EMPIRICAL ARTICLE
Development and Validation of the Chinese-Language Version of the Eating Pathology Symptoms Inventory Xiaoqi Tang, BS1,2 Kelsie T. Forbush, PhD3* P. Priscilla Lui, MA1
ABSTRACT Objective: Eating disorders are becoming increasingly prevalent among individuals from non-Western countries, yet few non-English-language measures of eating pathology exist. The current study sought to develop and validate a Chinese version of the Eating Pathology Symptoms Inventory1 with cross-cultural equivalence. Method: The Chinese version of the Eating Pathology Symptoms Inventory (CEPSI) was translated and backtranslated by native Chinese speakers, and administered to a pilot sample of native Chinese speaking students (N 5 45) from a Midwestern university in the United States. The measure was revised based on participant’s feedback, and administrated to a large sample of native Chinese speakers recruited from a Midwestern community (N 5 195; 49.2% women) to test the factor structure and convergent and discriminant validity of the measure. Results: As hypothesized, the CEPSI had a robust eight-factor structure, and demonstrated evidence for acceptable internal consistency (median coefficient
alphas were 0.80 for men and 0.79 for women, and alpha values ranged from 0.36 to 0.85 in men and 0.70 to 0.89 in women), and good convergent validity (correlations with relevant translated scales from the Eating Disorder Examination-Questionnaire and the Eating Attitudes Test-26 ranged from 0.22 to 0.58) and discriminate validity (correlations with a translated version of the Center for Epidemiological Studies – Depression Scale ranged from .12 to .30). Discussion: Results indicate that the CEPSI has high potential value as a new self-report measure of eating pathology that can be used in future research and clinical settings to assess eating disorderrelated psychopathology among Chinese C 2015 Wiley Perispeaking individuals. V odicals, Inc. Keywords: assessment; Chinese; cross-cultural; eating pathology symptoms inventory; EPSI; eating disorders (Int J Eat Disord 2015; 48:1016–1023).
This article was published online on 14 July 2015. An error was subsequently identified. This notice is included in the online and print versions to indicate that both have been corrected 27 August 2015. Accepted 12 April 2015 *Correspondence to: Kelsie Forbush, Ph.D., Department of Psychology, University of Kansas, Lawrence, KS 66045. E-mail: [email protected] 1 Department of Psychological Sciences, Purdue University, West Lafayette, Indiana 2 Department of Psychology, Brandeis University, Waltham, Massachusetts 3 Department of Psychology, University of Kansas, Lawrence, Kansas Published online 14 July 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/eat.22423 C 2015 Wiley Periodicals, Inc. V
ED-related psychopathology is becoming more prevalent in the People’s Republic of China. To ensure culturally competent ED assessment and intervention for Chinese populations, it is important to have measures that are psychometrically sound. Here we report the results of a study designed to develop and validate a Chineselanguage version of the Eating Pathology Symptoms Inventory (EPSI).1,15 Several translated measures of eating pathology are available for use among Chinese-speaking individuals, including the Eating Disorder Diagnostic Scale (EDDS),16 Eating Attitude Test (EAT),17 Eating Disorders Inventory (EDI),18 and Eating Disorder Examination (EDE).19 Although these measures have notable strengths, they also have certain limitations. First, most measures have not demonstrated consistent factor structure replicability or acceptable discriminant validity. For example, although few studies have examined the factor structure of the EDDS, ample evidence indicates that the factor structures of the EAT, EDI, and EDE-
1016
International Journal of Eating Disorders 48:7 1016–1023 2015
Introduction Eating disorders (EDs) are serious psychiatric disorders that are associated with significant medical morbidity,2,3 psychiatric comorbidity,4–8 and mortality.9–13 Although the prevalence of EDs is lower among individuals from non-Western cultures,14
DEVELOPMENT AND VALIDATION OF CEPSI
Questionnaire (EDE-Q) do not replicate well across different samples of English-speakers.15,20–28 Second, many self-report measures of eating pathology (viz., the EAT and EDE) were developed using small samples comprised mostly of young women, and may not accurately represent eating pathology as it presents in men. Finally, most self-report measures of eating pathology used methods that may have inadvertently decreased discriminant validity with depression and anxiety by using terms that tap negative affect (such as “I fear. . .,” or “I’m afraid of. . .,”). To address these limitations, we developed and validated a Chinese version of the Eating Pathology Symptoms Inventory (CEPSI). The EPSI was chosen because it has demonstrated excellent psychometric properties in both men and women, including internal consistency and test–retest reliability, as well as strong evidence for construct validity among native English-speakers.1,29 Based on the results from the English version of the EPSI, we hypothesized that the CEPSI would: (a) demonstrate strong psychometric properties (good-toexcellent convergent validity, discriminant validity, and internal consistency), (b) show a similar eightfactor structure, and (c) replicate the patterns of gender differences for EPSI scale scores among native English-speakers.
Method Procedures Translation and back-translation were undertaken by bilingual researchers who were native Chinese speakers and fluent in English.30 Translators had lived in China for an average (SD) of 18.5 (0.5) years. There are two different Chinese writing systems (Simplified and Traditional Chinese) and two commonly spoken Chinese languages (Mandarin and Cantonese). Simplified Chinese and Mandarin are used in Mainland China, whereas Traditional Chinese and Cantonese are used in Hong Kong and Macau. Although Simplified and Traditional characters can be converted easily, the Western colonization of Hong Kong and Macau may have led to greater similarities among these cultures and Western cultures, relative to Mainland China. To address these linguistic and cultural differences, we developed the CEPSI in both Simplified and Traditional Chinese writing systems. XT translated the CEPSI into Simplified and Traditional Chinese, which was subsequently backtranslated into English by PPL (who was blind to the original English-version of the EPSI). The English and backtranslated versions were compared for discrepancies, which were resolved by discussion.
International Journal of Eating Disorders 48:7 1016–1023 2015
The CEPSI was administered to a pilot sample of Chinese international students recruited from a large Midwestern university from an Introductory Psychology course (Study 1). With informed consent, each participant completed the CEPSI and a demographic questionnaire. Participants were asked to provide verbal feedback during in-person focus groups regarding items they found to be confusing or poorly worded. Focus groups were conducted in Cantonese and/or Mandarin, depending on participant preferences. To address participants’ feedback from focus group sessions, some items were slightly re-phrased to reduce potential confusion. Next, the revised measure was administered online to an independent sample of native Chinese speakers recruited from a Midwestern community. After providing informed consent, Study 2 participants completed the revised CEPSI and other measures online to test convergent and discriminant validity. Validity checks were embedded within the survey, and four participants were removed from the initial sample due to inconsistent responding. Participants Inclusion criteria for both studies included age 18 or older and native Chinese speaker. Exclusion criteria included neurological disorder or intellectual disability. Participants could complete consent forms and study measures in either Simplified or Traditional Chinese. Study 1. Participants (N 5 45; 49% women) had a mean (SD) age of 20.20 (1.79), and ranged from 18-28 years-ofage, and had spent an average (SD) of 1.84 (0.78) years in the United States. Forty-one participants (81.1%) chose Simplified Chinese surveys, whereas four participants (8.90%) completed surveys in Traditional Chinese. Mean (SD) body mass index (BMI) was 21.02 (3.27) kg/m2. Participants reported engaging in the following lifetime ED behaviors: 31.1% fasting to lose weight, 31.1% binge eating, 4.4% self-induce vomiting, and 2.2% laxative misuse. No participants endorsed misusing diuretics. Study 2. Participants (N 5 195; 49.2% women) had a mean (SD) age of 21.7 (3.08), and ranged from 18-33 years-of-age. Most participants (n 5 145; 94.9%) completed surveys in Simplified Chinese, rather than in Traditional Chinese (n 5 10; 5.1%). Mean (SD) body mass index was 22.09 (3.87) kg/m2. Participants engaged in the following lifetime ED behaviors: 31.3% binge eating, 27.8% fasting, 6.7% laxative misuse, 4.1% self-induced vomiting, and 0.5% diuretic misuse. Most participants had completed at least some college education (90.6%), and had spent an average (SD) of 1.77 (0.81) years in the USA. Five participants (2.6%) reported that they were prescribed antidepressant or anxiolytic medications. Measures Lifetime Eating Disorder Behaviors. All participants were asked dichotomous self-report questions (0 5 absent,
1017
TANG ET AL.
1 5 present) about their history of disordered eating behaviors, including binge-eating episodes, fasting, self-induced vomiting, and misuse of laxatives and diuretics. Chinese Version of the Eating Pathology Symptom Inventory (CEPSI). The original, 45-item version of the EPSI is a self-report questionnaire that has a robust eight factorstructure in English-speaking samples,1,29 and is characterized by Body Dissatisfaction, Binge Eating, Cognitive Restraint, Excessive Exercise, Restricting, Purging, Muscle Building, and Negative Attitudes toward Obesity. EPSI data have demonstrated strong convergent and discriminant validity and good test–retest reliability among native English speakers.1,29 Chinese Version of the Eating Attitude Test-26 (CEAT26). The Traditional Chinese version of the EAT-26 contains 26 items that assess Dieting, Bulimia and Food Preoccupation, and Oral Control31,32 Chinese Version of the Eating Disorder ExaminationQuestionnaire (CEDE-Q). The English-version of the EDE interview contains four subscales that assess Eating Concern, Weight Concern, Shape Concern, and Restraint, as well as several additional behavior-specific items to assess concrete disordered eating behaviors.33 After obtaining permission from the authors, we translated the Mandarin interview-version of this scale into a selfreport measure. Because the self-report version of this measure has not undergone a specific validation process, we only administered the more concrete, behavioral items of the EDE-Q, which included five items to assess driven exercise, binge eating, and purging, and the 5item Restraint scale. Chinese Version of the Center for Epidemiological Studies-Depression Scale (CCESD). The Englishlanguage version of the CESD consists 20 items to assess four dimensions of depression: Positive Affect, Depressed Affect, Somatic-Retarded Activity, and Interpersonal Difficulty.34 Statistical Analyses Data were analyzed using SPSS Version 21,35 SAS Version 9,36 and Mplus Version 7.37 Exploratory factor analyses (EFAs) were used to identify factors that could be used to form provisional scales. Oblique and orthogonal rotations were used to evaluate solutions. Items with loadings |.40| on a factor and |.35| on all other factors represented items for inclusion in a candidate scale. Next, we carried out confirmatory factor analyses (CFAs), using the same sample of Study 2 participants, to develop the final measure and to test whether the original eight-factor structure of the EPSI would replicate. A mean- and standard errors-adjusted Robust Weighted Least Squares estimator was used, which is appropriate for ordinal data. Coefficient alpha was computed to iden-
1018
tify the degree of interrelation among items on each final scale. Independent t-tests were used to compare CEPSI scores between men and women. Finally, Pearson’s r was used to test convergent validity with other measures of ED-related psychopathology, and discriminant validity with the CCESD.
Results Study 1
Based on participant feedback, two items were identified as confusing (“I thought my muscles were too small,” and “I don’t like how clothes fit the shape of my body”). We developed two versions of each of these items to determine which item participants in Study 2 would best understand. Eight additional items were slightly reworded to improve clarity. Given that we did not carry out statistical analyses of item content in the pilot study, no items were dropped from the item pool, and 47 items were administered to Study 2 participants. Study 2 EFA and CFA. EFA was carried out on the 47 candidate items. After dropping low- and cross-loading items, 37 items remained. Analyses of these remaining items indicated the presence of eight meaningful factors with a highly similar content, structure, and conceptual meaning to the English-language version of the EPSI (see Table 1). All of the items in the Chinese-version of the EPSI loaded on the same factor as the original English-version. Items that were not retained in the EFA included content related to disliking how clothes fit the shape of one’s body, trying on different outfits because one did not like how he or she looked, thinking one’s muscles were too small, using protein supplements, using or considering using diuretics to lose weight, skipping two meals in a row, and getting full more easily than most people. Given that few participants in the study chose to complete the Traditional Chinese version of the CEPSI, we re-ran the final EFA omitting these individuals. The interpretation of results from the Simplified-only subsample was identical to the combined Traditional-Simplified sample (results available upon request), therefore we report results from the full sample to maximize power. The 37item CEPSI’s eight-factor model demonstrated a good fit to the data in CFA (v2 5 955.45 [601], p < 0.001, CFI 5 0.911, TLI 5 0.901, RMSEA 5 0.057).
As shown in Table 2, the majority of CEPSI scale data demonstrated acceptable-to-good internal consistency. Median
Descriptive Statistics.
International Journal of Eating Disorders 48:7 1016–1023 2015
DEVELOPMENT AND VALIDATION OF CEPSI TABLE 1.
Exploratory factor analysis of chinese language version of the eating pathology symptoms inventory (CEPSI) Body Dissatisfaction
I did not like how my body looked I wished the shape of my body was different I was not satisfied with the size of my hips I did not like the size of my thighs I thought my butt was too big I ate a very large amount of food in a short period of time (e.g., within 2 hours) I stuffed myself with food to the point of feeling sick I ate until I was uncomfortably full If someone offered me food, I felt that I could not resist eating it I ate when I was not hungry I ate as if I was on auto-pilot I snacked throughout the evening without realizing it I did not notice how much I ate until after I had finished eating I pushed myself extremely hard when I exercised I engaged in strenuous exercise at least five days per week I exercised to the point of exhaustion I planned my days around exercising I felt that I needed to exercise nearly every day I tried to exclude “unhealthy” foods from my diet I tried to avoid foods with high calorie content I counted the calories of foods I ate I made myself vomit in order to lose weight I thought laxatives are a good way to lose weight I used diet pills I used diet teas or cleansing teas to lose weight I was disgusted by the sight of obese people I felt that overweight people are lazy I thought that obese people lack self-control I felt that overweight people are unattractive I was disgusted by the sight of an overweight person wearing tight clothes People would be surprised if they knew how little I ate People encouraged me to eat more People told me that I do not eat very much I got full after eating what most people would consider a small amount of food I thought about taking steroids as a way to get more muscular I used muscle building supplements I considered taking a muscle building supplement
Binge Eating
Excessive Exercise
Cognitive Restraint
Purging
Negative Attitudes toward Obesity
Restricting
Muscle Building
0.68 0.56 0.69 0.71 0.79 0.21
0.07 0.04 0.03 0.07 0.07 0.56
20.05 20.03 0.04 20.10 0.04 0.06
0.17 .30 20.18 0.12 20.09 20.19
20.04 20.23 0.24 20.06 0.06 0.13
0.02 0.13 0.07 0.04 20.08 20.02
0.07 20.07 0.08 20.16 0.04 20.06
0.00 0.03 20.10 0.08 20.05 0.10
0.08
0.48
0.02
0.25
0.12
20.02
0.05
20.01
20.07 0.17
0.62 0.60
0.06 0.00
0.12 0.06
0.18 20.03
20.09 20.05
0.03 20.01
20.02 0.08
20.07 0.02 0.00
0.80 0.84 0.70
20.12 0.08 20.10
20.01 20.07 0.04
20.11 0.02 20.11
0.11 20.02 0.00
20.04 0.07 0.03
20.05 20.13 0.01
0.12
0.64
0.12
20.14
0.02
0.03
20.01
20.02
0.09
0.00
0.58
0.11
20.03
0.07
0.03
0.19
20.07
20.01
0.80
0.02
0.04
0.05
20.03
20.07
0.03 20.14 0.08
0.13 20.03 0.00
0.56 0.75 0.59
0.02 0.21 0.10
20.17 0.04 20.10
0.10 20.09 0.03
0.05 20.14 20.09
0.14 0.02 0.01
20.13
0.10
0.05
0.56
0.01
0.04
0.14
20.02
0.11
20.15
0.21
0.71
0.13
0.08
0.05
20.13
0.11 20.06 20.06
20.04 20.03 0.14
0.21 0.08 20.10
0.55 20.11 0.16
0.06 0.62 0.54
20.15 0.03 0.07
0.12 0.10 20.03
20.15 20.07 0.16
0.09 0.17
0.05 20.08
20.06 20.22
0.17 .31
0.63 0.47
0.06 20.04
20.10 20.10
0.06 0.06
20.04 20.06 20.01 0.05 0.12
0.03 0.00 0.04 0.00 20.08
0.02 0.10 20.01 20.06 0.06
20.11 0.13 0.25 20.04 20.10
0.15 0.06 0.03 20.01 20.09
0.76 0.75 0.63 0.82 0.70
0.00 0.05 0.11 20.10 20.01
20.04 0.03 0.05 20.04 20.02
0.20
0.05
0.07
0.22
0.01
20.08
0.52
20.07
20.04 20.07 0.00
0.04 0.00 20.03
20.02 20.17 20.06
0.00 0.12 0.07
20.10 0.00 0.05
0.00 0.12 20.07
0.73 0.75 0.67
0.15 20.10 0.17
20.22
0.26
0.03
0.05
0.09
20.03
20.09
0.50
0.19 20.02
20.12 20.10
0.11 0.05
20.33 20.13
0.06 0.02
0.04 20.05
0.10 0.13
0.55 0.67
Note: Data were rotated using promax. Factor loadings |.40| are bolded. Items were presented in either Simplified or Traditional Chinese Characters to study participants. Data were translated to the original English-version for ease of interpretation in the current table. Items that were dropped from the final factor solution (presented above), included: two items from the Body Dissatisfaction scale (I did not like how clothes fit the shape of my body and I tried on different outfits, because I did not like how I looked), two items from the Muscle Building scale (I thought my muscles were too small and I used protein supplements), two items from the Purging scale (I used diuretics in order to lose weight and I considered taking diuretics to lose weight), and two items from the Restricting scale (I skipped two meals in a row and I got full more easily than most people). C items were reproduced by permission of the copyright holder (Dr. Kelsie Forbush). Further reproThe Eating Pathology Symptoms Inventory (EPSI) V duction of the EPSI and its derivatives (including translated versions) is prohibited without prior permission from Dr. Kelsie Forbush. Please contact [email protected] or visit https://psych.ku.edu/kelsie-t-forbush for more information. Correction made here after initial online publication.
International Journal of Eating Disorders 48:7 1016–1023 2015
1019
TANG ET AL. TABLE 2. Descriptive statistics for the Chinese-language versions of the eating pathology symptoms inventory (EPSI), eating disorder examination questionnaire (EDE-Q), and eating attitudes test-26 (EAT-26) in male and female native Chinese speakers Men Scale CEPSI Body Dissatisfaction Binge Eating Cognitive Restraint Purging Restricting Exercise Negative Attitudes toward Obesity Muscle Building CEDE-Q Restraint Scale Binge Eating Self-Induced Vomiting Laxative Misuse Driven Exercise CEAT-26
Women
Effect Size
M
SD
a
M
SD
a
d
7.92a 5.91a 3.08a 0.20a 4.08a 6.56a 8.94a 0.54a
5.01 4.74 2.41 0.72 3.16 4.22 4.07 1.23
0.85 0.84 0.67 0.36 0.78 0.81 0.81 0.62
9.23a 8.38b 3.88b 1.15b 4.63a 4.49b 8.29a 0.24b
4.77 5.43 2.64 1.73 2.96 3.57 4.73 0.65
0.85 0.89 0.74 0.75 0.76 0.82 0.89 0.70
20.27 20.48 20.32 20.72 20.18 0.53 0.15 0.30
3.53a 0.63a 0.00a 0.00a 2.71a 7.79a
6.14 2.37 0.00 0.00 6.14 5.15
0.81 – – – – 0.69
4.58a 1.17a 0.32a 0.13a 2.04a 10.34b
6.38 3.43 2.15 0.88 4.36 8.66
0.87 – – – – 0.84
0.17 0.18 0.21 0.21 0.13 0.36
Note: Independent t-tests were used to compare scale scores between men (vs.) women. For these comparisons, means not sharing the same subscript within a row differ from one another at p < .05. Cohen’s d for men (vs.) women. CEPSI 5 Chinese Eating Pathology Symptoms Inventory. CEDEQ 5 Chinese Eating Disorder Examination–Questionnaire. CEAT-26 5 Chinese Eating Attitudes Test-26. CESD 5 Chinese Center for Epidemiological StudiesDepression Scale. CEDE-Q Binge Eating, Self-Induced Vomiting, Laxative Misuse, and Driven Exercise were comprised of single items; thus, it is not possible to compute internal consistency for these variables.
coefficient alphas were 0.80 in men and 0.79 in women. However, coefficient alpha for the CEPSI Purging scale was particularly low in men, which may reflect the fact that very few men engaged in purging behaviors in the current sample (lifetime endorsement of purging behaviors in men ranged from 0 to 3.1% and from 1.1 to 10.6% in women). Similar to English-language normative data in the USA,29 native Chinese-speaking women scored significantly higher on Binge Eating, Cognitive Restraint, and Purging, whereas native Chinesespeaking men scored significantly higher on Excessive Exercise and Muscle Building than their opposite sex counterparts. In contrast to norms for native English speakers, native Chinese-speaking women did not score significantly higher than men on Body Dissatisfaction or Restricting, and native Chinese-speaking men did not score significantly higher than women on Negative Attitudes toward Obesity (see Table 2). Convergent and Discriminant Validity. Table 3 provides convergent correlations for CEPSI scales with the CEAT-26 total score, CEDE-Q Restraint scale, and CEDE-Q items measuring laxative misuse, excessive exercise, binge eating, and selfinduced vomiting. It also provides discriminant correlations for the CEPSI scales with the CCESD total score. As hypothesized, CEPSI Cognitive Restraint had strong positive correlations with EDE-Q Restraint and the CEAT-26; CEPSI Binge Eating demonstrated a moderate positive correla-
1020
tion with the CEDE-Q binge-eating item; CEPSI Purging had moderate to strong positive correlations with CEDE-Q laxative misuse and selfinduced vomiting items; and CEPSI Excessive Exercise had a strong positive correlation with the CEDE-Q Driven Exercise item. CEPSI Body Dissatisfaction had moderate correlations with several scales, including CEDE-Q Restraint and Binge Eating, and the CEAT-26. Similar to data obtained from native English-speaking samples, the CEPSI Restricting scale showed evidence for convergent validity with BMI. CEPSI Restricting was significantly negatively correlated with BMI (r 5 2.16, p < 0.05), whereas the EDE-Q Restraint scale was positively correlated with BMI (r 5 .21, p < .01), and the CEAT-26 total score was uncorrelated with BMI (r 5 .10, p > .10). Given that negative attitudes towards others’ bodies and desire for increased muscularity were not assessed by the other eating pathology measures used in this study, it is not surprising that CEPSI Negative Attitudes toward Obesity and Muscle Building demonstrated low-to-small correlations with these measures. Discriminant correlations between CEPSI scales and the CCESD total score were generally lower than correlations between CEPSI scales and other measures of eating pathology, demonstrating evidence for discriminant validity (see Table 3).
International Journal of Eating Disorders 48:7 1016–1023 2015
DEVELOPMENT AND VALIDATION OF CEPSI TABLE 3. Convergent and discriminant validity for the Chinese language version of the eating pathology symptoms inventory (CEPSI) CEPSI Scale CEDE-Q Restraint CEDE-Q Binge Eating CEDE-Q Self-Induced Vomiting CEDE-Q Laxative Misuse CEDE-Q Driven Exercise CEAT-26 CCESD
Body Dissatisfaction
Binge Eating
Cognitive Restraint
Purging
Restricting
Excessive Exercise
Negative attitudes toward Obesity
Muscle Building
0.39** 0.31** 0.13 0.15 0.18* 0.43** 0.25**
0.26** 0.41** 0.19** 0.13 0.19* 0.37** 0.30**
0.58** 0.25** 0.06 0.17* 0.28** 0.56** 0.14
0.26** 0.31** 0.37** 0.49** 0.14 0.41** 0.22*
0.11 20.06 0.09 0.17* 20.01 0.22* 0.20**
0.37** 0.19** 20.04 20.06 0.54** 0.29** 0.12
0.28** 0.19* 0.06 0.02 0.25** 0.34** 0.19**
20.01*** 0.07 20.04 20.04 0.06 0.08 0.20**
Note. Correlations |.30| are in boldface. CEPSI 5 Chinese Eating Pathology Symptoms Inventory. CEDE-Q 5 Chinese Eating Disorder Examination– Questionnaire. CEAT-26 5 Chinese Eating Attitudes Test-26. CCESD 5 Chinese Center for Epidemiological Studies-Depression Scale. Correlations representing medium effect sizes are bolded and correlations representing large effect sizes are also underlined.45,46 *p < .05, two-tailed. **p < .01, two-tailed.
Discussion The purpose of this study was to develop and validate a Chinese-language version of the EPSI. The CEPSI showed a robust 37-item, eight-factor structure that is consistent with previous findings from English-speaking individuals.1,29 As hypothesized, the majority of CEPSI scales demonstrated acceptable internal consistency. The low coefficient alpha for the CEPSI Purging scale in men may be due to the fact that certain compensatory weight-loss methods had extremely low base-rates in the current sample, which may have lowered reliabilities due to range restriction. In fact, no man in Study 2 endorsed CEPSI items pertaining to diuretic misuse, and rates of other purging behaviors were very low. The CEPSI demonstrated evidence for good convergent validity. The CEPSI Body Dissatisfaction, Binge Eating, Cognitive Restraint, Purging, and Excessive Exercise scales demonstrated moderate-to-large correlations with other measures of these same constructs, and CEPSI scales generally had lower correlations with depression than with other ED scales, showing evidence for discriminant validity. Native Chinese-speaking women scored significantly higher than men on CEPSI scales measuring Binge Eating, Cognitive Restraint, and Purging, whereas native Chinesespeaking men scored significantly higher on Excessive Exercise and Muscle Building. Contrary to results among native English-speaking participants,29 women did not score significantly higher than men on Body Dissatisfaction and men did not score significantly higher than women on Negative Attitudes toward Obesity. These discrepancies are consistent with previous research on body dissatisfaction in Asia and the USA, in which Caucasian college students demonstrated larger gender differences for body image and disordered eating than Asian students.39 With regard to the International Journal of Eating Disorders 48:7 1016–1023 2015
lack of gender difference for Negative Attitudes toward Obesity, it appears that people from nonWestern cultures exhibit less negative attitudes and biases toward overweight and obese individuals,40 yet we are unaware of studies that have tested for differences in social attitudes toward overweight and obesity between genders. Among the items that were dropped from the measure, content related to shape dissatisfaction, as manifested by dislike with how clothes fit one’s body, did not load significantly on the Body Dissatisfaction factor. It is possible that these items were not culturally meaningful, as there may not be as much variation in body shape or weight among native Chinese than native English-speaking samples. This means that participants may not have been dressing to hide or highlight individual differences in their body shape. Kawamura et al.41 found that Asian American men and women focused their body dissatisfaction more on their height and eyes than Euro-Americans, and these features are not subject to change by wearing different clothing. Additionally, two items from the English-version of the Muscle Building scale did not load on the translated Chinese-version. These results may reflect differences in body image ideals and the cultural salience of specific methods of muscle building between Chinese and Western men. For example, men in Hong Kong have been found to be significantly more satisfied with their muscularity than their USA counterparts,42 which might explain why “I thought my muscles were too small” did not load on the Muscle Building scale among Chinese participants. It is also possible that desire for muscularity among Asian men in our study was not associated with body dissatisfaction due to cultural differences in their concept of masculinity. Ng, Tan, and Hui’s research on Asian men’s views of masculinity attributes suggested that the majority of participants considered “having a good job” as the 1021
TANG ET AL.
most important factor in determining one’s masculinity.43 The current study provides several unique contributions to the field of EDs. First, this study provided evidence that the CEPSI has potential utility as a measurement tool with desirable psychometric properties that assesses ED symptoms in Chinese-speaking individuals. Second, most validation studies of translated versions of Englishlanguage ED measures have been based on samples of young women, whereas the CEPSI shows initial promise of being a valid and useful measure that can be used among both Chinese men and women. Yet, the present study is not without limitations. First, despite the fact that participants had not resided in the USA for very long, it is possible that the current findings are affected by their level of acculturation to American mainstream culture. A recent national study in Australia found that the longer new immigrants lived in Australia, the more similar their weight-related attitudes and behaviors were to their Australian-born counterparts.44 To better ascertain the effects of acculturation and cross-national differences, future studies should validate the CEPSI in samples of Chinese individuals residing in China. Second, given that the EPSI was developed for use in Western cultures, it is possible that the CEPSI may not contain certain culturally specific manifestations of eating pathology that are more common among Chinese individuals. Third, future research should test the CEPSI among native Chinese-speaking individuals with diagnosable EDs. Finally, more studies are needed to further validate the Traditional version. In summary, the CEPSI may contribute to improved assessment and early detection of EDs in Chinese populations. Given the comparable factorstructure of the CEPSI to the original English version, we believe the EPSI may be particularly useful in epidemiological cross-cultural research settings. Finally, although additional research is needed, the CEPSI may eventually be helpful for screening cases that exceed a certain cut-off score for additional assessment or referral.
References 1. Forbush KT, Wildes JE, Pollack LO, Dunbar D, Luo J, Patterson K, et al. Development and validation of the eating pathology symptoms inventory (EPSI). Psychol Assess 2013;25:859 2. Bulik CM, Sullivan PF, Kendler KS. Medical and psychiatric morbidity in obese women with and without binge eating. Int J Eat Disord 2002;32: 72–78.
1022
3. Bulik CM, Reichborn-Kjennerud T. Medical morbidity in binge eating disorder. Int J Eat Disord 2003;34(Suppl):S39–S46. 4. Fairburn CG, Harrison PJ. Eating disorders. The Lancet 2003;361:407–416. 5. von Lojewski A, Boyd C, Abraham S, Russell J. Lifetime and recent DSM and ICD psychiatric comorbidity of inpatients engaging in different eating disorder behaviours. Eat Weight Disord 2012; 17:e185–e93. 6. B€ uhren K, Schwarte R, Fluck F, Timmesfeld N, Krei M, Egberts K, et al. Comorbid psychiatric disorders in female adolescents with first-onset anorexia nervosa. Eur Eat Disord Rev 2014;22:39–44. 7. Grilo CM, White MA, Masheb RM. DSM-IV psychiatric disorder comorbidity and its correlates in binge eating disorder. Int J Eat Disord 2009; 42:228–234. 8. Gadalla TM. Psychiatric comorbidity in eating disorders: A comparison of men and women. J Men’s Health 2008;5:209–217. 9. Harris EC, Barraclough B. Excess mortality of mental disorder. Br J Psychiatry 1998;173:11–53. 10. Suokas JT, Suvisaari JM, Gissler M, L€ ofman R, Linna MS, Raevuori A, Haukka J. Mortality in eating disorders: A follow-up study of adult eating disorder patients treated in tertiary care, 1995–2010. Psychiatry Res 2013;210:1101– 1106. 11. Franko DL, Keshaviah A, Eddy KT, Krishna M, Davis MC, Keel PK, Herzog DB. A longitudinal investigation of mortality in anorexia nervosa and bulimia nervosa. Am J Psychiatry 2013;170:917–925. 12. Arcelus J, Mitchell AJ, Wales J, Nielsen S. Mortality rates in patients with anorexia nervosa and other eating disorders: A meta-analysis of 36 studies. Arch Gen Psychiatry 2011;68:724–731. 13. Crow SJ, Peterson CB, Swanson SA, Raymond NC, Specker S, Eckert ED, Mitchell JE. Increased mortality in bulimia nervosa and other eating disorders. Am J Psychiatry 2009;166:1342–1346. 14. Makino M, Tsuboi K, Dennerstein L. Prevalence of eating disorders: A comparison of western and non-western countries. Medscape Gen Med 2004;6. 15. Allen KL, Byrne SM, Lampard A, Watson H, Fursland A. Confirmatory factor analysis of the eating disorder examination-questionnaire (EDE-Q). Eat Behav 2011;12:143–151. 16. Lee SW, Stewart SM, Striegel-Moore RH, Lee S, Ho Sy, Lee PW, et al. Validation of the eating disorder diagnostic scale for use with hong kong adolescents. Int J Eat Disord 2007;40:569–574. 17. Lee AM, Lee S. Disordered eating and its psychosocial correlates among chinese adolescent females in hong kong. Int J Eat Disorders 1996;20:177–183. 18. Lee S, Lee AM, Leung T, Yu H. Psychometric properties of the eating disorders invetory (EDI-1) in a nonclinical chinese population in hong kong. Int J Eat Disord 1997;21:187–194. 19. Lau LL, Lee S, Lee E, Wong W. Cross-cultural validity of the eating disorder examination: A study of chinese outpatients with eating disorders in hong kong. J Psychiatry 2006;16:132–136. 20. Clausen L, Rosenvinge JH, Friborg O, Rokkedal K. Validating the eating disorder Inventory-3 (EDI-3): A comparison between 561 female eating disorders patients and 878 females from the general population. J Psychopathol Behav Assess 2011;33:101–110. 21. Engelsen BK, Hagtvet KA. The dimensionality of the 12-item version of the eating attitudes test. Confirmatory factor analyses. Scand J Psychol 1999;40: 293–300. ~a C. Dimensionality 22. Garcıa-Grau E, Fuste A, Mas N, G omez J, Bados A, Saldan of three versions of the eating disorder inventory in adolescent girls. Eur Eat Disord Rev 2010;18:318–327. 23. Hrabosky JI, White MA, Masheb RM, Rothschild BS, Burke-Martindale CH, Grilo CM. Psychometric evaluation of the eating disorder examinationquestionnaire for bariatric surgery candidates. Obesity 2008;16:763–769. 24. Limbert C. The eating disorder inventory: A test of the factor structure and internal consistency in a nonclinical sample. Health Care Women Int 2004; 25:165–178. 25. Mannucci E, Ricca V, Di Bernardo M, Moretti S, Cabras P, Rotella CM. Psychometric properties of EDE 12.0 D in obese adult patients without binge eating disorder. Eat Weight Disord 1997;2:144–149. 26. Peterson CB, Crosby RD, Wonderlich SA, Joiner T, Crow SJ, Mitchell JE, et al. Psychometric properties of the eating disorder examination-questionnaire: Factor structure and internal consistency. Int J Eat Disord 2007;40:386–389.
International Journal of Eating Disorders 48:7 1016–1023 2015
DEVELOPMENT AND VALIDATION OF CEPSI 27. White HJ, Haycraft E, Goodwin H, Meyer C. Eating disorder examination questionnaire: Factor structure for adolescent girls and boys. Int J Eat Disord 2014;47:99–104. 28. Williamson DA, Barker SE, Bertman LJ, Gleaves DH. Body image, body dysphoria, and dietary restraint: Factor structure in nonclinical subjects. Behav Res Ther 1995;33:85–93. 29. Forbush KT, Wildes JE, Hunt TK. Gender norms and validity for the eating pathology symptoms inventory (EPSI). Int J Eat Disord 2014;47:85–91. 30. Bracken BA, Barona A. State of the art procedures for translating, validating and using psychoeducational tests in cross-cultural assessment. School Psychol Int 1991;12:119–132. 31. Lee S, Kwok K, Liau C, Leung T. Screening chinese patients with eating disorders using the eating attitudes test in hong kong. Int J Eat Disord 2002;32: 91–97. 32. Lee AM, Lee S. Disordered eating and its psychosocial correlates among chinese adolescent females in hong kong. Int J Eat Disord 1996;20: 177–183 33. Tong J, Shi J, Wang J, Zhang H, Zhang SF, Wu XY, Hsu L. Validity and reliability of the chinese language version of the eating disorder examination (CEDE) in mainland china: Implications for the identity and nosology of the eating disorders. Int J Eat Disord 2011;44:76–80. 34. Chou Y-J. The structure of the Chinese CES-D scale: An application of confirmatory factor analysis to higher-order factor structure. Recent Dev Appl Quant Methods Soc Sci 1997;1–26. 35. IBM. IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY: IBM Corp., 2011.
International Journal of Eating Disorders 48:7 1016–1023 2015
R 9.2 User’s Guide. Cary, NC: SAS Institute Inc., 2008. 36. SAS Institute Inc. SAS/STATV 37. Muthen LK, Muthen BO. Mplus User’s Guide, 7th ed. Los Angeles, CA: Muthen & Muthen, 1998–2012. 38. Forbush KT, Wildes JE, Pollack LO, Dunbar D, Luo J, Patterson K, et al. Development and validation of the eating pathology symptoms inventory (EPSI). Psychological Assessment 2013;25:859–878. 39. Mintz LB, Kashubeck S. Body image and disordered eating among Asian American and Caucasian college students an examination of race and gender differences. Psychol Women Quarterly 1999;23:781–796. 40. Harper DC. Children’s attitudes to physical differences among youth from western and non-western cultures. Cleft Palate Craniofac J 1995;32:114– 119. 41. Kawamura KY. Body image attitudes of Asian American and Caucasian American women and men. US: ProQuest Information & Learning; 2001. 42. Jung J, Forbes GB, Chan P. Global body and muscle satisfaction among college men in the united states and hong Kong-china. Sex Roles 2010;63:104– 117. 43. Ng CJ, Tan HM, Low WY. What do asian men consider as important masculinity attributes? Findings from the asian men’s attitudes to life events and sexuality (MALES) study. J Men’s Health 2008;5:350–355. 44. Ball K, Kenardy J. Body weight, body image, and eating behaviours: Relationships with ethnicity and acculturation in a community sample of young australian women. Eat Behav 2002;3:205–216. 45. Cohen J. Statistical power analysis for the behavioral sciences. Routledge Academic, 2013. 46. Cohen J. A power primer. Psychol Bull 1992;112:155
1023
Development and Validation of the Chinese-Language Version of the Eating Pathology Symptoms Inventory Xiaoqi Tang, BS1,2 Kelsie T. Forbush, PhD3* P. Priscilla Lui, MA1
ABSTRACT Objective: Eating disorders are becoming increasingly prevalent among individuals from non-Western countries, yet few non-English-language measures of eating pathology exist. The current study sought to develop and validate a Chinese version of the Eating Pathology Symptoms Inventory1 with cross-cultural equivalence. Method: The Chinese version of the Eating Pathology Symptoms Inventory (CEPSI) was translated and backtranslated by native Chinese speakers, and administered to a pilot sample of native Chinese speaking students (N 5 45) from a Midwestern university in the United States. The measure was revised based on participant’s feedback, and administrated to a large sample of native Chinese speakers recruited from a Midwestern community (N 5 195; 49.2% women) to test the factor structure and convergent and discriminant validity of the measure. Results: As hypothesized, the CEPSI had a robust eight-factor structure, and demonstrated evidence for acceptable internal consistency (median coefficient
alphas were 0.80 for men and 0.79 for women, and alpha values ranged from 0.36 to 0.85 in men and 0.70 to 0.89 in women), and good convergent validity (correlations with relevant translated scales from the Eating Disorder Examination-Questionnaire and the Eating Attitudes Test-26 ranged from 0.22 to 0.58) and discriminate validity (correlations with a translated version of the Center for Epidemiological Studies – Depression Scale ranged from .12 to .30). Discussion: Results indicate that the CEPSI has high potential value as a new self-report measure of eating pathology that can be used in future research and clinical settings to assess eating disorderrelated psychopathology among Chinese C 2015 Wiley Perispeaking individuals. V odicals, Inc. Keywords: assessment; Chinese; cross-cultural; eating pathology symptoms inventory; EPSI; eating disorders (Int J Eat Disord 2015; 48:1016–1023).
This article was published online on 14 July 2015. An error was subsequently identified. This notice is included in the online and print versions to indicate that both have been corrected 27 August 2015. Accepted 12 April 2015 *Correspondence to: Kelsie Forbush, Ph.D., Department of Psychology, University of Kansas, Lawrence, KS 66045. E-mail: [email protected] 1 Department of Psychological Sciences, Purdue University, West Lafayette, Indiana 2 Department of Psychology, Brandeis University, Waltham, Massachusetts 3 Department of Psychology, University of Kansas, Lawrence, Kansas Published online 14 July 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/eat.22423 C 2015 Wiley Periodicals, Inc. V
ED-related psychopathology is becoming more prevalent in the People’s Republic of China. To ensure culturally competent ED assessment and intervention for Chinese populations, it is important to have measures that are psychometrically sound. Here we report the results of a study designed to develop and validate a Chineselanguage version of the Eating Pathology Symptoms Inventory (EPSI).1,15 Several translated measures of eating pathology are available for use among Chinese-speaking individuals, including the Eating Disorder Diagnostic Scale (EDDS),16 Eating Attitude Test (EAT),17 Eating Disorders Inventory (EDI),18 and Eating Disorder Examination (EDE).19 Although these measures have notable strengths, they also have certain limitations. First, most measures have not demonstrated consistent factor structure replicability or acceptable discriminant validity. For example, although few studies have examined the factor structure of the EDDS, ample evidence indicates that the factor structures of the EAT, EDI, and EDE-
1016
International Journal of Eating Disorders 48:7 1016–1023 2015
Introduction Eating disorders (EDs) are serious psychiatric disorders that are associated with significant medical morbidity,2,3 psychiatric comorbidity,4–8 and mortality.9–13 Although the prevalence of EDs is lower among individuals from non-Western cultures,14
DEVELOPMENT AND VALIDATION OF CEPSI
Questionnaire (EDE-Q) do not replicate well across different samples of English-speakers.15,20–28 Second, many self-report measures of eating pathology (viz., the EAT and EDE) were developed using small samples comprised mostly of young women, and may not accurately represent eating pathology as it presents in men. Finally, most self-report measures of eating pathology used methods that may have inadvertently decreased discriminant validity with depression and anxiety by using terms that tap negative affect (such as “I fear. . .,” or “I’m afraid of. . .,”). To address these limitations, we developed and validated a Chinese version of the Eating Pathology Symptoms Inventory (CEPSI). The EPSI was chosen because it has demonstrated excellent psychometric properties in both men and women, including internal consistency and test–retest reliability, as well as strong evidence for construct validity among native English-speakers.1,29 Based on the results from the English version of the EPSI, we hypothesized that the CEPSI would: (a) demonstrate strong psychometric properties (good-toexcellent convergent validity, discriminant validity, and internal consistency), (b) show a similar eightfactor structure, and (c) replicate the patterns of gender differences for EPSI scale scores among native English-speakers.
Method Procedures Translation and back-translation were undertaken by bilingual researchers who were native Chinese speakers and fluent in English.30 Translators had lived in China for an average (SD) of 18.5 (0.5) years. There are two different Chinese writing systems (Simplified and Traditional Chinese) and two commonly spoken Chinese languages (Mandarin and Cantonese). Simplified Chinese and Mandarin are used in Mainland China, whereas Traditional Chinese and Cantonese are used in Hong Kong and Macau. Although Simplified and Traditional characters can be converted easily, the Western colonization of Hong Kong and Macau may have led to greater similarities among these cultures and Western cultures, relative to Mainland China. To address these linguistic and cultural differences, we developed the CEPSI in both Simplified and Traditional Chinese writing systems. XT translated the CEPSI into Simplified and Traditional Chinese, which was subsequently backtranslated into English by PPL (who was blind to the original English-version of the EPSI). The English and backtranslated versions were compared for discrepancies, which were resolved by discussion.
International Journal of Eating Disorders 48:7 1016–1023 2015
The CEPSI was administered to a pilot sample of Chinese international students recruited from a large Midwestern university from an Introductory Psychology course (Study 1). With informed consent, each participant completed the CEPSI and a demographic questionnaire. Participants were asked to provide verbal feedback during in-person focus groups regarding items they found to be confusing or poorly worded. Focus groups were conducted in Cantonese and/or Mandarin, depending on participant preferences. To address participants’ feedback from focus group sessions, some items were slightly re-phrased to reduce potential confusion. Next, the revised measure was administered online to an independent sample of native Chinese speakers recruited from a Midwestern community. After providing informed consent, Study 2 participants completed the revised CEPSI and other measures online to test convergent and discriminant validity. Validity checks were embedded within the survey, and four participants were removed from the initial sample due to inconsistent responding. Participants Inclusion criteria for both studies included age 18 or older and native Chinese speaker. Exclusion criteria included neurological disorder or intellectual disability. Participants could complete consent forms and study measures in either Simplified or Traditional Chinese. Study 1. Participants (N 5 45; 49% women) had a mean (SD) age of 20.20 (1.79), and ranged from 18-28 years-ofage, and had spent an average (SD) of 1.84 (0.78) years in the United States. Forty-one participants (81.1%) chose Simplified Chinese surveys, whereas four participants (8.90%) completed surveys in Traditional Chinese. Mean (SD) body mass index (BMI) was 21.02 (3.27) kg/m2. Participants reported engaging in the following lifetime ED behaviors: 31.1% fasting to lose weight, 31.1% binge eating, 4.4% self-induce vomiting, and 2.2% laxative misuse. No participants endorsed misusing diuretics. Study 2. Participants (N 5 195; 49.2% women) had a mean (SD) age of 21.7 (3.08), and ranged from 18-33 years-of-age. Most participants (n 5 145; 94.9%) completed surveys in Simplified Chinese, rather than in Traditional Chinese (n 5 10; 5.1%). Mean (SD) body mass index was 22.09 (3.87) kg/m2. Participants engaged in the following lifetime ED behaviors: 31.3% binge eating, 27.8% fasting, 6.7% laxative misuse, 4.1% self-induced vomiting, and 0.5% diuretic misuse. Most participants had completed at least some college education (90.6%), and had spent an average (SD) of 1.77 (0.81) years in the USA. Five participants (2.6%) reported that they were prescribed antidepressant or anxiolytic medications. Measures Lifetime Eating Disorder Behaviors. All participants were asked dichotomous self-report questions (0 5 absent,
1017
TANG ET AL.
1 5 present) about their history of disordered eating behaviors, including binge-eating episodes, fasting, self-induced vomiting, and misuse of laxatives and diuretics. Chinese Version of the Eating Pathology Symptom Inventory (CEPSI). The original, 45-item version of the EPSI is a self-report questionnaire that has a robust eight factorstructure in English-speaking samples,1,29 and is characterized by Body Dissatisfaction, Binge Eating, Cognitive Restraint, Excessive Exercise, Restricting, Purging, Muscle Building, and Negative Attitudes toward Obesity. EPSI data have demonstrated strong convergent and discriminant validity and good test–retest reliability among native English speakers.1,29 Chinese Version of the Eating Attitude Test-26 (CEAT26). The Traditional Chinese version of the EAT-26 contains 26 items that assess Dieting, Bulimia and Food Preoccupation, and Oral Control31,32 Chinese Version of the Eating Disorder ExaminationQuestionnaire (CEDE-Q). The English-version of the EDE interview contains four subscales that assess Eating Concern, Weight Concern, Shape Concern, and Restraint, as well as several additional behavior-specific items to assess concrete disordered eating behaviors.33 After obtaining permission from the authors, we translated the Mandarin interview-version of this scale into a selfreport measure. Because the self-report version of this measure has not undergone a specific validation process, we only administered the more concrete, behavioral items of the EDE-Q, which included five items to assess driven exercise, binge eating, and purging, and the 5item Restraint scale. Chinese Version of the Center for Epidemiological Studies-Depression Scale (CCESD). The Englishlanguage version of the CESD consists 20 items to assess four dimensions of depression: Positive Affect, Depressed Affect, Somatic-Retarded Activity, and Interpersonal Difficulty.34 Statistical Analyses Data were analyzed using SPSS Version 21,35 SAS Version 9,36 and Mplus Version 7.37 Exploratory factor analyses (EFAs) were used to identify factors that could be used to form provisional scales. Oblique and orthogonal rotations were used to evaluate solutions. Items with loadings |.40| on a factor and |.35| on all other factors represented items for inclusion in a candidate scale. Next, we carried out confirmatory factor analyses (CFAs), using the same sample of Study 2 participants, to develop the final measure and to test whether the original eight-factor structure of the EPSI would replicate. A mean- and standard errors-adjusted Robust Weighted Least Squares estimator was used, which is appropriate for ordinal data. Coefficient alpha was computed to iden-
1018
tify the degree of interrelation among items on each final scale. Independent t-tests were used to compare CEPSI scores between men and women. Finally, Pearson’s r was used to test convergent validity with other measures of ED-related psychopathology, and discriminant validity with the CCESD.
Results Study 1
Based on participant feedback, two items were identified as confusing (“I thought my muscles were too small,” and “I don’t like how clothes fit the shape of my body”). We developed two versions of each of these items to determine which item participants in Study 2 would best understand. Eight additional items were slightly reworded to improve clarity. Given that we did not carry out statistical analyses of item content in the pilot study, no items were dropped from the item pool, and 47 items were administered to Study 2 participants. Study 2 EFA and CFA. EFA was carried out on the 47 candidate items. After dropping low- and cross-loading items, 37 items remained. Analyses of these remaining items indicated the presence of eight meaningful factors with a highly similar content, structure, and conceptual meaning to the English-language version of the EPSI (see Table 1). All of the items in the Chinese-version of the EPSI loaded on the same factor as the original English-version. Items that were not retained in the EFA included content related to disliking how clothes fit the shape of one’s body, trying on different outfits because one did not like how he or she looked, thinking one’s muscles were too small, using protein supplements, using or considering using diuretics to lose weight, skipping two meals in a row, and getting full more easily than most people. Given that few participants in the study chose to complete the Traditional Chinese version of the CEPSI, we re-ran the final EFA omitting these individuals. The interpretation of results from the Simplified-only subsample was identical to the combined Traditional-Simplified sample (results available upon request), therefore we report results from the full sample to maximize power. The 37item CEPSI’s eight-factor model demonstrated a good fit to the data in CFA (v2 5 955.45 [601], p < 0.001, CFI 5 0.911, TLI 5 0.901, RMSEA 5 0.057).
As shown in Table 2, the majority of CEPSI scale data demonstrated acceptable-to-good internal consistency. Median
Descriptive Statistics.
International Journal of Eating Disorders 48:7 1016–1023 2015
DEVELOPMENT AND VALIDATION OF CEPSI TABLE 1.
Exploratory factor analysis of chinese language version of the eating pathology symptoms inventory (CEPSI) Body Dissatisfaction
I did not like how my body looked I wished the shape of my body was different I was not satisfied with the size of my hips I did not like the size of my thighs I thought my butt was too big I ate a very large amount of food in a short period of time (e.g., within 2 hours) I stuffed myself with food to the point of feeling sick I ate until I was uncomfortably full If someone offered me food, I felt that I could not resist eating it I ate when I was not hungry I ate as if I was on auto-pilot I snacked throughout the evening without realizing it I did not notice how much I ate until after I had finished eating I pushed myself extremely hard when I exercised I engaged in strenuous exercise at least five days per week I exercised to the point of exhaustion I planned my days around exercising I felt that I needed to exercise nearly every day I tried to exclude “unhealthy” foods from my diet I tried to avoid foods with high calorie content I counted the calories of foods I ate I made myself vomit in order to lose weight I thought laxatives are a good way to lose weight I used diet pills I used diet teas or cleansing teas to lose weight I was disgusted by the sight of obese people I felt that overweight people are lazy I thought that obese people lack self-control I felt that overweight people are unattractive I was disgusted by the sight of an overweight person wearing tight clothes People would be surprised if they knew how little I ate People encouraged me to eat more People told me that I do not eat very much I got full after eating what most people would consider a small amount of food I thought about taking steroids as a way to get more muscular I used muscle building supplements I considered taking a muscle building supplement
Binge Eating
Excessive Exercise
Cognitive Restraint
Purging
Negative Attitudes toward Obesity
Restricting
Muscle Building
0.68 0.56 0.69 0.71 0.79 0.21
0.07 0.04 0.03 0.07 0.07 0.56
20.05 20.03 0.04 20.10 0.04 0.06
0.17 .30 20.18 0.12 20.09 20.19
20.04 20.23 0.24 20.06 0.06 0.13
0.02 0.13 0.07 0.04 20.08 20.02
0.07 20.07 0.08 20.16 0.04 20.06
0.00 0.03 20.10 0.08 20.05 0.10
0.08
0.48
0.02
0.25
0.12
20.02
0.05
20.01
20.07 0.17
0.62 0.60
0.06 0.00
0.12 0.06
0.18 20.03
20.09 20.05
0.03 20.01
20.02 0.08
20.07 0.02 0.00
0.80 0.84 0.70
20.12 0.08 20.10
20.01 20.07 0.04
20.11 0.02 20.11
0.11 20.02 0.00
20.04 0.07 0.03
20.05 20.13 0.01
0.12
0.64
0.12
20.14
0.02
0.03
20.01
20.02
0.09
0.00
0.58
0.11
20.03
0.07
0.03
0.19
20.07
20.01
0.80
0.02
0.04
0.05
20.03
20.07
0.03 20.14 0.08
0.13 20.03 0.00
0.56 0.75 0.59
0.02 0.21 0.10
20.17 0.04 20.10
0.10 20.09 0.03
0.05 20.14 20.09
0.14 0.02 0.01
20.13
0.10
0.05
0.56
0.01
0.04
0.14
20.02
0.11
20.15
0.21
0.71
0.13
0.08
0.05
20.13
0.11 20.06 20.06
20.04 20.03 0.14
0.21 0.08 20.10
0.55 20.11 0.16
0.06 0.62 0.54
20.15 0.03 0.07
0.12 0.10 20.03
20.15 20.07 0.16
0.09 0.17
0.05 20.08
20.06 20.22
0.17 .31
0.63 0.47
0.06 20.04
20.10 20.10
0.06 0.06
20.04 20.06 20.01 0.05 0.12
0.03 0.00 0.04 0.00 20.08
0.02 0.10 20.01 20.06 0.06
20.11 0.13 0.25 20.04 20.10
0.15 0.06 0.03 20.01 20.09
0.76 0.75 0.63 0.82 0.70
0.00 0.05 0.11 20.10 20.01
20.04 0.03 0.05 20.04 20.02
0.20
0.05
0.07
0.22
0.01
20.08
0.52
20.07
20.04 20.07 0.00
0.04 0.00 20.03
20.02 20.17 20.06
0.00 0.12 0.07
20.10 0.00 0.05
0.00 0.12 20.07
0.73 0.75 0.67
0.15 20.10 0.17
20.22
0.26
0.03
0.05
0.09
20.03
20.09
0.50
0.19 20.02
20.12 20.10
0.11 0.05
20.33 20.13
0.06 0.02
0.04 20.05
0.10 0.13
0.55 0.67
Note: Data were rotated using promax. Factor loadings |.40| are bolded. Items were presented in either Simplified or Traditional Chinese Characters to study participants. Data were translated to the original English-version for ease of interpretation in the current table. Items that were dropped from the final factor solution (presented above), included: two items from the Body Dissatisfaction scale (I did not like how clothes fit the shape of my body and I tried on different outfits, because I did not like how I looked), two items from the Muscle Building scale (I thought my muscles were too small and I used protein supplements), two items from the Purging scale (I used diuretics in order to lose weight and I considered taking diuretics to lose weight), and two items from the Restricting scale (I skipped two meals in a row and I got full more easily than most people). C items were reproduced by permission of the copyright holder (Dr. Kelsie Forbush). Further reproThe Eating Pathology Symptoms Inventory (EPSI) V duction of the EPSI and its derivatives (including translated versions) is prohibited without prior permission from Dr. Kelsie Forbush. Please contact [email protected] or visit https://psych.ku.edu/kelsie-t-forbush for more information. Correction made here after initial online publication.
International Journal of Eating Disorders 48:7 1016–1023 2015
1019
TANG ET AL. TABLE 2. Descriptive statistics for the Chinese-language versions of the eating pathology symptoms inventory (EPSI), eating disorder examination questionnaire (EDE-Q), and eating attitudes test-26 (EAT-26) in male and female native Chinese speakers Men Scale CEPSI Body Dissatisfaction Binge Eating Cognitive Restraint Purging Restricting Exercise Negative Attitudes toward Obesity Muscle Building CEDE-Q Restraint Scale Binge Eating Self-Induced Vomiting Laxative Misuse Driven Exercise CEAT-26
Women
Effect Size
M
SD
a
M
SD
a
d
7.92a 5.91a 3.08a 0.20a 4.08a 6.56a 8.94a 0.54a
5.01 4.74 2.41 0.72 3.16 4.22 4.07 1.23
0.85 0.84 0.67 0.36 0.78 0.81 0.81 0.62
9.23a 8.38b 3.88b 1.15b 4.63a 4.49b 8.29a 0.24b
4.77 5.43 2.64 1.73 2.96 3.57 4.73 0.65
0.85 0.89 0.74 0.75 0.76 0.82 0.89 0.70
20.27 20.48 20.32 20.72 20.18 0.53 0.15 0.30
3.53a 0.63a 0.00a 0.00a 2.71a 7.79a
6.14 2.37 0.00 0.00 6.14 5.15
0.81 – – – – 0.69
4.58a 1.17a 0.32a 0.13a 2.04a 10.34b
6.38 3.43 2.15 0.88 4.36 8.66
0.87 – – – – 0.84
0.17 0.18 0.21 0.21 0.13 0.36
Note: Independent t-tests were used to compare scale scores between men (vs.) women. For these comparisons, means not sharing the same subscript within a row differ from one another at p < .05. Cohen’s d for men (vs.) women. CEPSI 5 Chinese Eating Pathology Symptoms Inventory. CEDEQ 5 Chinese Eating Disorder Examination–Questionnaire. CEAT-26 5 Chinese Eating Attitudes Test-26. CESD 5 Chinese Center for Epidemiological StudiesDepression Scale. CEDE-Q Binge Eating, Self-Induced Vomiting, Laxative Misuse, and Driven Exercise were comprised of single items; thus, it is not possible to compute internal consistency for these variables.
coefficient alphas were 0.80 in men and 0.79 in women. However, coefficient alpha for the CEPSI Purging scale was particularly low in men, which may reflect the fact that very few men engaged in purging behaviors in the current sample (lifetime endorsement of purging behaviors in men ranged from 0 to 3.1% and from 1.1 to 10.6% in women). Similar to English-language normative data in the USA,29 native Chinese-speaking women scored significantly higher on Binge Eating, Cognitive Restraint, and Purging, whereas native Chinesespeaking men scored significantly higher on Excessive Exercise and Muscle Building than their opposite sex counterparts. In contrast to norms for native English speakers, native Chinese-speaking women did not score significantly higher than men on Body Dissatisfaction or Restricting, and native Chinese-speaking men did not score significantly higher than women on Negative Attitudes toward Obesity (see Table 2). Convergent and Discriminant Validity. Table 3 provides convergent correlations for CEPSI scales with the CEAT-26 total score, CEDE-Q Restraint scale, and CEDE-Q items measuring laxative misuse, excessive exercise, binge eating, and selfinduced vomiting. It also provides discriminant correlations for the CEPSI scales with the CCESD total score. As hypothesized, CEPSI Cognitive Restraint had strong positive correlations with EDE-Q Restraint and the CEAT-26; CEPSI Binge Eating demonstrated a moderate positive correla-
1020
tion with the CEDE-Q binge-eating item; CEPSI Purging had moderate to strong positive correlations with CEDE-Q laxative misuse and selfinduced vomiting items; and CEPSI Excessive Exercise had a strong positive correlation with the CEDE-Q Driven Exercise item. CEPSI Body Dissatisfaction had moderate correlations with several scales, including CEDE-Q Restraint and Binge Eating, and the CEAT-26. Similar to data obtained from native English-speaking samples, the CEPSI Restricting scale showed evidence for convergent validity with BMI. CEPSI Restricting was significantly negatively correlated with BMI (r 5 2.16, p < 0.05), whereas the EDE-Q Restraint scale was positively correlated with BMI (r 5 .21, p < .01), and the CEAT-26 total score was uncorrelated with BMI (r 5 .10, p > .10). Given that negative attitudes towards others’ bodies and desire for increased muscularity were not assessed by the other eating pathology measures used in this study, it is not surprising that CEPSI Negative Attitudes toward Obesity and Muscle Building demonstrated low-to-small correlations with these measures. Discriminant correlations between CEPSI scales and the CCESD total score were generally lower than correlations between CEPSI scales and other measures of eating pathology, demonstrating evidence for discriminant validity (see Table 3).
International Journal of Eating Disorders 48:7 1016–1023 2015
DEVELOPMENT AND VALIDATION OF CEPSI TABLE 3. Convergent and discriminant validity for the Chinese language version of the eating pathology symptoms inventory (CEPSI) CEPSI Scale CEDE-Q Restraint CEDE-Q Binge Eating CEDE-Q Self-Induced Vomiting CEDE-Q Laxative Misuse CEDE-Q Driven Exercise CEAT-26 CCESD
Body Dissatisfaction
Binge Eating
Cognitive Restraint
Purging
Restricting
Excessive Exercise
Negative attitudes toward Obesity
Muscle Building
0.39** 0.31** 0.13 0.15 0.18* 0.43** 0.25**
0.26** 0.41** 0.19** 0.13 0.19* 0.37** 0.30**
0.58** 0.25** 0.06 0.17* 0.28** 0.56** 0.14
0.26** 0.31** 0.37** 0.49** 0.14 0.41** 0.22*
0.11 20.06 0.09 0.17* 20.01 0.22* 0.20**
0.37** 0.19** 20.04 20.06 0.54** 0.29** 0.12
0.28** 0.19* 0.06 0.02 0.25** 0.34** 0.19**
20.01*** 0.07 20.04 20.04 0.06 0.08 0.20**
Note. Correlations |.30| are in boldface. CEPSI 5 Chinese Eating Pathology Symptoms Inventory. CEDE-Q 5 Chinese Eating Disorder Examination– Questionnaire. CEAT-26 5 Chinese Eating Attitudes Test-26. CCESD 5 Chinese Center for Epidemiological Studies-Depression Scale. Correlations representing medium effect sizes are bolded and correlations representing large effect sizes are also underlined.45,46 *p < .05, two-tailed. **p < .01, two-tailed.
Discussion The purpose of this study was to develop and validate a Chinese-language version of the EPSI. The CEPSI showed a robust 37-item, eight-factor structure that is consistent with previous findings from English-speaking individuals.1,29 As hypothesized, the majority of CEPSI scales demonstrated acceptable internal consistency. The low coefficient alpha for the CEPSI Purging scale in men may be due to the fact that certain compensatory weight-loss methods had extremely low base-rates in the current sample, which may have lowered reliabilities due to range restriction. In fact, no man in Study 2 endorsed CEPSI items pertaining to diuretic misuse, and rates of other purging behaviors were very low. The CEPSI demonstrated evidence for good convergent validity. The CEPSI Body Dissatisfaction, Binge Eating, Cognitive Restraint, Purging, and Excessive Exercise scales demonstrated moderate-to-large correlations with other measures of these same constructs, and CEPSI scales generally had lower correlations with depression than with other ED scales, showing evidence for discriminant validity. Native Chinese-speaking women scored significantly higher than men on CEPSI scales measuring Binge Eating, Cognitive Restraint, and Purging, whereas native Chinesespeaking men scored significantly higher on Excessive Exercise and Muscle Building. Contrary to results among native English-speaking participants,29 women did not score significantly higher than men on Body Dissatisfaction and men did not score significantly higher than women on Negative Attitudes toward Obesity. These discrepancies are consistent with previous research on body dissatisfaction in Asia and the USA, in which Caucasian college students demonstrated larger gender differences for body image and disordered eating than Asian students.39 With regard to the International Journal of Eating Disorders 48:7 1016–1023 2015
lack of gender difference for Negative Attitudes toward Obesity, it appears that people from nonWestern cultures exhibit less negative attitudes and biases toward overweight and obese individuals,40 yet we are unaware of studies that have tested for differences in social attitudes toward overweight and obesity between genders. Among the items that were dropped from the measure, content related to shape dissatisfaction, as manifested by dislike with how clothes fit one’s body, did not load significantly on the Body Dissatisfaction factor. It is possible that these items were not culturally meaningful, as there may not be as much variation in body shape or weight among native Chinese than native English-speaking samples. This means that participants may not have been dressing to hide or highlight individual differences in their body shape. Kawamura et al.41 found that Asian American men and women focused their body dissatisfaction more on their height and eyes than Euro-Americans, and these features are not subject to change by wearing different clothing. Additionally, two items from the English-version of the Muscle Building scale did not load on the translated Chinese-version. These results may reflect differences in body image ideals and the cultural salience of specific methods of muscle building between Chinese and Western men. For example, men in Hong Kong have been found to be significantly more satisfied with their muscularity than their USA counterparts,42 which might explain why “I thought my muscles were too small” did not load on the Muscle Building scale among Chinese participants. It is also possible that desire for muscularity among Asian men in our study was not associated with body dissatisfaction due to cultural differences in their concept of masculinity. Ng, Tan, and Hui’s research on Asian men’s views of masculinity attributes suggested that the majority of participants considered “having a good job” as the 1021
TANG ET AL.
most important factor in determining one’s masculinity.43 The current study provides several unique contributions to the field of EDs. First, this study provided evidence that the CEPSI has potential utility as a measurement tool with desirable psychometric properties that assesses ED symptoms in Chinese-speaking individuals. Second, most validation studies of translated versions of Englishlanguage ED measures have been based on samples of young women, whereas the CEPSI shows initial promise of being a valid and useful measure that can be used among both Chinese men and women. Yet, the present study is not without limitations. First, despite the fact that participants had not resided in the USA for very long, it is possible that the current findings are affected by their level of acculturation to American mainstream culture. A recent national study in Australia found that the longer new immigrants lived in Australia, the more similar their weight-related attitudes and behaviors were to their Australian-born counterparts.44 To better ascertain the effects of acculturation and cross-national differences, future studies should validate the CEPSI in samples of Chinese individuals residing in China. Second, given that the EPSI was developed for use in Western cultures, it is possible that the CEPSI may not contain certain culturally specific manifestations of eating pathology that are more common among Chinese individuals. Third, future research should test the CEPSI among native Chinese-speaking individuals with diagnosable EDs. Finally, more studies are needed to further validate the Traditional version. In summary, the CEPSI may contribute to improved assessment and early detection of EDs in Chinese populations. Given the comparable factorstructure of the CEPSI to the original English version, we believe the EPSI may be particularly useful in epidemiological cross-cultural research settings. Finally, although additional research is needed, the CEPSI may eventually be helpful for screening cases that exceed a certain cut-off score for additional assessment or referral.
References 1. Forbush KT, Wildes JE, Pollack LO, Dunbar D, Luo J, Patterson K, et al. Development and validation of the eating pathology symptoms inventory (EPSI). Psychol Assess 2013;25:859 2. Bulik CM, Sullivan PF, Kendler KS. Medical and psychiatric morbidity in obese women with and without binge eating. Int J Eat Disord 2002;32: 72–78.
1022
3. Bulik CM, Reichborn-Kjennerud T. Medical morbidity in binge eating disorder. Int J Eat Disord 2003;34(Suppl):S39–S46. 4. Fairburn CG, Harrison PJ. Eating disorders. The Lancet 2003;361:407–416. 5. von Lojewski A, Boyd C, Abraham S, Russell J. Lifetime and recent DSM and ICD psychiatric comorbidity of inpatients engaging in different eating disorder behaviours. Eat Weight Disord 2012; 17:e185–e93. 6. B€ uhren K, Schwarte R, Fluck F, Timmesfeld N, Krei M, Egberts K, et al. Comorbid psychiatric disorders in female adolescents with first-onset anorexia nervosa. Eur Eat Disord Rev 2014;22:39–44. 7. Grilo CM, White MA, Masheb RM. DSM-IV psychiatric disorder comorbidity and its correlates in binge eating disorder. Int J Eat Disord 2009; 42:228–234. 8. Gadalla TM. Psychiatric comorbidity in eating disorders: A comparison of men and women. J Men’s Health 2008;5:209–217. 9. Harris EC, Barraclough B. Excess mortality of mental disorder. Br J Psychiatry 1998;173:11–53. 10. Suokas JT, Suvisaari JM, Gissler M, L€ ofman R, Linna MS, Raevuori A, Haukka J. Mortality in eating disorders: A follow-up study of adult eating disorder patients treated in tertiary care, 1995–2010. Psychiatry Res 2013;210:1101– 1106. 11. Franko DL, Keshaviah A, Eddy KT, Krishna M, Davis MC, Keel PK, Herzog DB. A longitudinal investigation of mortality in anorexia nervosa and bulimia nervosa. Am J Psychiatry 2013;170:917–925. 12. Arcelus J, Mitchell AJ, Wales J, Nielsen S. Mortality rates in patients with anorexia nervosa and other eating disorders: A meta-analysis of 36 studies. Arch Gen Psychiatry 2011;68:724–731. 13. Crow SJ, Peterson CB, Swanson SA, Raymond NC, Specker S, Eckert ED, Mitchell JE. Increased mortality in bulimia nervosa and other eating disorders. Am J Psychiatry 2009;166:1342–1346. 14. Makino M, Tsuboi K, Dennerstein L. Prevalence of eating disorders: A comparison of western and non-western countries. Medscape Gen Med 2004;6. 15. Allen KL, Byrne SM, Lampard A, Watson H, Fursland A. Confirmatory factor analysis of the eating disorder examination-questionnaire (EDE-Q). Eat Behav 2011;12:143–151. 16. Lee SW, Stewart SM, Striegel-Moore RH, Lee S, Ho Sy, Lee PW, et al. Validation of the eating disorder diagnostic scale for use with hong kong adolescents. Int J Eat Disord 2007;40:569–574. 17. Lee AM, Lee S. Disordered eating and its psychosocial correlates among chinese adolescent females in hong kong. Int J Eat Disorders 1996;20:177–183. 18. Lee S, Lee AM, Leung T, Yu H. Psychometric properties of the eating disorders invetory (EDI-1) in a nonclinical chinese population in hong kong. Int J Eat Disord 1997;21:187–194. 19. Lau LL, Lee S, Lee E, Wong W. Cross-cultural validity of the eating disorder examination: A study of chinese outpatients with eating disorders in hong kong. J Psychiatry 2006;16:132–136. 20. Clausen L, Rosenvinge JH, Friborg O, Rokkedal K. Validating the eating disorder Inventory-3 (EDI-3): A comparison between 561 female eating disorders patients and 878 females from the general population. J Psychopathol Behav Assess 2011;33:101–110. 21. Engelsen BK, Hagtvet KA. The dimensionality of the 12-item version of the eating attitudes test. Confirmatory factor analyses. Scand J Psychol 1999;40: 293–300. ~a C. Dimensionality 22. Garcıa-Grau E, Fuste A, Mas N, G omez J, Bados A, Saldan of three versions of the eating disorder inventory in adolescent girls. Eur Eat Disord Rev 2010;18:318–327. 23. Hrabosky JI, White MA, Masheb RM, Rothschild BS, Burke-Martindale CH, Grilo CM. Psychometric evaluation of the eating disorder examinationquestionnaire for bariatric surgery candidates. Obesity 2008;16:763–769. 24. Limbert C. The eating disorder inventory: A test of the factor structure and internal consistency in a nonclinical sample. Health Care Women Int 2004; 25:165–178. 25. Mannucci E, Ricca V, Di Bernardo M, Moretti S, Cabras P, Rotella CM. Psychometric properties of EDE 12.0 D in obese adult patients without binge eating disorder. Eat Weight Disord 1997;2:144–149. 26. Peterson CB, Crosby RD, Wonderlich SA, Joiner T, Crow SJ, Mitchell JE, et al. Psychometric properties of the eating disorder examination-questionnaire: Factor structure and internal consistency. Int J Eat Disord 2007;40:386–389.
International Journal of Eating Disorders 48:7 1016–1023 2015
DEVELOPMENT AND VALIDATION OF CEPSI 27. White HJ, Haycraft E, Goodwin H, Meyer C. Eating disorder examination questionnaire: Factor structure for adolescent girls and boys. Int J Eat Disord 2014;47:99–104. 28. Williamson DA, Barker SE, Bertman LJ, Gleaves DH. Body image, body dysphoria, and dietary restraint: Factor structure in nonclinical subjects. Behav Res Ther 1995;33:85–93. 29. Forbush KT, Wildes JE, Hunt TK. Gender norms and validity for the eating pathology symptoms inventory (EPSI). Int J Eat Disord 2014;47:85–91. 30. Bracken BA, Barona A. State of the art procedures for translating, validating and using psychoeducational tests in cross-cultural assessment. School Psychol Int 1991;12:119–132. 31. Lee S, Kwok K, Liau C, Leung T. Screening chinese patients with eating disorders using the eating attitudes test in hong kong. Int J Eat Disord 2002;32: 91–97. 32. Lee AM, Lee S. Disordered eating and its psychosocial correlates among chinese adolescent females in hong kong. Int J Eat Disord 1996;20: 177–183 33. Tong J, Shi J, Wang J, Zhang H, Zhang SF, Wu XY, Hsu L. Validity and reliability of the chinese language version of the eating disorder examination (CEDE) in mainland china: Implications for the identity and nosology of the eating disorders. Int J Eat Disord 2011;44:76–80. 34. Chou Y-J. The structure of the Chinese CES-D scale: An application of confirmatory factor analysis to higher-order factor structure. Recent Dev Appl Quant Methods Soc Sci 1997;1–26. 35. IBM. IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY: IBM Corp., 2011.
International Journal of Eating Disorders 48:7 1016–1023 2015
R 9.2 User’s Guide. Cary, NC: SAS Institute Inc., 2008. 36. SAS Institute Inc. SAS/STATV 37. Muthen LK, Muthen BO. Mplus User’s Guide, 7th ed. Los Angeles, CA: Muthen & Muthen, 1998–2012. 38. Forbush KT, Wildes JE, Pollack LO, Dunbar D, Luo J, Patterson K, et al. Development and validation of the eating pathology symptoms inventory (EPSI). Psychological Assessment 2013;25:859–878. 39. Mintz LB, Kashubeck S. Body image and disordered eating among Asian American and Caucasian college students an examination of race and gender differences. Psychol Women Quarterly 1999;23:781–796. 40. Harper DC. Children’s attitudes to physical differences among youth from western and non-western cultures. Cleft Palate Craniofac J 1995;32:114– 119. 41. Kawamura KY. Body image attitudes of Asian American and Caucasian American women and men. US: ProQuest Information & Learning; 2001. 42. Jung J, Forbes GB, Chan P. Global body and muscle satisfaction among college men in the united states and hong Kong-china. Sex Roles 2010;63:104– 117. 43. Ng CJ, Tan HM, Low WY. What do asian men consider as important masculinity attributes? Findings from the asian men’s attitudes to life events and sexuality (MALES) study. J Men’s Health 2008;5:350–355. 44. Ball K, Kenardy J. Body weight, body image, and eating behaviours: Relationships with ethnicity and acculturation in a community sample of young australian women. Eat Behav 2002;3:205–216. 45. Cohen J. Statistical power analysis for the behavioral sciences. Routledge Academic, 2013. 46. Cohen J. A power primer. Psychol Bull 1992;112:155
1023
