Journal of American College Health
ISSN: 0744-8481 (Print) 1940-3208 (Online) Journal homepage: http://www.tandfonline.com/loi/vach20
An Examination of Influences on Body Dissatisfaction Among Asian American College Females: Do Family, Media, or Peers Play A Role? Sarah J. Javier MS & Faye Z. Belgrave PhD To cite this article: Sarah J. Javier MS & Faye Z. Belgrave PhD (2015): An Examination of Influences on Body Dissatisfaction Among Asian American College Females: Do Family, Media, or Peers Play A Role?, Journal of American College Health, DOI: 10.1080/07448481.2015.1031240 To link to this article: http://dx.doi.org/10.1080/07448481.2015.1031240
Accepted author version posted online: 31 Mar 2015.
Submit your article to this journal
Article views: 37
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=vach20 Download by: [University of Cambridge]
Date: 05 November 2015, At: 21:32
ACCEPTED MANUSCRIPT An Examination of Influences on Body Dissatisfaction Among Asian American College Females: Do Family, Media, or Peers Play A Role? Sarah J. Javier, MS1; Faye Z. Belgrave, PhD1 1
Department of Psychology, Virginia Commonwealth University, Richmond, Virginia
Address correspondence to Sarah Javier, MS, Virginia Commonwealth University, 806 West
Downloaded by [University of Cambridge] at 21:32 05 November 2015
Franklin St., Richmond, VA 23284, USA (e-mail: [email protected]). Received: 28 August 2014 Revised: 23 January 2015 Accepted: 14 March 2015 Abstract Objective: The etiology of body dissatisfaction and its correlates (e.g. disordered eating) among ethnic minorities is generally unknown. The purpose of this study was to replicate the Tripartite Model of Influence in an Asian American college female sample in order to examine this relationship. Participants: Participants were 80 undergraduate Asian American females between the ages of 18-25. Methods: Participants completed a survey which included the TIS, BPSS, and SATAQ-3. Results: Mediation analyses indicated that thin-ideal internalization fully mediated the relationship between media influence and body dissatisfaction and partially mediated the relationship between peer influence and body dissatisfaction. Family influence did not significantly predict body dissatisfaction. Conclusions: Asian American college females experience body dissatisfaction through mechanisms that have not been examined in detail. These factors must be considered when creating targeted health promotion strategies and developing best practices for eating disorder assessment and treatment protocols at university health centers.
1
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Keywords
Downloaded by [University of Cambridge] at 21:32 05 November 2015
Asian American, body dissatisfaction, thin-ideal internalization, tripartite influence
2
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Appearance becomes very salient during emerging adulthood as individuals transition from home to college environments, in which peers are influential.1 The prevalence of eating disorder development among college females is especially high, with rates up to 24% among college students.2 In addition, approximately 25% of college and university counseling centers reported an increase in clients presenting with disordered symptoms in recent years.3 What these
Downloaded by [University of Cambridge] at 21:32 05 November 2015
numbers do not indicate is the accurate prevalence of disordered eating and body dissatisfaction among Asian American (AA) college students. AA college women demonstrate levels of body dissatisfaction equivalent to White college women and manifest similar adverse outcomes including negative affect and disordered eating.4-5 Yet, AAs in general are less likely than other ethnic groups to utilize psychological treatment services,6 and college students are less likely to seek counseling if they endorse traditional Asian values.7 Body dissatisfaction that results from sociocultural image expectations can be understood in terms of the Tripartite Model of Influence (TMI).8 This model asserts that there are three major influences on body dissatisfaction: peer, family, and media. These directly influence body dissatisfaction by affecting an individual’s eating behaviors (e.g. restrictive eating) and negative affect.6 Further, these have an indirect effect on body dissatisfaction via internalization of societal beauty standards and excessive appearance comparison.8 The TMI has been validated only among predominantly White samples;9 however, we expected this framework to be useful in understanding the etiology of body dissatisfaction in AA college females, as this group also exhibits thin-ideal internalization.10 The purpose of this study was to replicate the TMI among a diverse sample of AA college women. We hypothesized that the relationship between each of the influences and body
3
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT dissatisfaction would be mediated by thin-ideal internalization. The model has not been evaluated with AAs, and understanding factors that contribute to body dissatisfaction is useful for understanding their adjustment to the college environment. AA women may face unique cultural challenges in both transitioning to college11 and seeking psychological services for outcomes related to body dissatisfaction.12 The study findings may prove useful for developing
Downloaded by [University of Cambridge] at 21:32 05 November 2015
disordered eating preventive interventions for AA college females. Methods Participants Participants were female undergraduate students from a large university in the MidAtlantic who identified as second-generation AA (N = 80) and who were between the ages of 1825. The mean age was 20.3 years, and most were Freshmen (50.6%). Although Freshmen appear to be oversampled, the current sample matches the demographics of the host institution. Most Freshmen at the institution are majors in humanities and sciences, including Psychology from which the Sona sample was drawn (See Procedure).13 AAs constitute 12% of the total host institution’s student population;14 however, the current study oversampled AA students (approximately 30%, or N = 80, of a larger study sample, N = 260). The sample included the following subgroups: Cambodian (2.5%), Chinese (17.5%), Indian (20.0%), Korean (11.3%), Pakistani (6.3%), Filipino (12.5%), Thai (3.8%), Vietnamese (18.8%), and other Asian category (7.3%). While South Asian individuals (e.g. Indian, Pakistani) are sometimes considered an ethnic group separate from East and Southeast Asians, they were included in the current study as these groups manifest similarities in eating disturbances.15 We assumed that the mechanism of the TMI would be similar for all groups of AAs.
4
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Measures Body Parts Satisfaction Scale (BPSS) Body dissatisfaction was measured using The Body Parts Satisfaction Scale (BPSS16), a list of 24 body parts (e.g., eyes, nose, mouth) about which individuals express satisfaction/dissatisfaction. In two college samples, reliability and validity of the BPSS for ethnic
Downloaded by [University of Cambridge] at 21:32 05 November 2015
minority and non-ethnic minority women showed sound psychometric properties.17 In the current sample, α = .92. Tripartite Influence Scale (TIS) The independent variables of the tripartite influences of peer, family, and media influence, were assessed using the TIS.18,8 This measure consists of 43 items that assess whether the Tripartite Influences lead to body dissatisfaction and disordered eating behaviors. The original measure8 was developed by combining items from different measures and scales that assessed family, media, and peer influence and demonstrated moderate reliability among college females (α = .69-.87). In the current sample, α = .88. Thin-ideal internalization Thin-ideal internalization was assessed using the 9-item internalization subscale from the Sociocultural Attitudes towards Appearance Questionnaire (3) (SATAQ-319). The internalization subscale has shown moderate consistency (α = .69) among young women.20 In the current study, α =.98. Demographic items
5
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Demographic items included birth date, Asian ethnicity, generational status, and selfreported weight and height. BMI is a salient predictor of body dissatisfaction,21 and thus was controlled for. The mathematical formula used for BMI was: weight (lb.) / [height (in.)]2 x 703.22 Procedure The study first received Institutional Review Board approval. Participants were recruited
Downloaded by [University of Cambridge] at 21:32 05 November 2015
through Sona,23 a web-based research platform, and student organizations. Participants received a link to the questionnaire on a secure website after meeting inclusion criteria. After completing an informed consent and accepting the study terms, participants completed the questionnaire and received course credit. Non-psychology students (N = 5) were recruited by researchers through brief, in-person presentations in which the research was presented as a study of body image among AAs. Interested participants completed an on-line version of the questionnaire after completing informed consent, and were entered into a raffle to receive a $100.00 gift card. Results Preliminary analysis Data was analyzed using SPSS version 20.0.24 Data were screened for outliers and missing data (N = 5) was dealt with using mode imputation. Statistical assumptions were checked. Select items on the BPSS were reverse-coded so that higher scores indicated higher levels of body dissatisfaction. The effect sizes for the mediation models were calculated according to recommendations by Kenny25 and Preacher and Kelley.26 Partial correlations were first computed and then, an effect size of .15 (95% CI [.12, .37]) was obtained for the model involving media influence (X), thin-ideal internalization (M), and body dissatisfaction (Y). indicating that 15% of the variance in
6
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT body dissatisfaction was attributable to the overlap of media influence and thin-ideal internalization. Main analysis The bootstrapping method has been used instead of a traditional mediation model to adequately deal with unmet assumptions of traditional mediation.27 This method involved four
Downloaded by [University of Cambridge] at 21:32 05 November 2015
steps: (1) A bootstrap sample of 80 cases was generated from the study sample of 80; (2) the regression coefficient b and indirect effect a was calculated based on this sample; (3) 1,000 estimates of the indirect effect of interest were obtained by repeating steps 1 and 2 1,000 times; (4) the mean of the 1,000 estimates from step (3) was calculated. Significance values and the presence or absence of zero in a 95% CI indicated whether the indirect effects were statistically significant (See Table 1) . The z-score for the specific indirect effect of family influence on body dissatisfaction through the mediator of thin-ideal internalization was non-significant, which was further confirmed by the presence of zero in the 95% confidence interval (see Table 1). Thus, contrary to study hypotheses, family influence was not related to thin-ideal internalization, t = .99, p = .33, and thin-ideal internalization was not related to body dissatisfaction, t = .48, p = .63. The z-score for the specific indirect effect of media influence on body dissatisfaction through the mediator of thin-ideal internalization was statistically significant, indicating that a 1point increase in media influence was associated with a .48 increase in body dissatisfaction through level of thin-ideal internalization (see Table 1). Thus, greater media influence was associated with greater thin-ideal internalization, t = 4.83, p < .001, and greater thin-ideal internalization was associated with greater body dissatisfaction, t = 4.06, p
ISSN: 0744-8481 (Print) 1940-3208 (Online) Journal homepage: http://www.tandfonline.com/loi/vach20
An Examination of Influences on Body Dissatisfaction Among Asian American College Females: Do Family, Media, or Peers Play A Role? Sarah J. Javier MS & Faye Z. Belgrave PhD To cite this article: Sarah J. Javier MS & Faye Z. Belgrave PhD (2015): An Examination of Influences on Body Dissatisfaction Among Asian American College Females: Do Family, Media, or Peers Play A Role?, Journal of American College Health, DOI: 10.1080/07448481.2015.1031240 To link to this article: http://dx.doi.org/10.1080/07448481.2015.1031240
Accepted author version posted online: 31 Mar 2015.
Submit your article to this journal
Article views: 37
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=vach20 Download by: [University of Cambridge]
Date: 05 November 2015, At: 21:32
ACCEPTED MANUSCRIPT An Examination of Influences on Body Dissatisfaction Among Asian American College Females: Do Family, Media, or Peers Play A Role? Sarah J. Javier, MS1; Faye Z. Belgrave, PhD1 1
Department of Psychology, Virginia Commonwealth University, Richmond, Virginia
Address correspondence to Sarah Javier, MS, Virginia Commonwealth University, 806 West
Downloaded by [University of Cambridge] at 21:32 05 November 2015
Franklin St., Richmond, VA 23284, USA (e-mail: [email protected]). Received: 28 August 2014 Revised: 23 January 2015 Accepted: 14 March 2015 Abstract Objective: The etiology of body dissatisfaction and its correlates (e.g. disordered eating) among ethnic minorities is generally unknown. The purpose of this study was to replicate the Tripartite Model of Influence in an Asian American college female sample in order to examine this relationship. Participants: Participants were 80 undergraduate Asian American females between the ages of 18-25. Methods: Participants completed a survey which included the TIS, BPSS, and SATAQ-3. Results: Mediation analyses indicated that thin-ideal internalization fully mediated the relationship between media influence and body dissatisfaction and partially mediated the relationship between peer influence and body dissatisfaction. Family influence did not significantly predict body dissatisfaction. Conclusions: Asian American college females experience body dissatisfaction through mechanisms that have not been examined in detail. These factors must be considered when creating targeted health promotion strategies and developing best practices for eating disorder assessment and treatment protocols at university health centers.
1
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Keywords
Downloaded by [University of Cambridge] at 21:32 05 November 2015
Asian American, body dissatisfaction, thin-ideal internalization, tripartite influence
2
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Appearance becomes very salient during emerging adulthood as individuals transition from home to college environments, in which peers are influential.1 The prevalence of eating disorder development among college females is especially high, with rates up to 24% among college students.2 In addition, approximately 25% of college and university counseling centers reported an increase in clients presenting with disordered symptoms in recent years.3 What these
Downloaded by [University of Cambridge] at 21:32 05 November 2015
numbers do not indicate is the accurate prevalence of disordered eating and body dissatisfaction among Asian American (AA) college students. AA college women demonstrate levels of body dissatisfaction equivalent to White college women and manifest similar adverse outcomes including negative affect and disordered eating.4-5 Yet, AAs in general are less likely than other ethnic groups to utilize psychological treatment services,6 and college students are less likely to seek counseling if they endorse traditional Asian values.7 Body dissatisfaction that results from sociocultural image expectations can be understood in terms of the Tripartite Model of Influence (TMI).8 This model asserts that there are three major influences on body dissatisfaction: peer, family, and media. These directly influence body dissatisfaction by affecting an individual’s eating behaviors (e.g. restrictive eating) and negative affect.6 Further, these have an indirect effect on body dissatisfaction via internalization of societal beauty standards and excessive appearance comparison.8 The TMI has been validated only among predominantly White samples;9 however, we expected this framework to be useful in understanding the etiology of body dissatisfaction in AA college females, as this group also exhibits thin-ideal internalization.10 The purpose of this study was to replicate the TMI among a diverse sample of AA college women. We hypothesized that the relationship between each of the influences and body
3
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT dissatisfaction would be mediated by thin-ideal internalization. The model has not been evaluated with AAs, and understanding factors that contribute to body dissatisfaction is useful for understanding their adjustment to the college environment. AA women may face unique cultural challenges in both transitioning to college11 and seeking psychological services for outcomes related to body dissatisfaction.12 The study findings may prove useful for developing
Downloaded by [University of Cambridge] at 21:32 05 November 2015
disordered eating preventive interventions for AA college females. Methods Participants Participants were female undergraduate students from a large university in the MidAtlantic who identified as second-generation AA (N = 80) and who were between the ages of 1825. The mean age was 20.3 years, and most were Freshmen (50.6%). Although Freshmen appear to be oversampled, the current sample matches the demographics of the host institution. Most Freshmen at the institution are majors in humanities and sciences, including Psychology from which the Sona sample was drawn (See Procedure).13 AAs constitute 12% of the total host institution’s student population;14 however, the current study oversampled AA students (approximately 30%, or N = 80, of a larger study sample, N = 260). The sample included the following subgroups: Cambodian (2.5%), Chinese (17.5%), Indian (20.0%), Korean (11.3%), Pakistani (6.3%), Filipino (12.5%), Thai (3.8%), Vietnamese (18.8%), and other Asian category (7.3%). While South Asian individuals (e.g. Indian, Pakistani) are sometimes considered an ethnic group separate from East and Southeast Asians, they were included in the current study as these groups manifest similarities in eating disturbances.15 We assumed that the mechanism of the TMI would be similar for all groups of AAs.
4
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Measures Body Parts Satisfaction Scale (BPSS) Body dissatisfaction was measured using The Body Parts Satisfaction Scale (BPSS16), a list of 24 body parts (e.g., eyes, nose, mouth) about which individuals express satisfaction/dissatisfaction. In two college samples, reliability and validity of the BPSS for ethnic
Downloaded by [University of Cambridge] at 21:32 05 November 2015
minority and non-ethnic minority women showed sound psychometric properties.17 In the current sample, α = .92. Tripartite Influence Scale (TIS) The independent variables of the tripartite influences of peer, family, and media influence, were assessed using the TIS.18,8 This measure consists of 43 items that assess whether the Tripartite Influences lead to body dissatisfaction and disordered eating behaviors. The original measure8 was developed by combining items from different measures and scales that assessed family, media, and peer influence and demonstrated moderate reliability among college females (α = .69-.87). In the current sample, α = .88. Thin-ideal internalization Thin-ideal internalization was assessed using the 9-item internalization subscale from the Sociocultural Attitudes towards Appearance Questionnaire (3) (SATAQ-319). The internalization subscale has shown moderate consistency (α = .69) among young women.20 In the current study, α =.98. Demographic items
5
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Demographic items included birth date, Asian ethnicity, generational status, and selfreported weight and height. BMI is a salient predictor of body dissatisfaction,21 and thus was controlled for. The mathematical formula used for BMI was: weight (lb.) / [height (in.)]2 x 703.22 Procedure The study first received Institutional Review Board approval. Participants were recruited
Downloaded by [University of Cambridge] at 21:32 05 November 2015
through Sona,23 a web-based research platform, and student organizations. Participants received a link to the questionnaire on a secure website after meeting inclusion criteria. After completing an informed consent and accepting the study terms, participants completed the questionnaire and received course credit. Non-psychology students (N = 5) were recruited by researchers through brief, in-person presentations in which the research was presented as a study of body image among AAs. Interested participants completed an on-line version of the questionnaire after completing informed consent, and were entered into a raffle to receive a $100.00 gift card. Results Preliminary analysis Data was analyzed using SPSS version 20.0.24 Data were screened for outliers and missing data (N = 5) was dealt with using mode imputation. Statistical assumptions were checked. Select items on the BPSS were reverse-coded so that higher scores indicated higher levels of body dissatisfaction. The effect sizes for the mediation models were calculated according to recommendations by Kenny25 and Preacher and Kelley.26 Partial correlations were first computed and then, an effect size of .15 (95% CI [.12, .37]) was obtained for the model involving media influence (X), thin-ideal internalization (M), and body dissatisfaction (Y). indicating that 15% of the variance in
6
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT body dissatisfaction was attributable to the overlap of media influence and thin-ideal internalization. Main analysis The bootstrapping method has been used instead of a traditional mediation model to adequately deal with unmet assumptions of traditional mediation.27 This method involved four
Downloaded by [University of Cambridge] at 21:32 05 November 2015
steps: (1) A bootstrap sample of 80 cases was generated from the study sample of 80; (2) the regression coefficient b and indirect effect a was calculated based on this sample; (3) 1,000 estimates of the indirect effect of interest were obtained by repeating steps 1 and 2 1,000 times; (4) the mean of the 1,000 estimates from step (3) was calculated. Significance values and the presence or absence of zero in a 95% CI indicated whether the indirect effects were statistically significant (See Table 1) . The z-score for the specific indirect effect of family influence on body dissatisfaction through the mediator of thin-ideal internalization was non-significant, which was further confirmed by the presence of zero in the 95% confidence interval (see Table 1). Thus, contrary to study hypotheses, family influence was not related to thin-ideal internalization, t = .99, p = .33, and thin-ideal internalization was not related to body dissatisfaction, t = .48, p = .63. The z-score for the specific indirect effect of media influence on body dissatisfaction through the mediator of thin-ideal internalization was statistically significant, indicating that a 1point increase in media influence was associated with a .48 increase in body dissatisfaction through level of thin-ideal internalization (see Table 1). Thus, greater media influence was associated with greater thin-ideal internalization, t = 4.83, p < .001, and greater thin-ideal internalization was associated with greater body dissatisfaction, t = 4.06, p
