Adaptation and validation of the Spanish version of the Dieting Peer Competitiveness Scale to adolescents of both genders

Abstract

This study examined the Dieting Peer Competitiveness Scale; it is an instrument for evaluating this social comparison in young people. This instrumental study has two aims: The objective of the first aim was to present preliminary psychometric data from the Spanish version of the Dieting Peer Competitiveness Scale, including statistical item analysis, research about this instrument’s internal structure, and a reliability analysis, from a sample of 1067 secondary school adolescents. The second objective of the study corresponds to confirmatory factor analysis of the scale’s internal structure, as well as analysis for evidence of validity from a sample of 1075 adolescents.

Keywords

adolescence eating disorders methodological study psychometric properties risk factors social comparison

Studying social comparison as a cognitive process resulting from interaction with others was begun by Festinger (1954). His Social Comparison Theory explains the process by which people attempt to understand whether they are behaving well or poorly and/or whether they are successful or not. Since Festinger’s research, social psychologists have recognized that individuals base this assessment by comparing a variety of their own characteristics, such as their health, professional success, and opinions, with others. Furthermore, observing others and comparing one’s behavior with them aid in deciding upon appropriate behavior given the situation (Buunk et al., 2002; Gibbons and Buunk, 1999; Moya, 2007), and in this way, links personal behavior with the appropriate manner to behave. According to Mettee and Smith (1977) (cited in Buunk et al., 2005), Social Comparison Theory is about “our quest to know ourselves, about the search for self-relevant information and how people gain self-knowledge and discover reality about themselves.”

Social comparison has been studied in the disease process (Bellizzi et al., 2006; French et al., 2006; Hodges and Bridget, 2010; Terol et al., 2008). In specific cases of eating behavior disorders, it is considered that social comparison centered on appearance and eating may be a risk factor in the development of these disorders (Morrison et al., 2003; Shaw and Waller, 1995). Because of this, the literature has sought to determine, in nonclinical populations, whether people with greater risks of developing an eating disorder (ED) also present greater social comparison in terms of behavior related to the symptomatology (Bamford and Halliwell, 2009; Corning et al., 2006; Halliwell and Harvey, 2006; Jackson and Chen, 2007; O’Brien et al., 2009; Thompson et al., 1999; Tsiantas and King, 2001). This research showed that people make more peer-related comparisons, primarily with respect to physical appearance, when they presented greater risk of developing an ED (Corning et al., 2006; Huon and Walton, 2000; Jackson and Chen, 2007; Thompson and Heinberg, 1992). For example, Huon and Walton (2000) found in a sample of 124 adolescents between 12 and 16 years that restrictive dieting was associated with greater peer-related comparison with respect to eating and physical appearance.

As for instruments used to evaluate social comparison in EDs, these can be highlighted: The Physical Appearance Comparison Scale (PACS) (Thompson et al., 1991); The Body Comparison Measure (BCM) (Fisher et al., 2002; Thompson et al., 1999); Social Comparisons to Models and Peers Scale (SCMPS) (Jones, 2001); and The Iowa-Netherlands Comparison Orientation Measure (INCOM) (Gibbons and Buunk, 1999). These evaluate how adolescents compare themselves with their peers as physical appearance or weight is concerned. Huon et al. (2002) developed an instrument that included not only social comparison evaluation in terms of physical appearance, but also with respect to peer-related eating, the Dieting Peer Competitiveness (DPC) Scale. This represents an advantage versus the other aforementioned instruments, because in EDs, when risk factors are evaluated not only body satisfaction is considered, but also eating habits are considered. This instrument therefore allows evaluating social comparison within these two factors. Furthermore, it is a brief instrument, easy to apply, and has suitable psychometric properties. And, as the authors point out, one of its strong points is that the DPC successfully distinguishes serious and nonserious dieters. It is particularly useful for identifying seriously committed dieters who might persist with their dieting, even when the physical and psychological consequences are severe.

The DPC Scale is a nine-item instrument asking youngsters about the frequency they compare themselves with their peer group in terms of physical appearance and eating, answered on a 5-point Likert scale (ranging from “it never happens to me” to “I do this very frequently”). Its items were grouped into two factors: social comparison with respect to physical appearance and social comparison with respect to eating. The original study has been validated in girls attending secondary schools whose mean age was 14 years. The reliability index for the first factor oscillated between .84 and .89, while for the second it was between .69 and .78. With respect to the validity index, the instrument’s scores were related with body dissatisfaction and drive for thinness. Furthermore, girls who dieted scored higher than those who did not.

The objective of this instrumental study was to adapt the Dieting Peer Competiveness Scale (Huon et al., 2002) to Spanish and analyze the psychometric properties in both boys and girls. This article presents the results of two studies. In the first, there was a statistical analysis of these and an exploration of the test’s dimensional structure. The second study examined the instrument’s structure by confirmatory factor analysis at the same time as it applied external validity tests to it. In order to produce this instrumental study, the guidelines proposed by Carretero-Dios and Pérez (2005, 2007) were followed according to Montero and León’s (2007) classification.

Method

Participants

In the first study, there were 1067 adolescents, with 47.2 percent boys (N = 504) and the remaining 52.8 percent girls (N = 563). Of these, 29.3 percent were first-year obligatory secondary education students, 25.9 percent were second-year students, 24.6 percent third-year students, and 20.2 percent fourth-year students (in Spain, obligatory secondary education is 4 years, comprising pupils of age 12–16 years). Their mean age was 13.97 years (standard deviation (SD) = 1.35). They came from nine different schools, of which public school students represented 78.2 percent of the total, state-run private schools were 14.1 percent, and privately run schools were the remaining 7.7 percent.

In the second study, participants were 1075 secondary school adolescents, of which 47.3 percent (N = 508) were boys and 52.7 percent were girls (N = 567). First-year students comprise 27.4 percent of this total, 30.5 percent were second-year students, 22.8 percent were third-year students, and 19.3 percent were fourth-year secondary education students. The students’ mean age was 13.95 years (SD = 1.53). Public school students represented 78.7 percent of the participants, state-run private schools contributed 14.6 percent, and private school students made up the remaining 6.7 percent.

Instruments

The DPC Scale by Huon et al. (2002) comprises nine items with an original two-factor structure: physical appearance-related comparisons (five items) and eating-related comparisons (four items). The answers are compiled by a 5-point Likert scale (ranging from “it never happens to me” to “I do this very frequently”), which allows for scores of 9–45; a high score in this instrument indicates that the person compares him/herself frequently with the peer group regarding physical appearance and/or eating. This questionnaire provides a cut-off score that can be applied as a criterion against which to indicate some degree of concern, and it differentiates between “Never” dieters and other dieters. A cut-off score of 33 produced perfect sensitivity, excellent specificity, and percentage correct classification (85%). As for the instrument’s internal consistency, the appearance-related social comparison was .89, and the second factor, eating-related social comparison, was .76.

Stages preceding the Spanish version adaptation of the DPC

The questionnaire’s translation and adaptation procedure took place using an iterative method (Bullinger et al., 1994; Guillemin et al., 1993):

Translation. Two bilingual persons (residents in Spain whose native language was English) were first instructed about the study’s conceptual framework, and then they translated two versions into Spanish independently. This created the first Spanish version.

Back translation. The resulting version was translated back to English by two separate bilingual individuals who had not previously been informed about the objectives of the construct to measure. From this, a version practically equal to the original resulted.

Expert review. A team composed by members of the investigative group (two experts in health psychology and one statistician) reviewed all versions and evaluated the comprehension, as well as the semantic, linguistic, and conceptual equivalency in such a manner that after modifying and adjusting the instructions, and some items, a consensus was reached.

Pilot program. With the aim of evaluating the comprehension, reliability, and welcome of both the items as well as the response scale, the questionnaire was administered to a pilot sample of 10 adolescents. The pilot sample was also interviewed, and opinions concerning different aspects related to understanding the instructions, the wording of the items, and so on were given. Some modifications to the Spanish version of the instrument resulted because of this.

Eating Attitude Test-40 (EAT-40) (Garner and Garfinkel, 1979) adapted to the Spanish population by Castro et al. (1991): This self-administered questionnaire is designed to detect the presence of abnormal eating attitudes, especially those related to fears of weight gain, the impetus to lose weight, and the presence of restrictive eating patterns. It consists of 40 items with a 6-point Likert scale (from “never” to “always”) whose total score varies between 0 and 120. The scale’s internal consistency in the original work of the Spanish adaptation was .93 (in the present study, this was .81).

The Body Dissatisfaction Subscale of the Eating Disorder Inventory-2 (EDI-2) (Garner, 1998): The EDI-2 is a self-report that evaluates symptoms related to eating behavior disorders. The authors confirmed that the 11 subscales can be used separately; among these is Body Dissatisfaction, consisting of nine items measured on a 6-point scale (from “never” to “always”) that presents a response range from 0 to 27 and evaluates the subject’s dissatisfaction with his or her body or parts of the body. It has an internal consistency of .92 in nonclinical samples, while for patients with EDs it is .90 (Garner, 1998): In the present study, the results showed an internal consistency of .88.

An ad hoc Perceived Pressure to Lose Weight Questionnaire containing five items referring to the perception a person has of social pressure upon himself or herself to behave in certain ways to either lose weight or not, in five different behaviors: to diet, to practice physical activity, to eat low-calorie food, to delete meals of the day, and to eat less food. Its items were evaluated with a Likert-type response scale from 1 to 5, with 1 being “strongly disagree” and 5 being “strongly agree,” whose total scores range from 4 to 20. For this study, the scale’s internal consistency was .76.

The dieting frequency (ad hoc) contained one item asking the adolescents about the frequency they dieted, concretely: “Have you ever been on a diet to lose weight?” The response alternatives to this item were the following: I do not diet, I diet for less than 1 week, I diet for less than 1 month, and I diet for more than 1 month.

Procedure

Single-stage cluster sampling took place, and nine schools were randomly chosen to collaborate in the study. Following participation acceptance, informed consent was requested from both parents and children. The questionnaires were administered collectively, anonymously, and voluntarily during school hours in the presence of the investigators. All participants completed all the measures. Once the questionnaires were collected, the sample was randomly divided, with one-half of the data used for study 1 and the other half used for study 2.

Data analysis

Based on the suggestions provided by Carretero-Dios and Pérez (2005), the following analyses were made: statistical analysis of the items, analysis of the scale’s internal structure by exploratory analysis of principal axes with varimax rotation, and reliability estimation with Cronbach’s α coefficient of the total scale. For the final solution, we considered the criteria of Stevens (1922) (cited in Martínez-Arias, 1996) for assessing variable saturations, which establishes at least 15 percent of shared variance by the factor and variable (saturation = .4) using SPSS V.18.0.

Confirmatory factor analysis was performed through structural equation modeling using AMOS version 19.0. To make the calculations, the comparative fit of different measurement models was analyzed (Batista-Foguet and Coenders, 2000; Timothy and Brown, 2006), constructed according to the starting theoretical proposal and the empirical results from Study 1.

Results

Study 1: preliminary psychometric study of the DPC

Study of the instrument’s dimensionality

To learn about the internal structure of DPC, exploratory factor analysis was conducted on the nine items by the extraction method of principal axes with varimax rotation in both girls and boys (Table 1). Before conducting the analysis, the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy and Bartlett’s test were conducted. The KMO index showed a value of .79 in girls and .74 in boys, while Bartlett’s test was statistically significant for both genders (girls: $χ_{(36)}^{2}$ = 843.547; p < .001; boys: $χ_{(36)}^{2}$ = 568.516; p < .001), confirming the adequacy of the correlation matrix for this analysis. The factor pattern finally remained composed of two factors that explained 36.3 percent of the variance in girls and 32.4 percent in boys; Factor 1 was composed of six items and Factor 2 was composed of three items. In the case of girls, Factor 1 explained 25.9 percent of the total variance while Factor 2 did so with 10.4 percent. For boys, Factor 1 explained 22.2 percent of the total variance and Factor 2 did so with 10.2 percent. Table 1 describes the rotated components matrix, within which the factor loadings grouped, in an adequate manner, all the items in each one of the factors that they measured. In all girls, the factor loadings were higher than .40 except for item 2, which was .33. For the case of boys, the factor loadings were higher than .40 except for item 1, which was .38.

Table 1.

Rotated component matrix.

Item	Girl’s components		Boy’s components
	Factor 1	Factor 2	Factor 1	Factor 2
1	.557	−.043	.381	−.102
3	.762	−.027	.700	.015
4	.583	−.114	.558	−.027
5	.638	−.025	.516	−.090
7	.682	−.135	.606	−.129
9	.431	.058	.574	.064
2	−.083	.330	−.033	.437
6	−.056	.553	−.197	.566
8	.100	.713	.069	.642

Reliability and statistical analysis of the items

The scale’s total reliability was calculated differentiating by gender, and Cronbach’s α resulted being .63 for girls and .55 for boys. The corrected item-test correlation index oscillated between .01 and .47 in the girls while in boys it ranged between −.03 and .45.

Statistical analysis of the items proceeded by calculating the mean, SD, and item-test corrected correlation index. Internal consistency with Cronbach’s α for each subscale as well as for the total scale was calculated. Table 2 shows the psychometric properties for each item within each subscale in terms of gender.

Table 2.

Psychometric properties of the group of items from each of the factors in the DPC questionnaire.

Items	Girls				Boys
	Complete scale (α = .63)				Complete scale (α = .55)
	Factor 1 (α =.78)				Factor 1 (α = .73)
	Mean	SD	R IT	α^a	Mean	SD	R IT	α^a
1	2.38	1.48	.490	.757	1.83	1.25	.315	.729
3	2.69	1.57	.651	.714	1.89	1.27	.568	.659
4	3.06	1.45	.509	.752	2.51	1.38	.473	.686
5	2.69	1.50	.551	.742	2.25	1.37	.443	.695
7	2.80	1.41	.597	.731	2.22	1.30	.512	.674
9	2.25	1.37	.374	.782	2.06	1.30	.464	.689
Items	Factor 2 (α = .53)				Factor 2 (α = .56)
	Mean	SD	R IT	α^a	Mean	SD	R IT	α^a
2	3.78	1.31	.247	.561	3.67	1.39	.322	.526
6	2.92	1.52	.370	.377	3.54	1.49	.381	.439
8	3.03	1.54	.415	.295	3.24	1.48	.406	.398

DPC: Dieting Peer Competitiveness; SD: standard deviation; R IT: item-total correlation.

Cronbach’s α if deleting the item.

Study 2: confirmatory factor analysis and external evidence of validity

Confirmatory factor analysis

The measurement model fit was assessed by the goodness-of-fit statistic χ². Because χ² tends to be significant in large samples, the χ²/gl ratio was calculated. If this ratio is inferior to 6, it is assumed that the model fit is adequate. This index is called the relative χ² because it is calculated to search for sample size independence. Carmines and McIver (1981) and Kline (1998) establish a value of 3 as an acceptable model. Moreover, and following the recommendation advising comparing various indices to ensure the fit of the proposed model, the following goodness-of-fit indices were kept in mind: the “comparative fix index” (CFI) (Schumacker and Lomax, 1996), with the calculated value of the root mean square error of approximation (RMSEA) also used, the model is considered to fit adequately when values between .05 and .08 are reached. Values inferior to .05 are considered indicative of excellent fit.

Confirmatory analysis was initially performed by gender for both girls and boys, and the Kolmogorov–Smirnov test confirmed that the questionnaire’s nine items fit the normal distribution. As for the girls, the overall fit shown by the scale was adequate; the Chi-square test was significant ( $χ_{(26)}^{2}$ = 76.5, p = .000), the χ²/gl ratio was 2.94, and the fit indices showed adequate results (CFI = .93; RMSEA = .059). Regarding the correlation between the factors and the factor loadings of the items, the two factors do not correlate with each other. With respect to the factor loadings, they oscillated between .52 and .74 for Factor 1, and .32 and .82 for Factor 2. The low factor loadings of items 2 and 6 from Factor 2 are noteworthy. In the case of boys, the overall fit shown by the scale was adequate; the Chi-square test was significant ( $χ_{(26)}^{2}$ , p ≤ .001), the χ²/gl ratio was 2.16, and the fit indices were adequate (CFI = .932; RMSEA = .048). As for the correlation of the factors and the factor loadings of the items, the two factors do not correlate with each other. As for the factor loadings, they ranged between .44 and .64 for Factor 1, and .28 and .54 for Factor 2. Noteworthy is the low factor loading of item 2 with respect to Factor 2.

External evidence of validity

This section’s objective was to incorporate evidence concerning the relationships between the measurements provided by the DPC and other theoretically related external variables: eating attitudes, body dissatisfaction, and pressure perceived by the peer group. The results show coherent relationship patterns with the theory. On one hand, Factor 1 of the DPC showed positive and significant relationships with the scores in abnormal eating attitudes (girls: r = .51, p ≤ .01; boys: r = .44, p ≤ .01) and body dissatisfaction (girls: r = .62, p ≤ .01; boys: r = .40, p ≤ .01), and the pressure by significant others to lose weight (girls: r = .36, p ≤ .01; boys: r = .33, p ≤ .01). In contrast, Factor 2 of the DPC only showed a significant relationship with eating attitudes in boys (r = .12, p ≤ .05) and girls (r = .12, p ≤ .05); however, it was low.

Just as the authors did in the instrument’s validation, the scores obtained in the DPC instrument were compared, classifying boys and girls according to the frequency they claimed to diet. The analyses showed that adolescents of both genders who claimed not to diet scored significantly lower in Factor 1 of the DPC (boys: F = 21.91, p < .001; girls: F = 47.20, p < .001). Only in the case of girls were there significant differences in Factor 1 scores between those who claimed to diet for less than 1 week (M = 17.81, SD = 5.93) in comparison with those who claimed to do so for more than 1 month (M = 20.25, SD = 6.28). Contrasting this, there were no significant differences in the scores reached in Factor 2 for either boys or girls.

Discussion

This study’s objective was to adapt the DPC to the Spanish population and analyze its psychometric properties. The results of these analyses showed that the psychometric properties of the adapted version of the DPC for Spanish adolescents are inferior with respect to the authors of the original version.

With respect to the factor structure, the two factors obtained are similar to those in the original version with the exception of one item, number 9, which in this work loads adequately in Factor 1, and furthermore presented theoretical coherence with it. One of this article’s proposals arises from this difference, modifying the names the original authors proposed for the two factors. In the original version, and according to its authors, Factor 1 refers to the comparison one makes with respect to his or her physical appearance and involves intangible aspects like wishes and aspirations. For its part, Factor 2 refers to the comparison one makes with respect to his or her eating habits and includes more tangible aspects that can be immediately changed. However, the distinction between the two factors is not so clear. For example, Factor 2 includes item 6, I don’t mind going out in a short skirt even if my friend is wearing a short skirt and looks better than I do, which does not refer to an eating-related comparison like the original version’s authors indicate. Furthermore, the distinction between the two factors in terms of including more or less tangible aspects is not completely clear either. For example, item 1 from Factor 1, I do not like wearing a bathing suit because I don’t think I look as good as the other girls, does not refer to a comparison implying greater desire or is more possible than item 2 from Factor 2, I don’t mind having junk food even if my friends are having healthy food.

For these reasons, and according to the items’ content, as well as considering the Social Comparison Theory (Festinger, 1954), in this article, we propose calling Factor 1 “affective social comparison,” which implies making a social comparison with the peer with an affective charge. We call Factor 2 “neutral social comparison” because it refers to this comparison with the peer group without an affective charge.

As for the reliability analysis, the results showed good internal consistency indices for the total scale and for Factor 1. On the other hand, the analysis of relationships with other variables showed evidence of validity to be convergent with the very ED symptomatology, body dissatisfaction, and the pressure to lose weight with Factor 1 of the DPC instrument, just as was expected according to the previous research (Bamford and Halliwell, 2009; Corning et al., 2006; Halliwell and Harvey, 2006; Jackson and Chen, 2007; O’Brien et al., 2009; Thompson et al., 1999).

However, with respect to Factor 2, the psychometric properties found were not satisfactory. With the aim of improving its internal consistency, following some of these actions in further research would be advisable: review the wording of the items, as eliminating the double negative might be advisable. In particular, with regard to item 2 (“I don’t mind having junk food even if my friends are having healthy food”), this item has low but acceptable factor loadings because of the sample size being larger than 300 (Guadalognoli and Velicer, 1988). Furthermore, in order to attempt to increase the internal consistency of Factor 2 (where item 2 belongs), we propose including other qualitatively discriminative items, which refer to a neutral social comparison, adapted to the Spanish population. In the same manner, once these modifications occur, the relationships of this factor with the scores on the scales that evaluate the symptomatology of EDs would also increase, and this way improve its validity, as well as could be able to change the scale’s factorial structure.

These results need to be treated cautiously because their exploratory nature and the limitations of this type of analysis must be considered. However, despite the noted limitations, our results suppose a useful starting point for evaluating peer-related social comparison with respect to eating and physical appearance attitudes in Spanish adolescents.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Bamford

Halliwell

(2009) Investigation the role of attachment in social comparison theories of eating disorders within a non-clinical female population. European Eating Disorders Review 17: 371–379.

Batista-Foguet

Coenders

(2000) Modelos de Ecuaciones Estructurales [Structural Equations Models]. Madrid: La Muralla/Hespérides.

Bellizzi

Blank

Oakes

(2006) Social comparison processes in autobiographies of adult cancer survivors. Journal of Health Psychology 11: 777–786.

Bullinger

Anderson

Cella

. (1994) Developing and evaluating cross-cultural instruments from minimum requirements to optimal models. Quality of Life Research 2: 451–459.

Buunk

Belmonte

Peiró

. (2005) Diferencias individuales en la comparación social: propiedades de la escala española de orientación hacia la comparación social [Individual differences in social comparison: Properties of the Spanish scale of social comparison orientation]. Revista Latinoamericana de Psicología 37: 561–579.

Buunk

Gibbons

Visser

(2002) The relevance of social comparison processes for prevention and health care. Patient Education and Counseling 47: 1–3.

Carmines

McIver

(1981) Analyzing models with unobserved variables: Analysis of covariance structures. In: George

Edward

(eds) Social Measurement. Thousand Oaks, CA: SAGE, pp. 65–115.

Carretero-Dios

Pérez

(2005) Normas para el desarrollo y revisión de estudios instrumentales [Standards development and revision of instrumental studies]. International Journal of Clinical and Health Psychology 5: 521–551.

Carretero-Dios

Pérez

(2007) Standards for the development and the review of instrumental studies: Considerations about test selection in psychological research. International Journal of Clinical and Health Psychology 7: 863–882.

10.

Castro

Toro

Salamero

Guimera

(1991). The eating attitudes test: validation of the spanish versión. Psychological Assessment, 7(2): 175–190.

11.

Corning

Krumm

Smitham

(2006) Differential social comparison processes in women with and without eating disorder symptoms. Journal of Counseling Psychology 53: 338–349.

12.

Festinger

(1954) A theory of social comparison processes. Human Relations 7: 117–140.

13.

Fisher

Dunn

Thompson

(2002) Social comparison and body image: An investigation of body comparison processes using multidimensional scaling. Journal of Social and Clinical Psychology 21: 566–579.

14.

French

Hevey

Sutton

. (2006) Personal and social comparison information about health risk. Journal of Health Psychology 11: 497–510.

15.

Garner

(1998) EDI-2: Inventario de Trastornos de la Conducta Alimentaria [EDI-2: Eating Disorders Inventory]. Madrid: TEA Ediciones.

16.

Garner

Garfinkel

(1979) The eating attitudes test: An index of the symptoms of anorexia nervosa. Psychological Medicine 9: 179–273.

17.

Gibbons

Buunk

(1999) Individual differences in social comparison: Development of a scale of social comparison orientation. Journal of Personality and Social Psychology 76: 129–142.

18.

Guadalognoli

Velicer

(1988) Relation of sample size to the stability of component patterns. Psychological Bulletin 103: 265–275.

19.

Guillemin

Bombardier

Beaton

(1993) Cross-cultural adaptation of health-related quality of life measures: Literature review and proposed guidelines. Journal of Clinical Epidemiology 46: 1417–1432.

20.

Halliwell

Harvey

(2006) Examination of a sociocultural model of disordered eating among male and female adolescents. British Journal of Health Psychology 11: 235–248.

21.

Hodges

Bridget

(2010) Social comparison within self-help groups: Views of parents of children with Duchenne Muscular Dystrophy. Journal of Health Psychology 15: 483–492.

22.

Huon

Walton

(2000) Initiation of dieting among adolescent females. The International Journal of Eating Disorders 28: 226–230.

23.

Huon

Piiram

Hayne

. (2002) Assessing body and eating peer-focused comparisons: The Dieting Peer Competitiveness (DPC) Scale. European Eating Disorders Review 10: 428–446.

24.

Jackson

Chen

(2007) Identifying the eating disorder symptomatic in China: The role of sociocultural factors and culturally defined appearance concerns. Journal of Psychosomatic Research 62: 241–249.

25.

Jones

(2001) Social comparison and body image: Attractiveness comparison to models and peers among adolescent girls and boys. Sex Roles 45: 645–664.

26.

Kline

(1998) Principles and Practice of Structural Equation Modeling. New York: Guilford Press.

27.

Martínez-Arias

(1996) Psicometría: Teoría de los tests psicológicos y educativos [Psychometrics: Theory of Psychological and Educational Tests]. Madrid: Síntesis.

28.

Mettee

Smith

(1977) Social comparison and interpersonal attraction: The case for dissimilarity. In: Suls

Miller

(eds) Social Comparison Processes: Theoretical and Empirical Perspectives. Washington, DC: Hemisphere, pp. 21–41.

29.

Montero

León

(2007) A guide for naming research studies in psychology. International Journal of Clinical and Health Psychology 7: 847–862.

30.

Morrison

Waller

Meyer

. (2003) Social comparison in the eating disorders. The Journal of Nervous and Mental Disease 191: 553–555.

31.

Moya

(2007) Relaciones interpersonales: Función e inicio [Interpersonal relations: Functions and initiation]. In: Morales

Moya

Gaviria

(eds) Psicología Social. Madrid: McGraw-Hill. 333-358

32.

O’Brien

Caputi

Minto

. (2009) Upward and downward physical appearance comparisons: Development of scales and examination of predictive qualities. Body Image 6: 201–206.

33.

Schumacker

Lomax

(1996) A Beginner’s Guide to Structural Equation Modeling. Mahwah, NJ: Erlbaum.

34.

Shaw

Waller

(1995) The media’s impact on body image: Implications for prevention and treatment. Eating Disorders 3: 115–123.

35.

Terol

Rodríguez-Marín

Pastor

. (2008) La evaluación psicológica en el proceso de enfermedad y adaptación a la enfermedad crónica. In: Morales

Huici

Gómez

(eds) Método, teoría e investigación en psicología social [Method, Theory and Research in Social Psychology]. Madrid: Prentice Hall, pp. 247–261.

36.

Thompson

Coovert

Stormer

(1999) Body image, social comparison, and eating disturbance: A covariance structure modelling investigation. The International Journal of Eating Disorders 26: 43–51.

37.

Thompson

Heinberg

Tantleff

(1991) The physical appearance comparison scale (PACS). The Behaviour Therapist 14: 174.

38.

Thompson

Heinberg

(1993) Preliminary test of two hypotheses of body image disturbance. International Journal of Eating Disorders, 14 (1): 59–63.

39.

Timothy

Brown

(2006) Confirmatory Factor Analysis for Applied Research. New York: The Guilford Press.

40.

Tsiantas

King

(2001) Similarities in body image in sisters: The role of sociocultural internalization and social comparison. Eating Disorders 9: 141–158.