Reliability and validity of a computerized questionnaire to measure fat intake in Belgium

This study evaluated the reliability and validity of a Belgian computerized questionnaire to measure fat intake. A total of 86 adults completed a 48-item computerized fat intake questionnaire, subsequently kept a 7-day diet record, and finally completed the questionnaire again. For validity, a Pearson correlation of 0.67 for total fat intake and of 0.60 for percent energy from fat were found between the computerized questionnaire and the diet records. Gross misclassification between the two methods was 5.8% for total fat intake and for percent energy from fat. Cohen’s K values were respectively 0.27 and 0.29. For reliability, intraclass correlation coefficients ranged from 0.70 to 0.87 between the two administrations of the computerized fat intake questionnaire. Gross misclassification between the administrations was 1.2% for total fat intake and for percent energy from fat, K values were respectively 0.64 and 0.50. These results indicate that this fat intake questionnaire has an acceptable reliability and validity.