ABSTRACT
OBJECTIVE@#The primary objective of this study was to examine the validity and reliability of a semi-quantitative food frequency questionnaire (FFQ) among Chinese children aged 12-17 years.@*METHODS@#A semi-quantitative 72-food item FFQ was developed for children aged 12-17 years. The reliability and validity of this FFQ were evaluated against 24-h dietary recalls (24 h DRs) to measure the consumption of foods and nutrients. We administered two FFQs and three DRs to children (N = 160) over a period of 1 month to evaluate the reliability and validity. Reliability was examined by quartile agreement and intraclass correlation coefficients (ICCs), and validity was examined by quartile agreement, Bland-Altman plots and correlation with DRs.@*RESULTS@#For reliability, the ICCs between the two FFQs ranged from 0.21 to 0.76 for foods and nutrients, and the quartile agreement ranged from 70.0% to 95.0% in the same or adjacent quartiles. Spearman's correlation coefficients of foods and nutrients between the second FFQ and the 24 h DRs ranged from -0.04 to 0.59. The Bland-Altman plots demonstrated good agreement across the range of intakes among nutrients. The quartile agreement ranged from 50.0% to 100.0%, with infrequent misclassification.@*CONCLUSION@#The FFQ assessment of dietary intakes demonstrated acceptable relative validity and high reproducibility for Chinese children aged 12-17 years.