Menu

Psychometric properties and factor structure of the Three

Psychometric properties and factor structure of the Three Factor Eating Questionnaire (TFEQ) in obese men and women. from the Swedish Obese (SOS) studyCorrespondence to: J Karlsson, aSUBJECTS: A total of 4377 middle aged, obese subjects in the Swedish Obese (SOS) study.METHODS: The total sample was randomly split into two data subsets and psychometric testing was performed separately in each sample. Multitrait analysis was conducted to test scaling assumptions and factor analysis was used to test the factor structure. of mental well being (MACL, HAD) were used for testing criterion based validity.RESULTS: The Cognitive Restraint factor was consistently reproduced and scaling analysis demonstrated strong item scale discriminant validity, while the item scale convergent validity was unsatisfactory. The internal structure of the Disinhibition scale was weak. Most Disinhibition and Hunger items grouped in one global factor labeled Uncontrolled Eating. A third cluster containing items on Emotional Eating was also identified. The obtained three factor structure was cross validated and replicated across subgroups by gender, age and BMI.CONCLUSION: The original TFEQ factor structure was not replicated. A short, revised 18 item instrument was constructed, representing the derived factors of Cognitive Restraint, Uncontrolled Eating and Emotional Eating. The most efficient items were used to Adidas Iniki Runner Boost BA9994 Unisex Beige Green White Malaysia Online both the convergent and discriminant validity of the scales.eating behavior; obesity; TFEQ; dietary restraint; emotional eating; factor structureThe concept of dietary restraint has played an important role in the study of human eating behavior. Restraint theory proposes that attempts to regulate food intake in order to control body weight and body shape cause episodic overeating and, moreover, the model posits a causal role of frequent dieting in the development of eating disorders and obesity.1 Research along this course is far reaching and has yielded conflicting results. The disinhibitory effect of dietary restraint has been supported by laboratory based research, however, the empirical evidence for a causal link between restraint and obesity is, at this point, still vague. Much attention has been given to the methodological and conceptual difficulties involved in the study of restraint and different views have been widely debated.2,3,4,5 However, controversy still exists concerning the most appropriate way to measure dietary restraint, and further careful evaluation of the reliability and construct validity of the different methods is needed in order to gain conceptual clarity and improve measurement precision.Three self rating questionnaires, the Restraint Scale (RS),6 the Three Factor Eating Questionnaire (TFEQ)7 and the Dutch Eating Behavior Questionnaire (DEBQ),8 have been extensively used in the study of eating behavior and related issues. The RS was developed on rational rather than psychometric grounds and the construct validity of the instrument has been seriously questioned, especially in obesity research.4 To improve the measurement of dietary restraint and to apply the concept to obesity, Stunkard and Messick7 used psychometric techniques to develop the TFEQ. Responses to the RS, the closely related Latent Obesity Questionnaire,9 and newly written items based on clinical observation were factor analyzed. Three factors entitled Cognitive Restraint of Eating, Disinhibition of Eating Control and Susceptibility to Hunger were derived and replicated in several mixed samples of dieters and free eaters. The third instrument, DEBQ, was constructed by Van Strien and co workers to reflect three conceptual models of eating behavior: restraint theory, externality theory and psychosomatic theory.8 Although the RS, TFEQ and DEBQ are conceptually related, these instruments reflect different approaches to the assessment of eating style and, most importantly, several studies have demonstrated essential differences between the RS and the restraint subscales of the TFEQ and the DEBQ.10,11While the methodological obscurities associated with the RS have been discussed in detail, less attention has been given to the construct validity of the alternative measures. Despite the widespread use of the TFEQ in current research, few studies have reported on the psychometric properties of the instrument, and little is known about its validity across diverse populations, eg obese vs normal weight, clinical vs non clinical samples. Ganley12 factor analyzed responses to the 55 item version of the instrument (items on weight fluctuation have been excluded in the final 51 item version) in a population sample of adult women mixed with a small portion of weight watchers. The derived four factor solution confirmed the construct validity of the Cognitive Restraint and Hunger factors, while items assigned to Disinhibition were split into two factors entitled Weight Lability and Emotional Eating. The factor structure of the 51 item TFEQ was tested in a student sample mixed with a small group of weight watchers.13 The stability of the Cognitive Restraint factor was again confirmed, while the Hunger and Disinhibition items grouped in one global factor. A third factor contained three items related to overeating during periods of negative mood (Emotional Eating).These findings cause some concern regarding the factor structure and the factor stability of the TFEQ. suggest that the Cognitive Restraint scale is a robust construct across different samples, while Disinhibition and Hunger are unstable and require further testing. There are also indications that an independent dimension reflecting emotional eating is nested within the Disinhibition scale. However, these validation studies have been carried out in samples of mostly normal weight Nike Air Max 2017 849560-002 Women Blue Grey Factory Outlet Malaysia and it is unclear if the findings are generalizable to other populations, eg obese, male or clinical subjects. Although the TFEQ was especially developed for application in obesity research, no study has yet examined the construct validity of the 51 item version of the instrument in well defined obese samples. In fact, the basic psychometric testing was conducted on earlier versions of the instrument, containing items on weight fluctuation now excluded from the final version.7The main purpose of the present study was to evaluate scaling properties and the construct validity of the TFEQ in large samples of obese subjects. In addition, we aimed at testing if more efficient scales could be constructed by item reduction. It has been demonstrated that lengthy questionnaires can be shortened while still maintaining or enhancing measurement precision and validity by constructing scales from the most efficient items.14 In fact, by deleting unspecific items, scale homogeneity and stability across different samples are likely to be strengthened.The following research questions were addressed:Do the TFEQ scales meet minimum required psychometric standards?Can the factor structure of the TFEQ be replicated in large samples of obese subjects?Which factor solution provides the best fit to the data?Is the factor structure stable across groups differing in sex, age and BMI?Can more efficient scales be constructed by item reduction?Data from 4377 middle aged, obese subjects were analyzed. were recruited by advertisements in the media to take part in the Swedish Obese (SOS) study, an ongoing, large scale study of the health implications of obesity. The project consists of a registry for analysis of cross sectional data and an intervention trial for evaluation of the longitudinal health effects of weight reduction. Obese subjects, aged 37 y, are included in the registry study. The study sample consisted of 1774 men and 2603 women in the registry. A total of 14.5% of the men and 23.Air Jordan 4 KAWS Retro 930155-100 Men White Gray Malaysia Sale% of the women were below the BMI inclusion criteria for the intervention study. According to WHO standards 11 of the subjects were pre obese (BMI 25 478 were obese class I (BMI 30 1994 were obese class II (BMI 35 and 1894 were obese class III (BMI40). All subjects completed a comprehensive series of mail out questionnaires on sociodemographics, weight and medical history, physical activity, sleep patterns, utilization of medical care, and a battery of generic and study specific quality of life instruments (the SOS Quality of Life Survey; Sullivan, Karlsson, Sj and Taft, in press).15 Details of the SOS study design, recruitment and assessment procedures have been reported previously.16,17 A general population group drawn from the western region of Sweden was used for comparisons of mental well being scores.18Three Factor Eating Questionnaire (TFEQ). The TFEQ is a self assessment questionnaire developed to measure cognitive and behavioral components of eating.7 The instrument contains 36 items with a yes response format, 14 items on a 1 response scale and one vertical rating. All item responses are dichotomized and aggregated into three scales. Cognitive Restraint (21 items), Disinhibition (16 items) and Hunger (14 items). Cognitive Restraint is designed to measure dietary restraint, that is, control over food intake in order to influence body weight and body shape. Disinhibition measures episodes of loss of control over eating, while the Hunger scale concerns subjective feelings of hunger and food cravings. The Swedish version of the TFEQ was translated according to standard procedures for cross cultural use.19

Share this Post!

About the Author : Amerson


0 Comment

Related post

  TOP