Examining Identity Intersectionality: Thai Marriage Migrants in the Netherlands

of intersectionality was developed for understanding how cumulative structural disadvantages, such as institutional racism and legal practices, affect minority people’s lives (e.g., Steffensmeier, Painter-Davis & Ulmer, 2017). Intersectionality theory emphasizes that for understanding the social experiences of minority members one should consider how different axes of stratification interrelate (Cole, 2009; Collins, 2015; Crenshaw, 1989). Intersecting minority positions would create unique negative experiences and would involve accumulated disadvantages in the form of double jeopardy based on multiple minority identities (Berdahl & Moore, 2006; King, 2003). For example, many studies have focused on the combination of race/ethnicity and gender, and the oppression of Black women is considered to differ from the oppression experiences of Blacks and of women (Crenshaw, 1989; Jordan-Zachery, 2007). Psychological conceptions of intersectionality focus on the individual meanings of belonging to multiple disadvantaged groups (Syed, 2010). Social psychologists are increasingly interested in multiple identities and the ways in which these identities are subjectively combined and organized (e.g., Amiot, de la Sablonnière, Terry & Smith, 2007; Roccas & Brewer, 2002: Settles & Buchanan, 2014). Psychological research on intersectionality tends to have a more qualitative focus or investigates statistical patterns in quantitative data and in relation to multiple group memberships (see Cole, 2009; Settles & Buchanan, 2014). Although the former approach tends to provide an indepth understanding, there typically is no systematic test. And in relation to the latter approach, it is argued that it is ‘a mistake to reduce the nuanced theoretical concept of intersectionality to include only the type of associations that can be modelled through the use of interaction effects’ (Cole, 2009, p. 177). Thus, we proceeded with mixed-methods, using the sequential combination of qualitative and quantitative approaches. Furthermore, to our knowledge there is no research that has used a systematic comparative design to examine whether intersectional identity (e.g., Black women) has different meanings compared to the component identities (Black or women). The current research tries to make a novel contribution to intersectional literature by using such a design for investigating in a qualitative (Study 1) and quantitative (Study 2) way the understandings that Thai female migrants in the Netherlands have about their intersectional identity (being a Thai woman) compared to their ethnic identity (being Thai) and their gender identity (being a woman). This design allows us to examine in a systematic way whether unique meanings and negative experiences are attached to the intersectional identity of ethnicity and gender. Furthermore, the focus on Thai women contributes to a further understanding of this relatively neglected group of immigrants and of female marriage migrants more generally. Marriage migration is closely connected to globalisation processes and occurs in many parts of the world. But in contrast to migration studies and anthropology (see Brettel, 2017) there is hardly any psychological research on marriage migrants; therefore, we do not know much about how marriage can affect the lives of migrant women and their sense of self.

In sociology, and in feminist and legal studies, the notion of intersectionality was developed for understanding how cumulative structural disadvantages, such as institutional racism and legal practices, affect minority people's lives (e.g., Steffensmeier, Painter-Davis & Ulmer, 2017). Intersectionality theory emphasizes that for understanding the social experiences of minority members one should consider how different axes of stratification interrelate (Cole, 2009;Collins, 2015;Crenshaw, 1989). Intersecting minority positions would create unique negative experiences and would involve accumulated disadvantages in the form of double jeopardy based on multiple minority identities (Berdahl & Moore, 2006;King, 2003). For example, many studies have focused on the combination of race/ethnicity and gender, and the oppression of Black women is considered to differ from the oppression experiences of Blacks and of women (Crenshaw, 1989;Jordan-Zachery, 2007).
Psychological conceptions of intersectionality focus on the individual meanings of belonging to multiple disadvantaged groups (Syed, 2010). Social psychologists are increasingly interested in multiple identities and the ways in which these identities are subjectively combined and organized (e.g., Amiot, de la Sablonnière, Terry & Smith, 2007;Roccas & Brewer, 2002: Settles & Buchanan, 2014. Psychological research on intersectionality tends to have a more qualitative focus or investigates statistical patterns in quantitative data and in relation to multiple group memberships (see Cole, 2009;Settles & Buchanan, 2014). Although the former approach tends to provide an indepth understanding, there typically is no systematic test. And in relation to the latter approach, it is argued that it is ' a mistake to reduce the nuanced theoretical concept of intersectionality to include only the type of associations that can be modelled through the use of interaction effects' (Cole, 2009, p. 177).
Thus, we proceeded with mixed-methods, using the sequential combination of qualitative and quantitative approaches. Furthermore, to our knowledge there is no research that has used a systematic comparative design to examine whether intersectional identity (e.g., Black women) has different meanings compared to the component identities (Black or women). The current research tries to make a novel contribution to intersectional literature by using such a design for investigating in a qualitative (Study 1) and quantitative (Study 2) way the understandings that Thai female migrants in the Netherlands have about their intersectional identity (being a Thai woman) compared to their ethnic identity (being Thai) and their gender identity (being a woman). This design allows us to examine in a systematic way whether unique meanings and negative experiences are attached to the intersectional identity of ethnicity and gender. Furthermore, the focus on Thai women contributes to a further understanding of this relatively neglected group of immigrants and of female marriage migrants more generally. Marriage migration is closely connected to globalisation processes and occurs in many parts of the world. But in contrast to migration studies and anthropology (see Brettel, 2017) there is hardly any psychological research on marriage migrants; therefore, we do not know much about how marriage can affect the lives of migrant women and their sense of self. tool for social justice projects (Collins, 2015). In (social) psychology, intersectionality is considered an important approach for social justice and equity (Rosenthal, 2016) and for investigating whether and how the combination of minority group memberships has unique social and personal meanings (Cole, 2009;Settles & Buchanan, 2014;Syed, 2010). For example, compared to White women, women of colour have been found to make a stronger and more meaningful connection between race/ethnicity and gender (Juan, Syed & Azmitia, 2016). How individuals understand themselves and perceive their disadvantages from an intersectional position (e.g., Thai women) can differ from the separate categories (Thai or women) that comprise the intersection. Thus, the categories forming an intersection can have meanings on their own (McCall, 2005;Weldon, 2006). One reason for these separate meanings is the different comparisons that are (implicitly) made. Social identity theory (Tajfel & Turner, 1979) emphasises the importance of intergroup comparisons for the meaning and value of a particular social identity. Comparison groups tend to be reciprocally paired, as in malefemale and black-white. This means that in our research we can expect that thinking about one's female identity will trigger comparisons with males and that thinking about Thai identity involves the Dutch. More importantly, it also can be expected that an intersectional position elicits other comparisons than the two separate identities and, thereby, other meanings.
For example, in research in the United States on perceived cultural stereotypes, it was found that there were unique stereotypes for ethnicity-by-gender that were not the result of adding ethnic stereotypes to gender stereotypes (Ghavami & Peplau, 2013). And in a study in New York City (see Deaux, 2006), Latina women were asked to describe themselves as women, as ethnics (predominantly Dominican), and then as ethnic women. For each of the three identities, different contrasts were made and different answers were given. Being a woman was described in terms of biological features and especially motherhood. The ethnicity question was answered by referring to cultural traditions, language, and pride in the country of origin. The intersectional category of Dominican women elicited many positive descriptions, such as strength, intelligence, independence, and passion, which were rarely given for either single label. However, it was also noted that as Dominican women they had to deal with discrimination by Americans, as well as Hispanic men. Another example is a study conducted by Heyse (2010) in Belgium among Russian-speaking female marriage migrants. In this study, it was concluded that intersectionality theory is a valuable tool in uncovering the specific negative stereotypes and discrimination that these women face.
Most social categories relate to structures of social stratification, which implies that some categories come with advantages, while others bring disadvantages. Dubrow (2008) argues that intersectional positions typically involve multiple disadvantaged categories (e.g., ethnic minority and female) that have negative psychological and behavioural consequences that differ from those of each category on its own. Studying Thai women in the Netherlands allows us to uncover meanings given to the ethnic and gender identities separately, as they are both inscribed in distinct structures of social stratification. In Western countries, there is gender inequality in many spheres of life (e.g., economic, political), and ethnic discrimination is present in a multitude of contexts (Wrench, 2016). Importantly, the intersection of gender and ethnicity might imply personal meanings and perceived disadvantages that differ from the ones attributed to gender identity and ethnic identity separately. An investigation in Korea showed that female marriage migrants often suffer from a sexualized image that hampers their upward social mobility (Lee, 2010; see also Faier, 2007).

Examining intersectionality
Research in this area has not tested the proposition of unique meanings and perceived disadvantages with a systematic comparative design (but see Deaux, 2006). We used such a design in both a qualitative (Study 1) and quantitative study (Study 2). In both studies, we randomly asked participants to reflect on and answer questions about what it means to be either a woman, a Thai, or a Thai woman. This allows us to systematically investigate the differences between an intersectional identity and the two separate identities and, thus, to assess how an intersecting identity influences self-conception and meaning making.
Intersectionality has been researched in a myriad of ways. Some researchers have analyzed survey data (Simien, 2006), others have analysed in-depth interviews (Hancock, 2004), investigated narratives (Arendt, 1998), or resorted to discourse analysis (Jordan-Zachery, 2008). There is quite a debate about the most appropriate method for studying intersectionality (e.g., Bowleg, 2008;Diamond & Butterworth, 2008;Dubrow, 2008;Shields, 2008), which, among epistemological issues, has to do with the question whether an additive approach (statistical interaction) is appropriate for examining intersectionality. Some studies indicate that such an approach is compatible with an intersectional framework (Cuadraz & Uttal, 1999;Juan et al., 2016), and some researchers have argued for the use of mixed-methods (Harper, 2011;Perkins, 2001). In our research, we used semi-directive interviews (Study 1) to uncover what it means to be a Thai woman in comparison to being Thai and being a woman. The findings from these interviews subsequently provided the input for the construction of a questionnaire (Study 2) to examine the generality of the findings across methods and samples. Thus, we used a sequential exploratory mixed-methods approach in which a phase of qualitative data collection and analysis was followed by a phase of quantitative data collection and analysis (Creswell, 2003). This is an appropriate strategy for exploring a relatively new phenomenon and for examining the generality of particular findings.
In the Netherlands, Thai immigrants represent the largest group immigrating with the purpose of marrying (48%; Sterckx, Dagevos, & Huijnk, 2014). Furthermore, it is the immigrant group with the proportionally largest number of females and the group that, on average, is the oldest at the time of arrival (two-thirds of the immigrants are over 30 years old). Additionally, it is proportionally the largest migrant group for which the marriage partner is Dutch (78%). In 2017, there were approximately 20,000 Thai people in the Netherlands, of which some 14,000 were female (of which 11,000 were first generation).

Study 1 Method
Participants were contacted between October and December 2016 at the Wat Buddhavihara, a Thai temple in the Dutch town of Purmerend. The temple is a relevant location to meet Thai female migrants because people from all socioeconomic backgrounds and occupations attend get-togethers in such a location (Suksomboon, 2009). In order to recruit respondents, we made three Sunday visits to the temple, and almost all women approached were willing to make time to be interviewed and/or introduced us to other women. The first (female, non-Thai) author of this article conducted thirty in-depth interviews, ten for each of the three identity categories. Six interviews were conducted by means of a video-call, and these respondents were either friends or colleagues of the women interviewed face to face.
The women interviewed were between 35 and 55 years old (M = 42, SD = 6.86), and their length of residence in the Netherlands varied between 5 and 35 years (M = 13.13, SD = 8.21). Twenty-five respondents were married to Dutch men, three were widowed (previously married to Dutch men), and two had been married to Dutch men but were divorced at the time of the interview. The final sample constituted of a heterogeneous group of Thai female marriage migrants in the Netherlands.

Study design
Before the start of the interview, each respondent was randomly assigned to one of the three 'identity groups': ethnic (Thai) identity, gender (female) identity, or intersectional (Thai woman) identity. The respondents assigned to the ethnic identity group were told that the research investigated the lives of Thai people in the Netherlands. The respondents in the female identity group were told that the research investigated the lives of women in the Netherlands. Finally, the respondents in the intersectional identity group were told that the research investigated the lives of Thai women in the Netherlands. With these three groups of respondents, open-ended interviews were conducted, and all respondents agreed for the interviews to be recorded. The average time of each interview was 40 minutes. A topic list was used that served as open-ended prompts, and each interview consisted of three parts: settling in the Netherlands, identity meanings, and marriage.

Data analysis
After transcribing the interviews verbatim, an inductive thematic analysis was performed (Kellehaer, 1993;Krippendorff, 2004). This approach is suitable for examining whether and how the broad meanings of the different identities differ. The transcripts of the interviews were first carefully read to identify narrative themes that emerged from the data. Subsequently, the transcripts were loaded into ATLAS.ti software and categorized using in vivo and emergent classification. The in vivo approach makes it possible to use the concepts as described by the respondents themselves, and the emergent aspect allows the identification of new concepts and narratives that can emerge while examining the data. The resulting themes were intensively discussed by the researchers in order to agree upon the core narrative in the interviews (Kellehear, 1993).

Results
Participants talked in the interviews about many different things, but in this paper, we focus on identity meanings and whether and how these differ between intersectional identity and the two separate identities. We will first discuss some of the key findings that are typical for the latter two identities and then for intersectionality.

Ethnic identity
When asked about what it means to be Thai, the participants (explicitly or implicitly) made a comparison with the Dutch and mentioned two main distinguishing cultural values: respect and generosity. Respect was mentioned in all the interviews as a very important attribute of being Thai. It referred to being respectful (modest and reserved) and was related to, for example, age differences and personal privacy. Kanya says the following: 'The most important aspect of being Thai are the norms and values. I mean respecting each other. Here, in the Netherlands, everybody has the same position in society. In Thailand, if someone is older than yourself, there is a word that you put in front of the name in order to show respect. For instance, my name is Kanya, and I am older than you, so you must call me Pi-Kanya. This is the most important part of my culture, and for me it means a lot.' The concept of being generous and offering help was the second characteristic that was used for describing what it means to be Thai. It was typically compared and contrasted with what the respondents described as the rather individualistic Dutch culture: 'Thai people always care for each other and help each other. Especially for parents or elderly people. In the Netherlands, it feels like everyone is alone' (Maluk).
For Maluk, the lack of generosity in the Netherlands contributed to feelings of isolation and lower chances of integrating into Dutch society. In contrast, respondents described how the generosity of fellow Thai migrants had contributed to settling in the Netherlands and coping with the initial stress of immigration.
Apart from cultural values, three practices were mentioned as enactments of Thai identity that symbolize what it means to be Thai compared to the Dutch. First, participating in Thai social gatherings on specific celebrations and holidays was mentioned. Second, Buddhism and going to the temple was discussed as an important aspect of being Thai. Respondents emphasized the importance of personal prayer at home and that the Buddhist temple represents a place for meeting other Thai and for praying for their family in Thailand and for Thailand as a country. Third, Thai food and cooking was considered an important marker of Thai identity and was sometimes presented as the most typical characteristic of Thai identity. Many Thai migrants work in Thai restaurants, and almost all of the women interviewed regularly cooked Thai food at home. Thinking about what it means to be Thai elicited various stories about Thai food as a central characteristic of Thai culture and identity.

Gender identity
When being asked about their female identity, the participating women were generally more hesitant and unsure about their answer. Compared to their Thai identity, the female identity was more difficult to talk about and with greater diversity in meanings. Interestingly, however, for these marriage migrants, the topic of female identity was almost unanimously interpreted in terms of one's role as a wife. These women migrated to the Netherlands for marriage, and being a wife was presented as one of the most meaningful aspects of their female identity. Being a wife inevitably involved a Dutch husband, and their role as a spouse was typically ' ethnicized', in the sense that they talked about themselves as a Thai wife.
The division of household duties and the relationship of trust between husband and wife were discussed as the most important aspects of their role as wife in the Netherlands. Almost all respondents stated that they took care of most of the household tasks, arguing that this is part of Thai culture and, thereby, differs from western culture. They explained that they did not feel any pressure to carry out these tasks but did it out of love and care for their family. The ' caring wife' was an important component of how they defined their womanhood, and it was predominantly discussed when asked about their gender identity and not when asked about the intersectional identity. Respondent Malee, who has been in the Netherlands for 12 years, reported how deeply engrained Thai wifely customs can be. Based on her own migratory experience, as well as the ones of Thai migrants in her network, she claimed: Being a wife implied a husband, with a further reported distinction between Dutch husbands and Thai men. Dutch husbands were described as being responsive, supportive, and thoughtful, and this was contrasted with Thai men, who were considered more secretive and selfcentered: 'Here the man respects the woman, and the man is trustworthy. Much more than the Thai man. For instance, when I come home my husband will tell me everything he has done during the time that I was at the temple. And he explains many days in advance what he is planning to do. In Thailand often men do not tell much, and that is not fair.' Furthermore, a Dutch husband was described by the majority of the respondents as an important facilitator of integration into Dutch society. He would guide his Thai wife toward learning the language and introduced her to Dutch family and friends and, consequently, to Dutch customs and practices.

Intersectional identity
When asked about being a Thai woman, this identity was mainly interpreted in terms of their migration status and of their role as a mother. First, whereas respondents described their Thai identity as a possible impediment to employment and their female identity tended to imply a level of dependency on their Dutch husband, the identity at the intersection of the two unveiled a recurring discourse on employment status and professional accomplishments and the need to be flexible and capable of adapting. Earning wages, having a professional activity, and gaining a certain level of independence defined what it meant to be a Thai female migrant. In Hathai's words:

'As a Thai woman, The Netherlands started feeling like my home because of my job. And sometimes getting a higher position in the restaurant, more responsibility. I always wanted to work and that is also why I really wanted to learn the Dutch language. But I can also type without looking and I can use a calculator well. I have done that all by myself. My husband and his family are very proud of me. Life as a woman back home would be very difficult. Here I can be a woman, I can be an immigrant from Thailand, and I can achieve things too.'
Second, the intersectional identity was also discussed in terms of the stereotypical Thai female marriage migrant, with the related feelings of exclusion and discrimination. Feelings of being stereotyped and discriminated were less discussed with the separate ethnic and gender identities. May, who has been living in the Netherlands for 11 years and who holds a university degree, described her experience of discrimination as follows: 'I feel discriminated until now. When I send out my CV people look at my foreign name and do not give me a chance. In some interviews they criticize my Dutch language skills. But for an English speaking job they also don't take me, and I speak English fluently! Sometimes I feel that people have a special idea of the Asian woman married to a foreigner. Like the Thai lady who only goes with a foreigner to enter the country. They look down on that. And you can-not really tell who you are, because you don't get the chance, people just think, "oh they look like that".' Respondents said they rarely dealt with these experiences with anger or despair, but rather with increased effort and perseverance.
Third, being a Thai woman in the Netherlands was talked about in terms of raising one's children in a Thai way. The idea of the Thai mother was an important aspect of the intersectional identity, and being asked about their Thai identity resulted in explaining what a Thai mother typically does in terms of passing on Thai culture and monitoring her children. Preeda, who has been in the Netherlands for 12 years, explained:

Discussion
The findings of Study 1 indicate that some specific meanings emerged when talking about being a Thai woman compared to talking about being a woman or being a Thai. These differences can be understood in terms of the implicit comparisons that were made when discussing the different identities (Tajfel & Turner, 1979). Thai identity and the importance of Thai culture was mainly discussed in contrast to Dutch culture, female identity was discussed in relation to one's role as a wife of a Dutch husband, and Thai female identity was discussed in terms of being a migrant and the possibilities for a mother to pass on Thai culture to her children. Furthermore, the intersectional identity elicited stories about professional accomplishments and independence, but it also raised concerns about being stereotyped and discriminated against and, therefore, the need to be flexible and to adapt to the host society.
These findings were used in Study 2 to develop an on-line questionnaire for testing in a quantitative way whether gender, ethnic, or gender-by-ethnic identity salience results in differential endorsement of Thai cultural characteristics, the importance attached to care at home and to being a mother, perceived discrimination, and selfstereotypes. The latter refers to being reserved, which was mentioned in the interviews and which is a common stereotype of women from East Asian cultures (Verkuyten & De Wolf, 2002).

Study 2 Method
Three versions of a survey were developed with identical questions, differing only in the introduction, in which respondents were told that the survey was about either their ethnic, their female, or their intersectional identity. A social media platform (Facebook) was used in recruiting participants and assigning the surveys randomly to Thai female migrants in the Netherlands. Facebook is a relevant platform to reach these respondents as it constitutes a main communication channel for keeping in touch with each other, as well as with family and kin in Thailand. A closed Facebook group containing 2200 Thai people living in the Netherlands was used to share the links to the surveys. In order to randomly assign the three different versions of the surveys, respondents were presented with three links and they were asked to select the link that corresponded to their month of birth. As such, women born in January, April, July, or October answered the survey investigating the meanings of ethnic identity. Women born in February, May, August, or November answered the survey investigating the meanings of gender identity. And women born in March, June, September, or December answered the survey investigating meanings of intersectional identity.

Sample
Before the start of the study, it was decided to recruit a minimum of 50 respondents per condition, which is sufficient for a design with three conditions and only outcomes measured. In total, 154 female respondents participated in the research (N = 50, ethnic identity group; N = 53, female identity group, N = 51, intersectional identity group). The mean age of the respondents was 38 years old (SD = 7.65), 72% of them were married, and their length of stay in the Netherlands varied between 1 year and 31 years (M = 7, SD = 5.76). The respondents had between 0 and 3 children and were fairly highly educated (M = 3.36, SD = .83), according to the highest achieved level of education (1 = primary school, 2 = secondary school, 3 = lower tertiary education, 4 = higher tertiary education). There were no statistically significant differences (p s > 42) between the three identity categories of respondents for age, length of residence, being married, number of children, and level of education.

Measures
The endorsement of Thai cultural characteristics was measured with seven items using 5-point scales (rho = .57): 'How important is respect to you?', 'How important is the concept of generosity to you?', and 'How important is Thai food to you?' The importance attached to care at home was measured with five items using 5-point scales (rho = .75): 'How important is it for you to take care of your husband?', 'How important is it for you to take care of house duties?', and 'How important is being a wife to you?' For assessing the importance attached to being a mother, two items (5-point scales) were used: 'How important is being a mother to you?' and 'How important is it for you to raise your children in the way you want?' These two questions were associated (r = .52, p < .001).
Perceived discrimination was measured with two items on 5-point scales: 'Do you feel that you are sometimes treated negatively (because you are a Thai/a woman/a Thai woman)?', and 'I sometimes have the feeling that people do not perceive me as an individual person but rather in terms of group stereotypes' (r = .53, p < .001).

Results
Using structural equation modeling in MPlus 7, we tested for between-group differences (Thai identity, female identity, or intersectional identity) in scores for the different latent constructs. This analysis is preferred over a multivariate analysis of variance (MANOVA) as it provides a more reliable test by taking random error of measurement of the observed variables into account. We estimated between-group differences by using the ethnic identity group as a reference category and setting its mean to zero. Table 1 presents the differences in scores for the different latent constructs and across the ethnic, gender, and intersectional identity groups. For the endorsement of Thai culture, respondents in the intersectional group scored lower compared to respondents in the ethnic identity group. This is in line with what was found in the interviews. There were no significant differences between the gender group and the other two groups.
For the importance attributed to taking care of the home as a meaningful identity aspect, there was a significant difference between the ethnic and intersectional groups: respondents found this significantly less important when their intersectional identity was salient compared to their ethnic identity, and the mean score for the gender group was in-between the other two scores. In the interviews, it was found that respondents talked most about the importance of taking care of the home as part of their female identity. However, this domestic role also seems to provide meaning to the ethnic identity, and this might be in line with the finding in the interviews that the female identity is ' ethnicized'.
Perception of discrimination differed between the intersectional group and the other two groups. Respondents thinking about their intersectional identity of being a Thai woman felt significantly more discriminated against than respondents thinking about their Thai identity or their female identity. This is in agreement with the interviews, in which respondents were more prone to talk about being discriminated against in society when asked what it means to be a Thai woman compared to being a Thai or being a woman. In contrast to the interviews, no significant differences between the three groups were found for the importance of being a mother.
For being reserved, there was a significant difference between the ethnic and gender identity groups. Respondents saw themselves as more reserved when their ethnic identity was salient compared to when they were asked to think about their gender identity. This finding is in agreement with the interviews and indicates that cultural stereotypes, such as being reserved and showing respect, are considered meaningful aspects of Thai self-identity compared to the gender and intersectional identity.

Discussion
Social psychologists are increasingly interested in multiple identities and the ways in which these are psychologically organized and managed (Amiot et al., 2007;Roccas & Brewer, 2002). Intersectionality is one of the analytical perspectives to investigate whether and how membership in one social group shapes and defines membership in another social group (Cole, 2009;Settles & Buchanan, 2014). Furthermore, intersectionality theory proposes that individuals in two disadvantaged groups (e.g., ethnic minority and female) are at increased risk for negative experiences, such as stereotyping and discrimination (Rosenthal, 2016). Research on intersectionality has either a quantitative or a qualitative approach, and to our knowledge, there are no studies that use a systematic comparative design (but see Deaux, 2006). In our research, we used such a design to examine both qualitatively (Study 1) and quantitatively (Study 2) whether an intersectional identity has unique personal meanings compared to the two identities separately. We focused on Thai female marriage migrants in the Netherlands as an understudied immigrant group. The findings indicate that (some of) the meanings of the intersectional identity (Thai woman) differ from the separate identities (Thai or woman). Social identities are defined in contrast to relevant others (Tajfel & Turner, 1979), and the Dutch were the most obvious group of comparison when talking about Thai identity; whereas, the role of being a mother was more central in talking about Thai female identity. When thinking about their intersectional identity, and compared to their Thai identity, the participants emphasized Thai cultural characteristics less and found it less important to take care of the home. Furthermore, respondents invited to think about their intersectional identity felt more often stereotyped and discriminated against in society than respondents thinking about their ethnic or female identity. This finding is in line with the notion of double-jeopardy that characterizes the lives of individuals in two disadvantaged positions (Settles & Buchanan, 2014). Thai women are sometimes faced with the stereotype of the Asian marriage migrant who is considered greedy, lazy, emotionally indifferent, and who uses sex as a commodity. These sorts of negative stereotypes have also been found in other research among female marriage migrants (e.g., Faier, 2007;Heyse, 2010;Lee, 2010). It indicates that negative stereotypes of intersectional positions can differ from those of the categories that form the intersection. However, it is important to note in the interviews, intersectional identity was also discussed in relation to employment, professional accomplishments, and independence. This suggests that for some Thai women their intersectional identity provided a sense of agency and control. This indicates in the context of migration, an intersectional identity does not only have to imply a sense of double jeopardy, but also might be considered a challenge and a meaningful resource. This possibility could be examined further in future research among different migrant groups and in various social contexts.
A related and noteworthy finding is that the female identity was described in terms of being a married wife. Marriage migration and female identity seemed to be closely intertwined, to the extent that all respondents interpreted 'being a woman' in terms of 'being a wife' and described their migration experiences largely in light of the important role of their husband for integrating in Dutch society (Brettel, 2017). Migration for the sake of marriage seems to give a particular definition to what it means to be a woman, in which the role of being the wife of a Dutch husband is central. Furthermore, being a wife to a Dutch husband also made their ethnic background meaningful. One's identity as a wife was ' ethnicized' and understood as being a wife in the Thai way. However, the emphasis on one's role as a wife and also a mother might not only have to do with marriage migration. Research on women of color has demonstrated that personal experiences with ethnicity/race and gender frequently occur within the family as a site in which traditional gender role dynamics are played out (Mehrotra & Calasanti, 2010). The family is often the most salient and influential setting in developing an understanding about what it means to be a woman within one's ethnic group (Juan et al., 2016).
Future research could examine the ways in which marriage migration and/or the family context in general is relevant and important for the development of an intersectional sense of self.
Some limitations of the research should be mentioned. For example, the sample size in both studies was not very large, and the samples do not allow us to draw generalizable conclusions. Additionally, identity meanings were examined in terms of attributed importance and not in terms of clusters or patterns of meanings, which can also differ for specific combinations of identities (Pouliasi & Verkuyten, 2007). Furthermore, even though the respondents reported feeling confident in reading and speaking Dutch or English, conducting the interviews in Thai and having the surveys translated into Thai might have made it easier for the respondents to express themselves. However, the use of Thai language can also make Thai identity salient, which would make it more difficult to examine the social meanings of intersectionality compared to gender and ethnic identity. Thai identity salience is also a limitation of our research. It is likely that, for most women, their Thai identity was made implicitly relevant because they were approached for participation in a Buddhist temple, as belonging to a community of people originating from Thailand, and by a non-Thai person. This procedure was unavoidable but limits the possibility of finding other and stronger meaningful differences between intersectional identity and the other two identities and, thereby, the conclusions that can be drawn.
In spite of these limitations, this research adds to the limited empirical base of intersectional identities in (social) psychology (Juan et al., 2016;Ghavami & Peplau, 2013). By using a comparative and mixed-method design, the research provides a unique empirical contribution to the emerging psychological literature on intersectionality. Furthermore, in considering a specific combination of identities, we have responded to the call for investigating intersectionality among individuals with different combinations of group memberships in different societal contexts (Settles & Buchanan, 2014). Moreover, we have focused on an understudied migrant population and have shed some light on the experiences of Thai female marriage immigrants in the Netherlands. In social psychology, there is little interest in specific subgroups of migrants and the unique challenges they face, such as marriage migrants. A focus on various groups of immigrants might raise new social psychological questions and can contribute to a further development of social psychology as a societal relevant discipline. For example, marriage migration raises questions about the importance of different migration motives for the acculturation process of migrants and their developing sense of self. People can marry out of love, to improve their economic situation, or to legally stabilize their temporary status in society, and there is arranged marriage migration in which family assets, family honor, and religious and cultural practices and observances are important (Brettel, 2017;Timmerman, 2006). In general, social psychological research has paid little attention to the reasons that people have for migrating, although these reasons might in various and profound ways shape how migrants develop an understanding of themselves, their family, and the broader society in which they settle.