University of Wales , Newport, UK.
University of Wolverhampton , UK.
J Sports Sci Med. 2004 Dec 1;3(4):244-53. eCollection 2004 Dec.
Theory testing and construct measurement are inextricably linked. To date, no published research has looked at the factorial validity of an existing eating attitude inventory for use with exercisers. The Eating Attitude Test (EAT) is a 26-item measure that yields a single index of disordered eating attitudes. The original factor analysis showed three interrelated factors: Dieting behavior (13-items), oral control (7-items), and bulimia nervosa-food preoccupation (6-items). The primary purpose of the study was to examine the factorial validity of the EAT among a sample of exercisers. The second purpose was to investigate relationships between eating attitudes scores and selected psychological constructs. In stage one, 598 regular exercisers completed the EAT. Confirmatory factor analysis (CFA) was used to test the single-factor, a three-factor model, and a four-factor model, which distinguished bulimia from food pre-occupation. CFA of the single-factor model (RCFI = 0.66, RMSEA = 0.10), the three-factor-model (RCFI = 0.74; RMSEA = 0.09) showed poor model fit. There was marginal fit for the 4-factor model (RCFI = 0.91, RMSEA = 0.06). Results indicated five-items showed poor factor loadings. After these 5-items were discarded, the three models were re-analyzed. CFA results indicated that the single-factor model (RCFI = 0.76, RMSEA = 0.10) and three-factor model (RCFI = 0.82, RMSEA = 0.08) showed poor fit. CFA results for the four-factor model showed acceptable fit indices (RCFI = 0.98, RMSEA = 0.06). Stage two explored relationships between EAT scores, mood, self-esteem, and motivational indices toward exercise in terms of self-determination, enjoyment and competence. Correlation results indicated that depressed mood scores positively correlated with bulimia and dieting scores. Further, dieting was inversely related with self-determination toward exercising. Collectively, findings suggest that a 21-item four-factor model shows promising validity coefficients among exercise participants, and that future research is needed to investigate eating attitudes among samples of exercisers. Key PointsValidity of psychometric measures should be thoroughly investigated. Researchers should not assume that a scale validation on one sample will show the same validity coefficients in a different population.The Eating Attitude Test is a commonly used scale. The present study shows a revised 21-item scale was suitable for exercisers.Researchers using the Eating Attitude Test should use subscales of Dieting, Oral control, Food pre-occupation, and Bulimia.Future research should involve qualitative techniques and interview exercise participants to explore the nature of eating attitudes.
理论检验和结构测量是密不可分的。迄今为止，没有发表的研究探讨过现有用于锻炼者的饮食态度量表的因子有效性。饮食态度测试（EAT）是一种 26 项的测试，可得出一个关于饮食失调态度的综合指标。原始的因素分析显示了三个相互关联的因素：节食行为（13 项）、口腔控制（7 项）和神经性贪食症-食物关注（6 项）。研究的主要目的是在锻炼者样本中检验 EAT 的因子有效性。第二个目的是调查饮食态度得分与选定心理结构之间的关系。在第一阶段，598 名经常锻炼者完成了 EAT。验证性因子分析（CFA）用于检验单因子、三因子模型和四因子模型，该模型区分了贪食症和食物关注。单因子模型的 CFA（RCFI=0.66，RMSEA=0.10）、三因子模型（RCFI=0.74；RMSEA=0.09）的模型拟合较差。四因子模型的拟合情况尚可（RCFI=0.91，RMSEA=0.06）。结果表明，五项表现出较差的因子负荷。剔除这 5 项后，重新分析了三个模型。CFA 结果表明，单因子模型（RCFI=0.76，RMSEA=0.10）和三因子模型（RCFI=0.82，RMSEA=0.08）的拟合情况较差。四因子模型的 CFA 结果显示出可接受的拟合指标（RCFI=0.98，RMSEA=0.06）。第二阶段探讨了 EAT 得分、情绪、自尊以及与锻炼相关的自我决定、享受和能力等动机指标之间的关系。相关结果表明，抑郁情绪得分与贪食症和节食症得分呈正相关。此外，节食与锻炼的自我决定呈负相关。总的来说，研究结果表明，21 项四因子模型在锻炼参与者中显示出有前途的有效性系数，需要进一步研究来调查锻炼者样本中的饮食态度。关键点心理测量学测量的有效性应进行彻底调查。研究人员不应假设在一个样本上进行的量表验证在不同人群中会显示出相同的有效性系数。饮食态度测试是一种常用的量表。本研究表明，修订后的 21 项量表适用于锻炼者。使用饮食态度测试的研究人员应使用节食、口腔控制、食物关注和贪食症的子量表。未来的研究应涉及定性技术，并采访锻炼参与者，以探讨饮食态度的性质。
