Comparative validity of the Internet and paper-and-pencil versions of the Night Eating Questionnaire

• Published in: Comprehensive psychiatry Vol. 75; pp. 53 - 61
• Main Authors: Tu, Chao-Ying, Tseng, Mei-Chih Meg, Chang, Chin-Hao, Lin, Chao-Cheng
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.05.2017; Elsevier Limited
• Subjects: Chronobiology Disorders - diagnosis; Chronobiology Disorders - psychology; Clinical medicine; Drug use; Eating - psychology; Eating disorders; Feeding and Eating Disorders - diagnosis; Feeding and Eating Disorders - psychology; Female; Gastrointestinal surgery; Humans; Internet; Language; Male; Mental health; Methods; Onsite; Patients; Psychiatry; Psychometrics; Quality; Quantitative psychology; Reproducibility of Results; ROC Curve; Surveys and Questionnaires; Translations; Validity
• ISSN: 0010-440X; 1532-8384
• DOI: 10.1016/j.comppsych.2017.03.001

Abstract Objective This study examined the psychometric properties of the Internet and paper-and-pencil versions of the Mandarin Chinese version of the Night Eating Questionnaire (C-NEQ) and compared these measures' validity. Method The C-NEQ was evaluated through two different media: 626 participants completed the C-NEQ on the Internet and 160 participants completed the paper-form C-NEQ at the psychiatric outpatient clinics. A subgroup completed both versions of the C-NEQ ( n = 50). The Night Eating Syndrome History and Inventory was used to identify individuals with night eating syndrome (NES). Results The paper-and-pencil and Internet versions of the C-NEQ both showed good internal consistency, reliability, and concurrent validity. Reliability between the Internet and the paper-and-pencil versions of the C-NEQ was excellent (ICC = .96). Diagnostic analysis of the C-NEQ's performance using the Receiver Operation Curve method showed excellent results in both versions; the area under the curve did not differ significantly between the versions. Regarding detecting NES, the Internet version had a higher optimal cutoff point than the paper-and-pencil version (23 and 22, respectively). Conclusions The Internet and paper-and-pencil versions of the C-NEQ both showed strong reliability and validity; however the two versions appear to differ marginally regarding usage in NES detection.