South Asian women in the United Kingdom: the role of skin colour dissatisfaction in acculturation experiences and body dissatisfaction

South Asian women living in Western cultures may experience skin colour dissatisfaction, as fair skin is an important South Asian appearance ideal, whilst visible ethnic differences in their skin colour may lead to appearance-related ethnic teasing from members from the mainstream culture. This study investigates whether appearance-related ethnic teasing is indirectly associated with body dissatisfaction via skin colour dissatisfaction and explores the relationship between appearance-related ethnic teasing, cultural identifi- cation and skin colour dissatisfaction amongst first-generation South Asian women living in the United Kingdom. South Asian women ( N = 98; 18–55 years, M = 24.60) completed an online questionnaire that measured appearance-related ethnic teasing, skin colour dissatisfaction, cultural identification, and body dissatisfaction. Appearance-related ethnic teasing was indirectly linked with greater body dissatisfaction via greater skin colour dissatisfaction. Appearance-related ethnic teasing was linked with stronger British identification, a greater sense of having an integrated identity and greater skin colour dissatisfaction. South Asian identification was associated with greater skin colour dissatisfaction. These findings suggest that skin colour dissatisfaction is an important link between appearance-related ethnic teasing and acculturating South Asian women’s body image.


Introduction
In South Asian cultures, the pursuit of fair skin might be linked with skin colour dissatisfaction, which often leads to harmful behaviours such as skin lightening (Harper & Choma, 2019), as having fair skin is associated with higher caste (Ayyar & Khandare, 2013;Gupta, 2000) and marriage prospects (Nagar, 2018). Particularly for South Asian women living in Western cultures, constant exposure to White women with fair skin might reinforce the fair skin ideal (Chapkis, 1986;Clayton, 2021). Further, research suggests that South Asian women who perceived larger discrepancies between their skin tone and the White ideal tended to have a greater desire for lighter skin (Sahay & Piran, 1997). This could lead to skin colour dissatisfaction, which might be associated with body dissatisfaction (Bond & Cash, 1992;Goel et al., 2021); yet, research examining these relationships amongst South Asian women is limited (e.g., Harper & Choma, 2019;Swami et al., 2013). Thus, the current study examines the relationships between levels of acculturation, skin colour dissatisfaction and body dissatisfaction amongst South Asian women in the United Kingdom (UK). We define South Asia as consisting of Afghanistan, Bangladesh, Bhutan, India, Pakistan, Maldives, Nepal and Sri Lanka.
In addition, acculturating women's (i.e., women who have moved to another culture) body image is also shaped by experiences in their host country, such as appearance-related ethnic teasing. Specifically, this refers to experiences of teasing based on appearance due to ethnic differences, including clothing and physical features (e.g., hair and skin). Initial research suggests that appearance-related ethnic teasing was associated with skin colour dissatisfaction amongst Singaporean Indian women 3 (Pillai & Sündermann, 2020), and body dissatisfaction amongst South Asian women in the United States (US; Reddy & Crowther, 2007;Sahi Iyer & Haslam, 2003). However, no research thus far has explored the role of skin colour dissatisfaction in the relationship between teasing and body dissatisfaction amongst acculturating South Asian women.
One mechanism that could potentially link teasing and skin colour dissatisfaction is cultural identification, both with the https://doi.org/10.1016/j.bodyim.2022.07.007 1740-1445/© 2022 The Author(s). Published by Elsevier Ltd. CC_BY_4.0 mainstream culture (e.g., British) and the ethnic culture (e.g., South Asian). Social identity theory (Tajfel & Turner, 2004) suggests that individuals' social identity (e.g., South Asian cultural identity) may be threatened when members of their ingroup are devalued. As such, people's identification to their cultural identity (i.e., cultural identification) may be affected following discriminatory experiences, such as teasing. Previous research found that perceived rejection (in a Romanian and Morrocan sample; Badea et al., 2011) and discrimination (in a Russian and Estonian sample; Jasinskaja-Lahti et al., 2009) were negatively associated with people's mainstream cultural identification. However, when appearance-related ethnic teasing was examined specifically, this was associated with greater cultural conflict between South Asian women's mainstream and ethnic identities (Reddy & Crowther, 2007). Reddy and Crowther (2007) argued that appearance-related ethnic teasing might have led women to adopt mainstream appearance ideals as an attempt to avoid teasing, thus increasing their mainstream cultural identification. Supporting this, a recent qualitative study suggested that South Asian women who experienced appearance-related ethnic teasing preferred adopting assimilation an acculturation strategy (Goel et al., 2021). However, cultural identification was not measured in these studies. As such, it is possible that appearance-based and non-appearance-based rejection may be differently associated with cultural identification, as appearance is a visible cue that displays cultural differences. However, the associations between teasing, cultural identification, and skin colour dissatisfaction have not yet been examined.
Cultural identification (with the ethnic or mainstream cultures, or both) has implications for acculturating individuals' wellbeing (Berry, 2006). Particularly for first-generation South Asian women in the US, greater identification to their ethnic identity acted as a buffer to discrimination (Inman et al., 2014). However, when considering body image, the opposite might be plausible. Given the importance of fair skin in the South Asian culture (Bakhshi & Baker, 2011), women's identification with the South Asian identity (i.e., SA identification) might be associated with greater skin colour dissatisfaction. Similarly, South Asian women's identification with the British identity (i.e., British identification) might also result in greater skin colour dissatisfaction, as distinct differences in physical features exist between the appearance of ethnically diverse women and White women (e.g., skin tone; Cheney, 2011). However, previous research suggests some benefits of adopting an integrated identity (i.e., strong identification with the mainstream identity and one's ethnic identity; Berry & Hou, 2017), including lower levels of pathological eating (Furnham & Patel, 1994). Extending this to body image, an integrated identity may act as a buffer to the negative body image effects associated with each cultural identity. Nevertheless, it is worth noting that some research examining wellbeing outcomes beyond the scope of body image found both positive and negative psychological outcomes of adopting an integrated identity (e.g., Brown et al., 2013). As such, the relationship between cultural identification and skin colour dissatisfaction amongst acculturating South Asian women in the UK remains unclear.
The current study aims to investigate the indirect relationships between teasing 4 and body dissatisfaction via skin colour dissatisfaction, and to explore the relationships between teasing, cultural identification and skin colour dissatisfaction. As acculturation and wellbeing outcomes are moderated by factors such as generation status (Inman et al., 2014), the current study uses a cross-sectional design with first-generation South Asian women living in the UK. Given the influence of socioeconomic status (SES) on the adoption of Western appearance ideals in non-Western cultures (Swami et al., 2010) and the associations between caste hierarchies and skin colour preferences, we controlled for subjective SES throughout. We predicted that higher levels of teasing would be indirectly associated with greater body dissatisfaction via greater skin colour dissatisfaction (H1). We also hypothesised that experiences of teasing would be linked with stronger British identification (H2a), in line with Reddy and Crowther (2007) suggestions, and British identification would be associated with greater skin colour dissatisfaction (H2b), due to the unattainable nature of the Western ideal for ethnically diverse women (Cheney, 2011). Given the cultural importance of fair skin (Bakhshi & Baker, 2011), we predicted that SA identification would be associated with greater skin colour dissatisfaction (H2c). Finally, we hypothesised that having a stronger sense of an integrated identity would be associated with lower skin colour dissatisfaction (H2d), as previous acculturation research suggested its role as a buffer to negative wellbeing outcomes (Berry & Hou, 2017).

Procedure
Following institutional ethical approval, South Asian women over 18 years old, who were born in South Asia and have lived in the UK for at least 18 months were recruited. This criterion was selected based on changes in attractiveness preferences previously found amongst migrants who had lived in the UK for 18 months (Tovee et al., 2006); hence, we expect any appearance-related acculturation effects to have developed after this period. Recruitment was conducted before the Covid-19 pandemic through the university participant pool, social media and via snowball sampling, with the incentive of being entered into a £ 25 prize draw. Individuals with a previous or current eating disorder were advised not to take part due to the sensitive nature of the questionnaire. Participants provided informed consent, completed the questionnaire online via Qualtrics and were debriefed after completion.

Measures
Data 5 were collected as part of a larger project. The measures relevant to the current study are detailed below.

Appearance-related ethnic teasing
Teasing was measured using the 'frequency of teasing' subscale from the Measure of Ethnic Teasing (MET; Reddy & Crowther, 2007), which assessed how often people experienced teasing on their appearance (e.g., skin colour, body hair, clothing, etc.) since moving to the UK. The subscale consisted of eight items (e.g., "Since you moved to the UK, did other people make jokes about your hair because of your race or ethnicity?"), rated on a five-point scale (Never to Frequently). These scores were averaged with higher scores indicating greater experiences of teasing. The MET demonstrated high reliability (α = 0.81) with previous acculturating South Asian samples (e.g., Reddy & Crowther, 2007).

Cultural identification
Two single-item measures were used to assess SA identification and British identification (e.g., "I identify with the British identity"). Participants indicated their responses on a seven-point scale (ranging from very strongly disagree to very strongly agree). The continuous integrated identity variable was calculated by multiplying the scores of both items (Baysu et al., 2011), reflecting the extent to which an integrated identity was adopted. Single-item measures demonstrated good validity in previous social identification (Postmes et al., 2013) and acculturation research (Baysu et al., 2011).

Skin colour dissatisfaction
Skin colour dissatisfaction was measured using an item from the MET (Reddy & Crowther, 2007). Participants were asked the question "how often do you wish you could change your skin colour?" and this specifically referred to their desire after moving to the UK. Participants rated this on a five-point scale (Never to Frequently). Single-item measures demonstrated good validity in previous body image research (Frederick et al., 2016;Sandhu & Frederick, 2015).

Body dissatisfaction
Body dissatisfaction was measured using the Body Satisfaction Scale (BSS; Slade et al., 1990), Participants rated how satisfied they were with 14 areas of their bodies (e.g., face, eyes, chest, tummy, etc.) on a seven-point scale (ranging from very satisfied to very unsatisfied). These scores were reverse coded, such that a higher score reflected greater body dissatisfaction, and were averaged to provide an overall dissatisfaction score. The BSS demonstrated good reliability (α = 0.90) and has been previously used in acculturation research recruiting South Asian samples (e.g., Furnham & Adam-Saib, 2001).

Demographic measures
Subjective SES was measured using the MacArthur Scale of Subjective Social Status (Adler et al., 2000). Participants positioned themselves on an imaginary ladder with ten rungs that represents their position in their own country, with higher scores reflecting higher subjective SES. Participants also provided their age, length of residence in the UK and country of birth.

Data analysis
We ran the following mediation models using Hayes (2014) PROCESS v3.5 SPSS, with 5000 bootstrapped samples, using the percentile method and a HC3 estimator (Hayes & Cai, 2007). Listwise deletion was used to deal with missing data within each model. Correlations were computed using Spearman's rho due to non-normality in the data, with pairwise deletion to deal with missing data. Subjective SES was controlled for throughout the analysis.
To examine the indirect relationship between teasing and body dissatisfaction via skin colour dissatisfaction (H1), we ran a mediation model (Model 1; Fig. 1) where teasing was directly associated with skin colour dissatisfaction and body dissatisfaction, and skin colour dissatisfaction was directly associated with body dissatisfaction.
To establish the relationships between teasing and cultural identification, and cultural identification's association with skin colour dissatisfaction (H2a-d), we ran a mediation model (Model 2; Fig. 2) where teasing was directly associated with British identification, SA identification and integrated identity, and these cultural identification variables were directly associated with skin colour dissatisfaction. The direct association between teasing and skin colour dissatisfaction was also included. Table 1 contains the means, standard deviations, and correlations of all the variables after excluding two outliers 6 (N = 96). Controlling for subjective SES (M = 6.44, SD = 1.94) in Model 1 (Fig. 1), teasing was positively associated with skin colour dissatisfaction (β = 0.33, 95 % CI [.17,.91], R 2 =.11), with experiences of teasing linked with greater skin colour dissatisfaction. Skin colour dissatisfaction was positively associated with body dissatisfaction (β = 0.25, 95 % CI [.11,.63], R 2 =.16), suggesting that women who were dissatisfied with their skin colour experienced greater body dissatisfaction. Teasing and body dissatisfaction were not directly associated (β = 0.11, 95 % CI [−0.28,.84]) but there was an indirect association via skin colour dissatisfaction (b =.21, 95 % CI [.04,.44]). There was a significant relationship between subjective SES as a covariate and body dissatisfaction (β = −0.22, 95 % CI [−0.26, −0.04]), with higher subjective SES linked with lower body dissatisfaction.

Discussion
Previous research suggests that appearance-related ethnic teasing was associated with body dissatisfaction and skin colour dissatisfaction amongst acculturating (Reddy & Crowther, 2007) and non-acculturating (Pillai & Sündermann, 2020) South Asian women. Our study adds to existing literature by establishing the role of skin colour dissatisfaction as an important link between teasing and body dissatisfaction, supporting our hypothesis (H1). This suggests that South Asian women's body image was associated with the mainstream culture's reaction to them. As hypothesised, we also found that stronger SA identification was associated with greater skin colour dissatisfaction (H2c), which suggests that South Asian women with stronger SA identification might have a greater desire to change their skin colour. As previous research found South Asian women's desire to have lighter skin positively related to the discrepancy between their skin tone and the White ideal (Sahay & Piran, 1997), it is possible that South Asian women's desire to change their skin colour was related to wanting lighter skin tones. However, we cannot confirm this as desire for lighter skin specifically was not assessed in our study. Nevertheless, these findings highlight the importance of skin colour to South Asian women and suggest that South Asian women's body image is shaped by both internal (i.e., South Asian identification) and external (i.e., teasing) processes. However, it is worth noting that teasing explained a low proportion of the variance in skin colour dissatisfaction in our study, meaning that there are likely other contributing factors. Future research should seek to also take into account other non-acculturative factors, such as marital status (Nagar, 2018), family comments (Sharda, 2020), sibling complexion (Rondilla & Spickard, 2007) and exposure to the media (Hussein, 2010).
In contrast to previous findings that found perceived rejection negatively associated with mainstream cultural identification (Badea et al., 2011), our findings suggest that teasing was associated with greater British identification. This provides support for our hypothesis (H2a) and is consistent with findings that linked teasing with conflicting cultural values (Reddy & Crowther, 2007). Since teasing targets visible ethnic differences in appearance, South Asian women who were teased might try to fit in by increasing their identification with the mainstream cultural identity and adopting its appearance norms (Cheney, 2011). It would be useful in future research to measure both appearance-based and non-appearance-based rejection to further explore these differences.
We additionally found teasing associated with greater adoptions of an integrated identity, which was not initially hypothesised in our study. However, it is possible that adopting an integrated identity potentially reflected the conflict in cultural values found by Reddy and Crowther (2007). Considering that SA identification did not vary significantly with experiences of teasing in our study, it is possible that stronger British identification associated with teasing, on top of women's SA identification might have resulted in the adoption of both identities (i.e., integrated identity). A recent study found that South Asian women living in the US utilised code switching (i.e., assimilating more with either their ethnic or the mainstream culture depending on contexts; Goel et al., 2021) to avoid scrutiny, which potentially explains the adoption of an integrated identity when navigating discriminatory experiences such as teasing. However, future studies are needed to probe these possibilities, using multiple-item and multifaceted (e.g., values, relationships and behaviours) measures of integration.
Contrary to our hypotheses (H2b and H2d), we found no evidence for any relationships between skin colour dissatisfaction and British   Note. SA identification = South Asian identification; β = standardised regression coefficients; B = unstandardized regression coefficients; SE B = standard errors for the unstandardised regression coefficients; CI = confidence interval; LL = lower limit; UL = upper limit. Bootstrapped model parameters are shown, controlling for subjective SES.
identification or the integrated identity. Previous research on pathological eating suggests that the extent to which South Asian girls were adapted to the host culture was a better predictor than group membership alone (Furnham & Patel, 1994). Therefore, it is possible that skin colour dissatisfaction might be linked with the extent to which individuals immerse themselves in the mainstream and their ethnic culture (thus being exposed to their respective appearance ideals), more than the extent to which they identify with these cultural identities. Future studies should also examine cultural immersion (i.e., the extent to which people engage in cultural behaviours and practices). Our findings have practical implications. For instance, body image interventions for women in the UK should focus on both internal and external aspects (e.g., de-emphasising the importance of skin colour as a cultural ideal, as well as promoting racial/ethnic inclusivity), in order for them to also be effective for South Asian women. Nevertheless, several limitations should be considered. First, we cannot draw any causal inferences, as our dataset is crosssectional. Second, we note that two of the four major variables in our study (i.e., skin colour dissatisfaction and cultural identification) were measured using single-item measures, despite having been used in previous research (e.g., Baysu et al., 2011). Whilst we have addressed issues with skewness statistically using bootstrapping, future research would benefit from assessing these variables using multiple-item measurement scales for greater reliability and validity. Third, we were not able to robustly consider other covariates due to our sample size. Future research with a larger sample should seek to include covariates, such as appearance-related pressures from parents, length of residence and age. Fourth, we did not consider sub-cultural differences in appearance ideals between the South Asian countries, as the group sizes between countries were highly unequal in our study. Nevertheless, our sample of first-generation South Asian women living in the UK has highlighted some important findings on acculturation and body image.

Conclusion
Appearance-related ethnic teasing was associated with acculturating South Asian women's body image, as it was found to be linked with greater skin colour dissatisfaction, stronger British identification and stronger integrated identity directly, and with greater body dissatisfaction indirectly via skin colour dissatisfaction. Whilst future research should seek to investigate other acculturation experiences that might contribute to body image, our study offers initial evidence on the importance of skin colour dissatisfaction as an important link between appearance-related ethnic teasing and body image amongst first-generation South Asian women living in the UK.

Data Availability
The data that supports the findings of this study are openly available in the Open Science Framework at https://osf.io/qtgx5/ view_only=1d8f90293887426e96db731aefc20b28