Comparability and Reliability of Paper- and Computer-Based Measures of Psychosocial Constructs for Adolescent Fruit and Vegetable and Dietary Fat Intake

Background This study investigated the comparability and reliability of computer- and paper-based measures of psychosocial constructs related to fruit and vegetable and dietary fat intake among adolescents. Methods An ethnically diverse sample of 76 adolescents was studied (mean age 13 years). Scales measured use of change strategies, self-efficacy, decisional balance, family influences, and peer influences separately for the two dietary outcomes. Results Comparability analyses indicated that responses for each of the 12 diet-related scales were not significantly different between the computer- and paper-based surveys. Internal consistencies were generally high (α from .61 to .97) with slightly better reliability on the computer- vs paper-based surveys. Test–retest reliabilities were adequate to good for most multiple-item scales (interclass correlation coefficients from .43 to .85 and .48 to .90 for paper and computer formats, respectively). Conclusions Computer- and paper-based measures of psychosocial constructs are appropriate and ready for use in either format for studies of dietary behaviors in young people.