In light of increasing rates of overweight and obesity worldwide, there is a critical need for accurate self-report measures of disinhibited and restrained eating behaviors across the weight spectrum. Item response theory was used to determine whether differences in disinhibited and restrained eating between healthy weight and overweight or obese individuals were due to item bias (i.e., differential item functioning). Study 1 participants were healthy weight (n = 510) or overweight or obese (n = 304) adults recruited from the community. Study 2 participants were healthy weight (n = 778) or overweight or obese (n = 320) college students. Study 1 participants completed the Eating Disorder Examination-Questionnaire (EDE-Q), Eating Disorder Inventory-3, Dutch Eating Behaviors Questionnaire, Restraint Scale, and Three-Factor Eating Questionnaire. Study 2 participants completed the Eating Pathology Symptoms Inventory (EPSI). Items on the Restraint Scale demonstrated the most evidence for bias (60% of items), whereas the majority of other scales demonstrated low to moderate levels of item bias (17-38% of items). However, EDE-Q Restraint and EPSI Binge Eating, Cognitive Restraint, Excessive Exercise, Muscle Building, and Negative Attitudes Toward Obesity scales did not show any evidence of differential item functioning among weight groups. Participants with the same level of disordered eating responded differently to certain eating disorder self-report items due to weight-bias, rather than true between-groups differences. Nevertheless, EDE-Q Restraint, EPSI Cognitive Restraint, and EPSI Binge Eating did not exhibit any evidence of bias and are ideal for assessing restrained and disinhibited eating across the weight spectrum in both research and clinical settings. (PsycINFO Database Record (c) 2020 APA, all rights reserved).

中文翻译：

---

在自我报告的饮食禁忌和饮食禁忌方面，健康体重或超重个体之间的差异是否反映了现实或“偏见”？

鉴于全球超重和肥胖症的发病率不断上升，迫切需要针对体重范围内的禁忌和约束饮食行为进行准确的自我报告测量。项目反应理论用于确定健康体重和超重或肥胖个体在禁忌饮食和约束饮食方面的差异是否归因于项目偏见（即，不同的项目功能）。研究1的参与者是从社区招募的健康体重（n = 510）或超重或肥胖（n = 304）的成年人。研究2的参与者是健康体重（n = 778）或超重或肥胖（n = 320）的大学生。研究1的参与者完成了饮食失调问题调查问卷（EDE-Q），饮食失调清单3，荷兰人饮食行为问卷，约束量表和三因素饮食问卷。研究2的参与者填写了《饮食病理症状清单》（EPSI）。约束量表上的项目显示出最多的偏见证据（占项目的60％），而大多数其他量表显示出项目偏见的水平低至中度（占项目的17-38％）。但是，EDE-Q抑制和EPSI暴饮暴食，认知抑制，过度运动，肌肉锻炼和对肥胖的负面态度量表没有显示任何证据表明体重组之间存在差异项功能。饮食失调水平相同的参与者对某些饮食失调自我报告项目的反应因体重偏见而不是真正的群体间差异。尽管如此，EDE-Q约束，EPSI认知约束，和EPSI Binge Eating没有表现出任何偏见的证据，是在研究和临床环境中评估整个体重谱范围内禁忌和禁忌饮食的理想选择。（PsycINFO数据库记录（c）2020 APA，保留所有权利）。