A food frequency questionnaire for youth: Psychometric analysis and summary of eating habits in adolescents

Purpose: To examine the reliability and validity of the Eating Habits Quest ionnaire (EHQ) for adolescents and assess eating habits in the context of t he Food Pyramid. Methods: Subjects were 446 students (81.2% female) attending three middle s chools (sixth to eighth grades). Over half (56.9%) were African-American. R eliability was assessed with 48-h and 2-week retests on two subsamples (n = 62) and validity was assessed by comparison with three 24-h recalls on an additional subsample (n = 24) and by factor analysis. The EHQ is a self-adm inistered questionnaire consisting of 83 food items and questions assessing food habits and food preparation style. Results: internal validity, determined with factor analysis, found that 10 factors explained 81.3% of the variance in eating habits. Overall perfect a greement between food groups reported on the 24-h recall and on the EHQ was 56%. internal consistency, assessed by Cronbach alpha, ranged from .60 to .89. Test-retest correlations were highest at 48 h, ranging from .46 to .85 for the 10 factors. Adolescents ate more servings of fats and sugars (11.2 ) and meats (6.0) and fewer breads (5.2) and vegetables (1.6) than recommen ded in Food Pyramid guidelines. In addition, subjects consumed more high-fa t protein than lo tv-fat protein servings. Conclusions: The EHQ is a valid and reliable food frequency questionnaire t hat has been tested with African-American and white adolescents in the Sout heast. It showed that adolescents consume higher than recommended daily ser vings of fats, sweets, and meats and lower than recommended servings of veg etables and breads. (C) Society for Adolescent Medicine, 2000.