHL assessment tools may contribute to this research gap. OBJECTIVE This study developed and validated test-based scales of adolescents’ MHL. METHODS Items developed were vetted iteratively via community reviews and cognitive interviews to establish content and face validity. Adolescents (n=355) completed a questionnaire including the revised MHL items. The scales were validated using Rasch measurement models. Convergent validity was assessed by correlating the summed scores of the three scales with existing functional and internet-related HL measures. Criterion validity was assessed by modeling logistic regressions predicting HL-related behaviors from each scale after controlling for demographics. Effect sizes were estimated. A short form was also validated. RESULTS The final MHL scales (Recognition/Identification, Influence/Critical analysis, Action/Reaction) fit their Rasch models. The 9-item Recognition/Identification and 9-item Influence/Critical analysis had good convergent validity with functional and internet-related health literacy measures and were positively related to reading instructions before taking medicine and questioning the truthfulness of online information. The 12-item MHL Scales-Short Form also had good convergent and criterion validity. However, convergent and criterion validity were not established for the 3-item Action/Reaction Scale. CONCLUSIONS The Recognition/Identification and Influence/Critical analysis scales and the MHL Scales-Short Form may be used to determine the impact of MHL on the media use/health outcomes relationships and ultimately inform the development of interventions and policies to affect these relationships in multiple settings.