Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Oct 1.
Published in final edited form as: J Cross Cult Psychol. 2011 Sep 26;43(7):1082–1094. doi: 10.1177/0022022111420146

Ethnicity, gender socialization, and children’s attitudes towards gay men and lesbian women

Henny MW Bos 1,4, Charles Picavet 2, Theo G M Sandfort 3
PMCID: PMC3496397  NIHMSID: NIHMS306815  PMID: 23162164

Abstract

The aim of the present study was to assess whether children’s attitudes towards gay men and lesbian women differ in relation to their ethnic backgrounds, and whether ethnic differences are a result of perceived differential gender socialization practices. Data were collected from children in eight Dutch elementary schools by means of a paper-and-pencil questionnaire administered in the classroom. All children (mean age 11.47; N = 229) lived in the Netherlands; 50.2% had non-Western and 49.8% Western ethnic backgrounds. Children with non-Western ethnic backgrounds reported more negative attitudes towards gays and lesbians. These children perceived more parental pressure to behave in accordance with their gender and showed more negative attitudes towards gender-nonconforming behaviour by peers. Hierarchical regression analyses revealed that cultural differences in attitudes towards gay men and lesbian women are partly mediated by differentially perceived parental pressure to behave in accordance with their gender.

Keywords: Children, Ethnicity, Attitudes towards homosexuality, Gender role beliefs

Even though childhood is a critical formative period for the development of prejudice (Aboud, 2005; Fishbein, 1996), hardly any studies have explored children’s attitudes towards lesbian women and gay men. Factors such as ethnicity that affect such attitudes have also not been explored among children. The present study is one of the first studies to address this. Because attitudes towards lesbian women and gay men generally seem to be affected by attitudes about gender-nonconformity, we will explore the role of children’s evaluation of gender nonconformity as well as the pressure they perceive from peers and parents to behave in accordance with one’s gender.

Most studies exploring attitudes towards homosexual people have been carried out among adults. These studies have identified several correlates of acceptance of homosexuality, such as gender, socioeconomic status (SES), knowing and having contact with a homosexual person, ideas of what the proper roles are for men and women in society, and ethnicity (e.g., Cullen, Wright, & Alessandri, 2002; Herek, 2002; Lewis, 2003; Nierman, Thompson, Bryan, & Mahaffey, 2007; Ratcliff, Lassiter, Markman, & Snyder, 2009; Schulte & Balltle, 2004; Whitley, 2001). Individuals with more traditional ideas about gender-nonconformity are, for example, more likely to perceive gay men and lesbian women as violating cultural standards of masculinity and femininity, and to express negative attitudes towards gay men and lesbian women (e.g. Herek, 1993; Laner & Laner, 1980; Whitley, 2001).

Several US studies found that African-American adults have more negative attitudes towards same-sex sexual relationships than Caucasian-Americans (see for overview: Lewis, 2003). These differences are likely to result from differences in religious and educational background between African- Americans and Caucasian- Americans (Schulte & Balltle, 2004). In a study comparing Chileans and Americans, Nierman and colleagues (2007) found that Chileans were more prejudiced towards gay men and lesbian women, but when ideas of what proper roles are for men and women in society are taken into account, participants’ nationality (Chilean versus American) was no longer related to attitudes toward lesbian women and only moderately related to attitudes toward gay men. These findings indicate that gender role beliefs are helpful in understanding differences in acceptance of homosexuality between adults from various ethnic backgrounds.

Research has shown that at a young age children already have well-developed beliefs about how girls and boys should behave (Ruble & Martin, 1998). These beliefs affect, for instance, young children’s judgements of who should be included and excluded from gender stereotyped activities. During early adolescence (i.e. between the ages of 10 and 13 years), gender norms and conventions regarding behaviour, interests and appearance play an even stronger role (Alfieri, 1996). As a consequence, children in early adolescence start to judge cross-gender behaviour more harshly. Non-normative appearance, activities and preferences are often negatively sanctioned by peers (Carr, 1998; Lobel, Bempechat, Gewirtz, Shoken-Topaz, & Bashe, 1993; Lobel, 1994). Compared to girls, boys seem to have a stronger tendency to exclude other boys on the basis of cross-gender activities (Theimer, Killen, & Stangor, 2001).

Previous work has demonstrated that children with an ethnic minority background perceive more pressure to conform to traditional gender norms. Corby, Hodges and Perry (2007) found that Black and Hispanic children in the US perceived more pressure to conform to traditional gender norms than White children. To explain this difference, they argued that non-Western cultures are more collectivist rather than individualistic, and group conformity is stronger in collectivistic than in individualistic cultures (Triandis, 1995; Williams & Best, 1990). It is likely that a child’s ethnic minority status self-generates pressure to adhere to the norms of his or her culture (Arndt, Greenberg, Schimmel, Pyszcynski, & Solomon, 2002). Although Corby and colleagues (2007) did not distinguish between perceived parental versus perceived peer pressure regarding to gender conformity, it would be relevant to have a better understanding of the relative importance of parents and peers, and to separately assess the pressure that children perceive from both.

According to social learning theories (e.g., Bandura, 1977), parents play an important role in a child’s development of beliefs regarding gender stereotypes. Tenenbaum and Leaper (2002) showed in a meta-analysis how parents affect the gender-related thinking of their children. Children of parents who hold strong gender stereotypes and traditional attitudes towards women are, for example, more likely to hold beliefs that are similar to those of their parents (Lippa, 2002). Numerous studies have also shown that ethnicity is related to belief systems about how children should be raised (Bugental & Johnston, 2000; Huijbregts, Leseman, & Tavecchio, 2008; Kağıtçıbaşı, 1997; Rubin & Chunk, 2006). It is likely that children with non-Western ethnic backgrounds perceive more parental pressure to behave in accordance with their gender compared to those with Western ethnic backgrounds. This is because social norms regarding men and women in non-Western contexts differ more stringently, compared with those in a Western context.

Eleven percent of the population in the Netherlands, where the present study was carried out, are first- or second- generation immigrants from non-Western countries; the majority of them have a Turkish, Moroccan, Indonesian, Surinamese, or Antillean background (CBS, 2008). These groups with a non-Western background differ from adults with a traditional Dutch background in various ways, including what is seen as important regarding children’s personal and social development. For example, in non-Western groups collectivistic childrearing goals such as respect for authorities and conformity to rules are more valued (Huijbregts et al., 2008; Pels, Distelbrink & Postma, 2009). Most non-Western parents think that Dutch children are given too much independence and that their attitudes toward sexuality are too liberal. Some parents adhere to their own traditional values in parenting. Others try to combine the best of both worlds. Most non-Western parents do not adopt Dutch parenting styles because they are felt to be at odds with their own values and ideals (Pool, Geense & Lucassen, 2005). The distance between the world of men and women, and boys and girls in non-Western families is also larger than in Dutch families; Boys are treated differently than girls because women are believed to have other roles and obligations in life than men. Dutch parents are generally more egalitarian with regard to gender (Vennix & Vanwesenbeeck, 2005; Pels, Distelbrink & Postma, 2009). One could therefore expect that, even though these children are embedded within the same Dutch cultural context, there are differences in the gender socialization of children of non-Western ethnic groups and children with a Dutch ethnic background.

Although the role of parents seems apparent, there has been some disagreement about the importance of parents on children’s development (Harris, 1995; Maccoby, 1998). For example, Harris (1995) argued that children have their own culture which is separate from that of adults, and may often be more influenced by peers than by adults. Accordingly, children would be expected to adopt their peer’s ideas of what the proper roles for boys and girls are.

The aim of our study was to examine children’s attitudes towards gay men and lesbian women and explore the role of ethnic background by comparing children with a Dutch background with children with a non-Western ethnic backgrounds. We also assessed the role of children’s ethnicity on their opinions about gender-nonconforming behaviour, and on their perceived parental and peer pressure to behave in accordance with their gender. We further hypothesized that the expected ethnicity-related differences in attitudes toward gay men and lesbian women are mediated by the pressure children perceive to behave in accordance with their gender and by their evaluation of cross-gender behaviour.

Methods

The sample comprised 229 children: 112 boys (48.9%) and 117 girls (51.1%). All were between the ages of 10 and 13. The mean age was 11.47 (SD = .71); there was no significant difference in age between boys (M = 11.49, SD = .72) and girls (M = 11.44, SD = .70), F (1, 114) = .21, p > .05. Although the parents’ social economic backgrounds and educational levels were not assessed, we expect that given the schools’ locations, most children came from lower SES backgrounds.

One hundred and fifteen children (50.2%) reported that their mother and/or father were born in a non-Western country. From this group, 37 children reported that their mother was born in Turkey, 32 in Morocco, 22 in Surinam, 4 in the Netherlands Antilles, 3 in Indonesia, and 3 reported that their mother was not born in the Netherlands or another Western country but they did not specify the country. Regarding fathers, 38 children reported that their father was born in Turkey, 38 in Morocco, 17 in Surinam, 2 in the Netherlands Antilles, 5 in Indonesia, and 7 reported that their father was not born in either the Netherlands or another non-Western country but they did not specify which country. One hundred fourteen children (49.8%) reported that both their parents were born in the Netherlands; no children reported that their parents were born in other Western countries.

Procedure

Data were gathered in the two highest grades of eight elementary schools in Amsterdam, the Netherlands. All schools were located in low SES neighbourhoods. Consent was obtained in line with Dutch regulations. Before data collection started, the board of each school sent a letter to all parents containing information about the purpose, nature and date, of the study. In this letter parents were also informed about the topics that were included in the questionnaire. The letter made clear to the parents that their child’s participation in the study was voluntary and that if they wanted that their child should not be involved in this the study they could let this know to the research team by returning the letter and indicate that they did not wish their child to participate in the study. All parents allowed their child to participate. A research assistant (a student in Educational Sciences at the University of Amsterdam) administered the questionnaires in the classroom during school hours. The research assistant explained the topic of the questionnaire to the children and reminded them of the voluntary nature of their participation. All children were willing to participate.

Measurements

Negative attitudes towards gay men and lesbian women

Negative attitudes towards gay men and lesbian women were assessed by two scales developed by de Graaf, Meyer, Poelman and Vanwesenbeeck (2005), who also used these scales in a study among adolescents in the age of 12 years old. Children were asked to rate on a 5-point Likert-type response score (1= strongly disagree, 5= strongly agree) whether they agree with the following two statements: ‘It is disgusting when two boys/men kiss each other’, and ‘Boys/Men who are in love with other boys/men are sick’). To assess attitudes towards lesbian women the words ‘boys’ and ‘men’ in these items were replaced by ‘girls’ and ‘women’. Cronbach’s alphas in our study were .77 for attitudes towards gay men, and .73 for attitudes towards lesbian women.

Evaluation of gender-nonconforming behaviour

Children were asked to read the following hypothetical situation about exclusion of a peer based on cross-gender behaviour: ‘There is a group of five boys who are good friends. One of the boys X sometimes talks and acts like a girl. One day, the group decides to engage in a typical activity for boys. All the other boys in the group decide not to ask X to join in because he acts like a girl’ (an adjusted version was used for the scenario for girls). Using 5-point Likert scales, children were asked: (1) ‘How much do you agree or disagree with the decision not to ask X to join in?’ (1 = disagree completely, 5 = agree completely) (i.e., agreement with peer exclusion); (2) ‘How much should X change himself in order to be accepted by the group?’ (1 = not at all, 5 = very much) (i.e., prescription of conformity). The peer in the hypothetical situation was from the same sex as the student who filled in the questionnaire. This vignette and the questions assessed agreement with peer exclusion and prescription of conformity were developed by Killen, Crystal and Watanabe (2002). They were part of an instrument to assess children’s reasons for group exclusion of atypical peers; because of the focus of our study we only included the vignette about gender-nonconforming behaviour.

Perceived pressure from parents and peers to behave in accordance with one’s gender

The degree to which children felt pressure from parents and peers to conform to traditional gender norms was assessed with a modified version of two scales developed by Egan and Perry (2001). Each scale consisted of four items (e.g. ‘I think that my parents/peers would be upset if I told them that I want to play with girls’ toys’ (boys’ version)). In the original format, these items are formulated as bipolar statements and the respondent first has to decide the kind of child he or she is and then report if the description is ‘sort of true’ or ‘really true’ of him or her. However, because van den Bergh and Marcoen (1999) showed that this response format is too complex for younger children, we changed the answering format. Children were asked to rate on a 4-point scale (1 = not true at all, 4 = very true) ‘how true of you’ each statement was. Bos and Sandfort (2010) already used this version of these two scales in another study. Cronbach’s alphas in the present study were .80 and .79 for perceived parental and for peer pressure, respectively.

Ethnicity

In the Netherlands, the standard procedure for assessing a child’s ethnicity is to ask in which country the mother and father were born. A child is regarded as having a non-Western ethnic background if the mother and/or the father was not born in the Netherlands or in another Western country; if mother and father were both born in the Netherlands or another Western country, the child is considered as having a Western ethnic background. This procedure was applied in the present study.

Results

Intercorrelations between all studied variables by gender and by ethnic background (non-Western and Dutch ethnic background) are presented in Table 1, respectively.

Table 1.

Intercorrelations between negative attitudes towards gay men and lesbian women, agreement with peer exclusion based on gender- nonconformity, prescription of conformity, and perceived pressure from parents and peers to behave in accordance with one’s gender

Intercorrelations for boys (below diagonal) and girls (above diagonal)
Intercorrelations for children with a Western (below diagonal) and non-Western ethnic background (above diagonal)
1. 2. 3. 4. 5. 6. 1. 2. 3. 4. 5. 6.
1. Negative attitudes towards gay men - .96*** −.11 .08 .13 −.07 - .93*** .13 .17 .16 −.05
2. Negative attitudes towards lesbian women .93*** - −.11 .08 .14 −.08 .93*** - .07 .15 .08 −.12
3. Agreement with peer exclusion .30** .25** - .21* −.07 −.14 .12 .05 - .41*** .32*** .25**
4. Prescription of conformity .28** .25** .44*** - .33*** .13 .08 .00 .40*** - .46*** .34**
5. Parental pressure .49*** .45*** .28** .45*** - .06 .36*** .36*** .18 .35*** - .79***
6. Peer pressure .15 .12 .14 .28** .67*** - .32** .27 .12 .19* .63*** -
Open in a new tab
***

p < .001

**

p < .01

*

p < .05

Differences between children with non-Western and Dutch ethnic backgrounds

To test whether children with a non-Western ethnic background differ from their counterparts with a Dutch ethnic background with respect to attitudes towards gay men and lesbian women, opinions about gender-nonconforming behaviour, and perceived parental and peer pressure to behave in accordance with one’s gender we conducted a 2 (ethnicity: 1= non-Western, 2 = Dutch) by 2 (gender: 1 = boys, 2 = girls) multivariate analysis of variance (MANOVA). This analysis revealed a significant main effect for ethnicity (Wilks’s λ = .72 F (6, 219) = 14.28, p = .001), and gender (Wilks’s λ = .65, F (6, 219) = 19.97, p = .001). The interaction of ethnicity and gender was not significant, Wilks’s λ = .94, F (7, 219) = 2.15, p = .050. Table 2 shows the means and standard deviations separately for children with a non-Western and Dutch ethnic background and for boys and girls. In this table we also included the statistical information regarding the ANOVAs for gender differences and differences on ethnicity.

Table 2.

Means (and standard deviations) for negative attitudes towards gay men and lesbian women, evaluation of gender- nonconforming behaviour, and pressure to conform to traditional gender norms (from parents and peers)

Children’s ethnic background
Non-Western
Western
Gender
F - value
Boys Girls Total Boys Girls Total Boys Girls Gender Ethnicity
Negative attitudes towards gay men 3.61 (1.37) 3.12 (1.62) 3.37 (1.51) 2.23 (0.97) 2.00 (0.99) 2.11 (0.98) 2.95 (1.38) 2.55 (1.44) 4.77* 56.67***
Negative attitudes towards lesbian women 3.42 (1.35) 3.25 (1.52) 3.34 (1.43) 2.16 (0.89) 2.03 (1.03) 2.09 (0.97) 2.82 (1.31) 2.63 (1.43) 0.92 58.92***
Agreement with peer exclusion 2.74 (1.02) 1.67 (0.81) 2.21 (1.06) 2.25 (1.05) 1.76 (0.80) 1.99 (0.95) 2.50 (1.06) 1.72 (0.80) 40.38*** 2.63
Prescription of conformity 3.10 (1.42) 2.12 (1.00) 2.62 (1.32) 2.19 (1.09) 2.05 (0.92) 2.11 (1.00) 2.67 (1.35) 2.09 (0.96) 14.13*** 11.00***
Parental pressure 3.47 (1.11) 2.39 (1.03) 2.93 (1.119) 2.54 (1.00) 1.82 (0.74) 2.16 (0.94) 3.03 (1.15) 2.10 (0.94) 48.43*** 33.72***
Peer pressure 3.75 (1.02) 2.55 (1.11) 3.16 (1.22) 3.50 (0.82) 2.53 (0.81) 2.99 (0.95) 3.63 (0.94) 2.54 (0.97) 74.19*** 1.25
Open in a new tab
***

p < .001

**

p < .01

*

p < .05

Attitudes towards gay men and lesbian women

Follow-up 2 (ethnicity: 1= non-Western, 2 = Dutch) by 2 (gender: 1 = boys, 2 = girls) ANOVAs showed a significant main effect for ethnicity on attitudes towards gay men and lesbian women: compared to children with a Dutch background, children with non-Western backgrounds had more negative attitudes towards gay men, F (1, 219) = 56.67. p = .001, and lesbian women, F (1, 219) = 59.92, p = .001. Boys were also more negative about gay men than girls, F (1, 219) = 4.77, p = .030. Boys and girls did not differ on their attitudes towards lesbian women, F (1, 219) = .92, p = .339.

Children’s evaluation of gender-nonconforming behaviour

Children with non-Western and Dutch backgrounds differed significantly in whether they thought that the excluded child should change him- or herself to fit into the group (prescription of conformity): children with non-Western backgrounds reported more often that a child with gender nonconforming behavior should change this behavior in order to be accepted by the peer group, F (1, 219) = 11.00, p = .001. Children with non-Western and Western backgrounds did not differ on agreement with peer exclusion based on gender-nonconforming behavior, F (1, 219) = 2.636, p = .107.

Furthermore, compared to girls, boys reported higher scores on agreement with peer exclusion, F (1, 219) = 40.38, p = .001, and also they scored higher on prescription of conformity, F (1, 219) = 14.13, p = .001.

Perceived pressure to behave in accordance with one’s gender

Significant main effects of ethnicity, F (1, 219) = 33.72, p = .001, and gender, F (1, 219) = 48.43, p = .001, were also found on perceived parental pressure to behave in accordance with one’s gender: children with non-Western backgrounds perceived significantly more parental pressure to behave in accordance with their gender than those with Dutch backgrounds, and boys perceived more pressure from their parents than girls.

For perceived peer pressure to behave in accordance with one’s gender we only found a significant main effect of gender: boys perceived more peer pressure than girls to behave in accordance with their gender, F (1, 219) = 74.19, p = .001. There was no significant main effect of ethnicity, F (1, 219) = 1.25, p = .266, on perceived peer pressure.

Mediation model

To test whether the effect of ethnicity on attitudes towards gay men and lesbian women was mediated by children’s evaluation of gender non-conforming behaviour, and the perceived pressure of parents and peers to conform to traditional gender norms, we followed the guidelines for mediation analysis suggested by Baron and Kenny (1986). We tested this mediation model for the boys and girls collectively, because we had found no interactions between ethnicity and gender on any of the studied variables.

According to Baron and Kenny (1986) several conditions must be met in order to demonstrate the existence of a mediating pathway. There must be significant relationships between: (1) the independent variable (ethnicity) and the dependent variables (negative attitudes towards gay men and negative attitudes towards lesbian women); (2) the independent variable and the potential mediators (evaluation of gender-nonconforming behaviour, and perceived parental and peer pressure to conform to traditional gender norms); and (3) the potential mediators and the dependent variables. The final requirement for mediation is that the effect of ethnicity on attitudes towards gay men and lesbian women disappears or is significantly reduced when entered in conjunction with the mediators as opposed to when it is entered alone.

We had already established that the first condition is met: ethnicity is related to attitudes towards gay men and attitudes towards lesbian women. For the second condition, we demonstrated that ethnicity was related to prescription of conformity and perceived parental pressure. To examine the third condition, we computed a multiple regression analyses with negative attitudes towards gay men and lesbian women as dependent variables and prescription of conformity and perceived parental pressure (the only two potential mediators that met the first and second conditions) as independent variables. Perceived parental pressure was found to make a significant and independent contribution to the variance explained in negative attitudes towards gay men (β = .31, p = .001; R2 = .12, p = .001) and lesbian women (β = .28, p = .001; R2 = .12, p = .001). Children with negative attitudes towards gay men and lesbian women felt stronger parental pressure to behave in accordance with their gender. In these analyses prescription of conformity was not significantly related to negative attitudes towards gay men (β = .07, p = .306) or lesbian women (β = .05, p = .451).

To test our mediation model, we conducted two hierarchical regression analyses with attitudes towards gay men and attitudes towards lesbian women as dependent variables. Ethnicity was entered in Step 1, and perceived parental pressure was entered in Step 2 (because this was the only possible mediator that met the criteria of significant association). Summary statistics for these regression models are presented in Table 3.

Table 3.

Summary of hierarchical regression analyses on negative attitudes towards gay men and towards lesbian women

Negative attitudes towards gay men
Negative attitudes towards lesbian women
B Std. Error β B Std. Error β
Step 1
Ethnic background 1 −1.27 .17 −.45*** −1.26 .16 −.46***
Step 2
Ethnic background 1 −1.06 .17 −.37*** −1.10 .17 −.40***
Pressure from parents .27 .08 .22*** .20 .08 .16**
Open in a new tab
1

1 = non-Western, 2 = Western. Note: for negative attitudes towards gay men: R2 = .20, F (1, 227) = 56.81*** for step 1, ΔR2 = .04, Δ F (2, 227) = 12.22 *** for step 2. For attitudes towards lesbian women: R2 = .21, F (1, 227) = 59.81*** for step 1, ΔR2 = .02, Δ F (2, 227) = 6.85 ** for step 2.

***

p < .001

**

p < .01

*

p < .05

The inclusion of perceived parental pressure in Step 2 (β = .22, p = .001) produced a significant change in the coefficient of determination for negative attitudes towards gay men, ΔR2 = .04, p = .001. The β for ethnicity in the prediction of negative attitudes towards gay men decreased from −.45 (p = .001) to −.37 (p = .001). Ethnicity and perceived parental pressure together accounted for 24% of the variance in attitudes towards gay men.

For negative attitudes towards lesbian women, perceived parental pressure in Step 2 (β = .16, p = .009) also produced a significant change in the coefficient of determination, ΔR2 = .02, p = .009. The β for ethnicity in the prediction of negative attitudes towards lesbian women decreased from −.46 (p = .001) to −.40 (p =.001). For negative attitudes towards lesbian women, ethnicity and perceived parental pressure together accounted for 23% of the variance.

The Sobel tests (MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002) indicated that perceived parental pressure to behave in accordance with one’s gender is a significant mediator in the relationship between ethnicity and attitudes towards gay men (Sobel test z = −3.88, SE = .09, p = .001) and in the relationship between ethnicity and attitudes towards lesbian women (Sobel z = −2.28, SE = .08, p =.01).

Discussion

In our study we found that children’s ethnicity is strongly associated with attitudes towards gay men and lesbian women: compared to children with a Dutch background, children with non-Western backgrounds had less positive attitudes. Children with non-Western backgrounds also felt more strongly that children with gender-nonconforming behaviour should adjust in order to be accepted by their peers, and they felt more pressure from parents to conform to traditional gender norms. Perceived parental pressure to conform to one’s gender was found to account for part of the differences in children’s ethnicity on attitudes towards gay men and lesbian women.

In contrast to perceived parental pressure, no main effect of ethnicity was found on peer pressure to conform to gender norms. This finding emphasizes that regarding the pressure that children feel to gender conformity from their environment it is important to assess this separately for the pressure from peers and parents. Furthermore, our finding revealed that parents ethnic background does matter regarding to pressure to behave conform once gender. Ethnicity, however, does not play a role on this type of pressure felt from peers. It should be noted, however, that we did not differentiate between perceived pressure from peers with non-Western backgrounds and perceived pressure from peers with Western backgrounds.

The finding that children with non-Western ethnic backgrounds feel more parental pressure to conform to traditional gender norms is likely to be a reflection of the parents’ more traditional gender role beliefs resulting from more stringent differences in non-Western cultures between the social norms for men and women (Hofstede, 1998). Some prudence is required regarding this explanation, however, as the parents’ attitudes towards gender role beliefs were not included in this study. The higher score on perceived parental pressure to conform to traditional gender norms might also reflect the relatively more collectivistic character of the minority cultures involved in this study in which group conformity is presumably stronger (Fiske, Kitayama, Markus, & Nisbett, 1998; Matsumoto & Hee Yoo, 2006; Nierman et al., 2007). Furthermore, the greater perceived parental pressure to conform to traditional gender norms reported by children with non-Western backgrounds might reflect the fact that a person’s minority status tends to self-generate pressure to adhere to the norms of his or her culture (Arndt et al., 2002).

We found that perceived parental pressure to conform to traditional gender norms mediated the effect of ethnicity on attitudes towards gay men and lesbian women. It should be noted, however, that the mediation effect was partial. This suggests that additional factors have to be identified to explain the difference between non-Western and Dutch children’s negative attitudes towards gay men and lesbian women. These factors may be related to religious beliefs: the majority of the non-Western children in this study had a Turkish or a Moroccan ethnic background, which is strongly influenced by Islamic values, in which homosexuality is unacceptable and labelled a sin (Oksal, 2008). Subsequent studies should also include religious backgrounds as a potential factor explaining the observed differences.

Furthermore, the findings in our study that children with non-Western ethnicities perceive more parental pressure to conform to traditional gender norms and hold a more negative attitudes towards gay men and lesbian women might be related to differences in SES between families with non-Western and those with Western backgrounds. Studies on gender role beliefs and attitudes towards gay men and lesbian women show that these attitudes are influenced by SES (Lewis, 2003). Although we did not collect data about the SES of the parents, all children came from schools in low SES neighbourhoods, and thus most of them are likely to come from lower SES backgrounds. As a consequence, it is unlikely that differences in SES family background explain the differences we found in this study. Although not the focus of our study, it should be mentioned that significant differences were found between boys and girls in attitudes towards gay men. Compared to girls, boys were more negative about gay men. This gender difference is consistent with what is found among adults (e.g., Herek, 2002a, 2002b). In the Netherlands were the current study was carried out this gender difference is also found among (young) adults (de Graaf, Meijer, Poelman, & Vanwesenbeeck, 2005; Keuzenkamp, 2010; Keuzenkamp et al. 2006; Kuyper & Bakker, 2006).

We also found gender differences in perceived parental and peer pressure to conform to one’s gender, and their opinions about gender-nonconforming behaviour: boys perceived more parental and peer pressure than girls to behave in accordance with their gender. Girls were less negative about the gender-nonconforming behaviour of peers and more often disagreed with the exclusion of these peers. Previous work has shown that girls are more sensitive than boys to issues of exclusion. They are, for example, more concerned with fairness than boys are (Killen & Stangor, 2001). Furthermore, girls more frequently experience exclusion than boys (Theimer et al., 2001). Therefore, girls may identify with excluded peers and as a consequence reject the exclusion of peers based on cross-gender behaviour or other reasons. Our findings of gender differences in level of agreement with peer exclusion based on gender-nonconforming behaviour and prescription of conformity could be related to the fact that in the present study boys evaluated the gender-nonconformity of boys while girls evaluated the gender-nonconformity of girls. This would be in line with the overall tendency that gay men are evaluated more negatively than lesbian women (de Graaf, Meijer, Poelman, & Vanwesenbeeck, 2005; Herek, 2002a, 2002b. Keuzenkamp, 2010; Keuzenkamp et al. 2006; Kuyper & Bakker, 2006).

Interactions between ethnicity and gender were not significant on all studied variables. This indicates that the differences between children with non-Western and Western backgrounds in attitudes towards gay men and lesbian women, perceived parental pressure to behave in accordance with one’s gender, and prescription of conformity are similar for boys and girls.

Several limitations of this study should be noted. First, we combined children with a Turkish, Moroccan, Surinamese, Antillean and Indonesian background into one group (i.e., children with non-Western backgrounds) and compared them with children with a Dutch native background. Given the relatively low numbers of any single non-Western group, it was not possible to test differences between these groups. However, this procedure might mask the considerable diversity within the group of children with non-Western backgrounds. A second limitation is the way we categorized children in terms of their ethnicity. We included children with one parent who was born outside the Netherland in the non-Dutch group (the groups with one versus two parents born outside the Netherlands were too small to analyze separately) and we did not take into account how long the parents already lived in the Netherlands (this information was not collected). We expect this limitation to attenuate the actual differences based on ethnicity. A third limitation is that our findings are based on children’s self-reports: it would be informative to expand this assessment by including the perceptions of, for example, teachers and parents. A fourth limitation concerns the instrument that was used to assess the children’s evaluation of gender nonconforming behaviour: Children were asked if it is right for children who are planning a same-gender activity to exclude a peer with known cross-gender interests. It could be that children’s judgments about this exclusion might be based on the assumption that the children in the vignette who are excluding the gender-nonconforming child are doing this based on their idea that the cross-gender-typed child might not enjoy the planned activity. In this case the exclusion can be seen as considerate behavior. Furthermore, it is a limitation that children’s agreement with peer exclusion was measured only with a single item. This was also the case regarding children’s prescription of conformity with the excluded peer. Finally, the study was not longitudinal. A longitudinal design would be especially interesting as it would allow us to understand how gender role beliefs and attitudes towards gay men and lesbian women develop over time, and how this development is influenced by ethnicity.

Our findings indicate that children in early adolescence who are living in the Netherlands but have a non-Western background hold more negative attitudes towards gay men and lesbian women. Furthermore, we found support for the hypothesis that children’s perceived parental pressure to conform to traditional gender norms at least partially accounts for differences in attitudes towards gay men and lesbian women between children with non-Western and those with a Dutch ethnic backgrounds. These findings strongly suggest that in order to improve children’s attitudes towards homosexuality, it is crucial to address their gender role beliefs as well, including those of their parents.

References

  1. Aboud FE. The development of prejudice in childhood and adolescence. Malden, MA: Blackwell Publishing; 2005. [Google Scholar]
  2. Alfieri T. Gender stereotypes during adolescence: Developmental changes and the transition to junior high school. Developmental Psychology. 1996;32:1129–1137. [Google Scholar]
  3. Arndt J, Greenberg J, Schimel J, Pyszczynski T, Solomon S. To belong or not to belong, that is the question: Terror management and identification with gender and ethnicity. Journal of Personality and Social Psychology. 2002;83:26–43. [PubMed] [Google Scholar]
  4. Bandura A. Social learning theory. London: Prentice-Hall; 1977. [Google Scholar]
  5. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology. 1986;51:1173–1182. doi: 10.1037//0022-3514.51.6.1173. [DOI] [PubMed] [Google Scholar]
  6. Bos H, Sandfort Th. Children’s gender identity in lesbian and heterosexual two-parent families. Sex Roles. 2010;62:114–126. doi: 10.1007/s11199-009-9704-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bugental DB, Johnston C. Parental and child cognitions in the conext of the family. Annual Review of Psychology. 2000;51:315–344. doi: 10.1146/annurev.psych.51.1.315. [DOI] [PubMed] [Google Scholar]
  8. Carr CL. Tomboy resistance and conformity: Agency in social psychological gender theory. Gender and Society. 1998;12:528–553. [Google Scholar]
  9. Centraal Bureau voor de Statistiek (CBS) [National Statistics Office, the Netherlands] Bevolking naar herkomstgroepering en generatie [Dutch Population: Origins and generations] 2011 Retrieved March 14, 2011, from http://www.statline.cbs.nl/
  10. Corby BC, Hodges EV, Perry DG. Gender identity and adjustment in Black, Hispanic, and White preadolescents. Developmental Psychology. 2007;43:261–266. doi: 10.1037/0012-1649.43.1.261. [DOI] [PubMed] [Google Scholar]
  11. Cullen Jenifer M, Wright, Lester W, Alessandri M. The personality variable openness to experience as it relates to homophobia. Journal of Homosexuality. 2002;42:119–134. doi: 10.1300/J082v42n04_08. [DOI] [PubMed] [Google Scholar]
  12. De Graaf H, Meijer S, Poelman J, Vanwesenbeeck I. Sexual health of adolescents in the Netherlands in 2005. Utrecht: Rutgers Nisso Groep; 2005. Seks onder je 25e Seksuele gezondheid van jongeren in Nederland anno 2005. [Google Scholar]
  13. Egan SK, Perry DG. Gender identity: A multidimensional analysis with implications for psychological adjustment. Developmental Psychology. 2001;37:451–463. doi: 10.1037//0012-1649.37.4.451. [DOI] [PubMed] [Google Scholar]
  14. Fishbein HD. Peer prejudice and discrimination; Evolutionary, culture, and developmental dynamics. Boulder, CO: Westview Press; 1996. [Google Scholar]
  15. Fiske AP, Kitayama s, Markus HR, Nisbett RE. The cultural matrix of social psychology. New York: McGraw-Hill; 1998. [Google Scholar]
  16. Harris J. Where is the child’s environment? A group socialization theory of development. Psychological Review. 1995;102:458–489. [Google Scholar]
  17. Herek GM. The context of antigay violence: Notes on cultural and psychological heterosexism. New York: Columbia University Press; 1993. [Google Scholar]
  18. Herek GM. Gender gaps in public opinion about lesbians and gay men. Public Opinion Quarterly. 2002;66:40–66. [Google Scholar]
  19. Hofstede G. Masculinity/femininity as a dimension of culture. In: Hofstede G, editor. Masculinity and femininity: The taboo dimension of national cultures. Thousand Oaks, CA: Sage; 1998. pp. 3–38. [Google Scholar]
  20. Huijbregts SK, Leseman PPM, Tavecchio LWC. Cultural diversity in center-based childcare: Childrearing beliefs of professional caregivers from different cultural communities in the Netherlands. Early Childhood Research Quaterly. 2008;23:233–244. [Google Scholar]
  21. Kağıtçıbaşı Ç. Individualism and collectivism. In: Berry JW, editor. Handbook of cross-cultural psychology. Vol. 3: Social behavior and applications. Boston, MA: Allyn and Bacon; 1997. pp. 1–49. [Google Scholar]
  22. Keuzenkamp S. Monitoring acceptance of homosexuality in the Netherlands. The Hague: The Netherlands Institute for Social Research; 2010. [Google Scholar]
  23. Killen M, Stangor C. Children’s reasoning about social inclusion and exclusion in gender and race peer group contexts. Child Development. 2001;72:174–186. doi: 10.1111/1467-8624.00272. [DOI] [PubMed] [Google Scholar]
  24. Killen M, Crystal D, Watanabe H. The individual and the group: Japanese and American children’s evaluation of peer exclusion, tolerance of difference, and prescription for conformity. Child Development. 2002;73:1788–1802. doi: 10.1111/1467-8624.t01-1-00506. [DOI] [PubMed] [Google Scholar]
  25. Kuyper L, Bakker F. Attitudes towards homosexuality. A literature search. The Hague: The Netherlands Institute for Social Research; 2006. De houding ten opzichte van homoseksualiteit. Een beschrijvende literatuurstudie. [Google Scholar]
  26. Laner MR, Laner RH. Sexual preference or personal style? Why lesbians are disliked. Journal of Homosexuality. 1980;5:339–356. doi: 10.1300/J082v05n04_01. [DOI] [PubMed] [Google Scholar]
  27. Lewis GB. Black-White differences in attitudes toward homosexuality and gay rights. Public Opinion Quarterly. 2003;67:59–78. [Google Scholar]
  28. Lippa RA. Gender, nature and nurture. Mahwah, NJ: Lawrence Erlbaum Associates, Inc; 2002. [Google Scholar]
  29. Lobel TE, Bempechat J, Gewirtz JC, Shoken-Topaz T, Bashe E. The role of gender-related information and self-endorsement of traits in preadolescents’ inferences of judgments. Child Development. 1993;64:1285–1294. [PubMed] [Google Scholar]
  30. Lobel TE. Sex typing and the social perception of gender stereotypic and nonstereotypic behaviour; The uniqueness of feminine males. Journal of Personality and Social Psychology. 1994;66:379–385. doi: 10.1037//0022-3514.66.2.379. [DOI] [PubMed] [Google Scholar]
  31. Maccoby EE. The two sexes; Growing up apart, coming together. Cambridge, MA: Belknap Press; 1998. [Google Scholar]
  32. MacKinnon CM, Lockwood JM, Hoffman SG, West SG, Sheets V. A comparison of methods to test mediation and other intervening variable effects. Psychological Methods. 2002;7:83–110. doi: 10.1037/1082-989x.7.1.83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Martin CL, Ruble DN, Szykrybalo J. Recognizing the centrality of gender identity and stereotype knowledge in gender development and moving toward theoretical integration: reply to Bandura and Bussey (2004) Psychological Bulletin. 2004;130:702–710. doi: 10.1037/0033-2909.130.5.702. [DOI] [PubMed] [Google Scholar]
  34. Matsumoto D, Hee Yoo S. Toward a new generation of cross-cultural research. Perspective on Psychological Science. 2006;1:234–250. doi: 10.1111/j.1745-6916.2006.00014.x. [DOI] [PubMed] [Google Scholar]
  35. Nierman AJ, Thompson SC, Bryan A, Mahaffey A. Gender role beliefs and attitudes towards lesbian and gay men in Chile and the US. Sex Roles. 2007;57:61–67. [Google Scholar]
  36. Oksal A. Turkish family members’ attitudes towards lesbians and gay men. Sex Roles. 2008;58:514–525. [Google Scholar]
  37. Pels T, Distelbrink M, Postma L. Childrearing in a context of migration: Reserach review on childrearing of migrant families in the Netherlands. Utrecht: Verwey-Jonker Instituut; 2009. Opvoeding in de migratiecontext: Review van onderzoek naar de opvoeding in gezinnen van nieuwe Nederlanders. [Google Scholar]
  38. Pool M, Geense P, Lucassen N. Migrant families: Parents and their teenagers. Den Haag: Nederlandse Gezinsraad; 2005. Allochtone gezinnen: Opvoeding van tieners. [Google Scholar]
  39. Ratcliff JJ, Lassiter GD, Markman Kd, Snyder CJ. Gender differences in attitudes toward gay men and lesbians: The role of motivation to respond without prejudice. Personality and Social Psychology, Bulletin. 2009;22:1325–1337. doi: 10.1177/0146167206290213. [DOI] [PubMed] [Google Scholar]
  40. Rubin KH, Chunk O, editors. Parenting beliefs, behaviors, and parent-child relations: A cross-cultural perspective. NY: Psychology Press; 2006. [Google Scholar]
  41. Ruble DN, Martin CL. Gender development. Hoboken: John Wiley and Sons Inc; 1998. [Google Scholar]
  42. Schulte L, Battle J. The relative importance of ethnicity and religion in predicting attitudes towards gays and lesbians. Journal of Homosexuality. 2004;47:127–142. doi: 10.1300/J082v47n02_08. [DOI] [PubMed] [Google Scholar]
  43. Tenenbaum HR, Leaper C. Are parents’ gender schemas related to their children’s gender-related cognitions? A meta- analysis. Developmental Psychology. 2002;38:615–630. doi: 10.1037//0012-1649.38.4.615. [DOI] [PubMed] [Google Scholar]
  44. Theimer CE, Killen M, Stangor C. Young children’s evaluations of exclusion in gender-stereotypic peer contexts. Developmental Psychology. 2001;37:18–27. [PubMed] [Google Scholar]
  45. Triandis HC. Individualism and collectivism. Boulder, CO: Westview; 1995. [Google Scholar]
  46. Van den Bergh BRH, Marcoen A. Harter’s self-perception profile for children: Factor structure, reliability, and convergent validity in a Dutch-speaking Belgian sample of fourth, fifth, and sixth grades. Psychologica Belgica. 1999;39:29–47. [Google Scholar]
  47. Vennix P, Vanwesenbeeck I. Nederlanders. Delft: Eburon; 2005. Seksualiteit en relaties van Turkse en Marokkaanse. [Google Scholar]
  48. Williams JE, Best DL. Gender and self viewed cross-culturally. Newbury Park, CA: Sage; 1990. [Google Scholar]
  49. Whitley BE. Gender-role variables and attitudes towards homosexuality. Sex Roles. 2001;45:691–612. [Google Scholar]

RESOURCES