Stick a fork in it: A psychometric evaluation of a food frequency questionnaire and an activity frequency questionnaire

Kelley E Borradaile, University of Pennsylvania

Abstract

Introduction. Child and adolescent obesity are major public health concerns with consequences extending into psycho-social and economic areas. Despite the amount of resources devoted to the prevention of obesity, many studies are weak and inconsistent, attributable, partly, to the short length of intervention and study design. As a result, policymakers and researchers have struggled to find effective means by which to identify the components of interventions that work, in an attempt to curb the obesity epidemic in children. Questionnaires that direct attention to what works have been developed and are purported to be cheap, reliable, and valid measures of diet and activity. ^ Purpose. This dissertation used the theory of energy balance to psychometrically evaluate the Youth/Adolescent Questionnaire (YAQ) and the Youth/Adolescent Activity Questionnaire (YAAQ) against body mass index (BMI). ^ Sample. Over 1,300 4th--6th grade Philadelphia students that were part of an ongoing cluster randomized obesity prevention trial were assessed annually over three consecutive school years. Participants were administered the YAQ and YAAQ and heights and weights were collected to calculate BMI z-scores. ^ Methods. The YAQ and YAAQ were assessed descriptively and were subjected to tests of reliability and validity both cross-sectionally and over two years using analysis of covariance and individual growth models. Factor analytic procedures were applied to the YAQ and YAAQ to extract an alternative set of dimensions, which were subjected to a set of confirmatory procedures. The new scales were assessed using the same psychometric analyses to determine if they were more valid than the original scales. ^ Results. The YAQ and the YAAQ, while reliable, were not valid when considered against BMI z-score. Similarly, the re-scaled questionnaires were not valid against BMI z-score. No substantive differences existed between subgroups (gender, race, age, relative weight category) with respect to the reliability and validity of these questionnaires.^ Conclusion. The YAQ and the YAAQ do not relate to BMI z-score in a manner consistent with the theory of energy balance. ^

Subject Area

Education, Tests and Measurements|Health Sciences, Nutrition|Health Sciences, Public Health|Psychology, Psychometrics

Recommended Citation

Kelley E Borradaile, "Stick a fork in it: A psychometric evaluation of a food frequency questionnaire and an activity frequency questionnaire" (January 1, 2007). Dissertations available from ProQuest. Paper AAI3260882.
http://repository.upenn.edu/dissertations/AAI3260882

Share

COinS