Keywords

Development of Gender Prejudice from Childhood to Adulthood: A Spanish Perspective

The development of intergroup prejudice is a subject of great interest in social psychology (e.g., Aboud, 2005, 2008; Bigler & Liben, 2007; Enesco & Guerrero, 2011). In fact, there is broad consensus that implementing preventive interventions with children is one of the best ways to prevent prejudice when such children become adults (Stathi, Cameron, Hartley, & Bradford, 2014). In the specific case of gender prejudice, many efforts have recently been made in countries such as Spain to implement measures aimed at reducing gender-based discrimination. In Spain, two pieces of legislation have guided such efforts: (1) the Act on Comprehensive Protective Measures Against Gender-Based Violence (Ley Orgánica 1/2004 de Medidas de Protección Integral contra la Violencia de Género), a pioneering act, and (2) the Act on Effective Equality Between Women and Men (Ley Orgánica 3/2007 para la igualdad efectiva de mujeres y hombres). In education, for example, a special gender equality program promoted by the regional government of Andalusia (in southern Spain) has been active since 2006 (Consejería de Educación de la Junta de Andalucía, 2006, 2007). Although some changes derived from this legal framework have corrected the power imbalance between men and women to some extent, discrimination against women still exists in the patriarchal Spanish society. This is shown by the persistence of gender-based violence (WHO, 2002), the pay gap (Sanz de Miguel, 2010), and sexual discrimination at the workplace (Martín Artiles, 2006). According to social psychologists, sexist ideology has a pervasive role in perpetuating these inequalities (Glick & Fiske, 2001; Pratto & Walker, 2004). We argue that a better knowledge of the factors that influence the development of sexist attitudes will facilitate the design of specific interventions aimed at reducing such attitudes from early stages of human development.

Internal and External Factors that Explain the Development of Prejudice and Stereotypes During Childhood

The interest in the study of intergroup prejudice has promoted multidisciplinary research in the fields of social and developmental psychology (Enesco & Guerrero, 2011). The growing interest of both disciplines in exploring intergroup prejudice has led to an increase of theoretical and empirical studies on this subject in recent years (e.g., see the monograph devoted to the study of intergroup prejudice in the Spanish journal Anales de Psicología 2011, as well as international journals such as the European Journal of Social Psychology 2010, and the International Journal of Behavioral Development 2007). These and other previous studies have led to the two main theoretical approaches that explain the emergence of prejudice during childhood (see Martin & Ruble, 2010).

The first approach focuses on analyzing the cognitive processes of individuals that lead to the development of prejudiced beliefs (e.g., Aboud, 2005; Martin & Ruble, 2004). These models mainly focus on the influence of internal variables following classic cognitive theories such as genetic-developmental theories (Piaget, 1966; Kohlberg, 1966) and theories based on the concept of schema (Bem, 1981; Markus & Oyserman, 1989). They also highlight the influence of motivational factors associated with individuals’ need of identifying with a group and having a positive status (Tajfel, 1981). The need for identification is one of the basic motivations during childhood. For instance, Kohlberg (1966) states that gender identity is achieved by the age of 2 years. According to cognitive theories based on the concept of schema (Bem, 1981; Markus & Oyserman, 1989), once individuals self-categorize as members of a group, they process and interpret information based on such membership (Martin & Halverson, 1983). Cognitive models of the development of prejudice also give great importance to the levels of cognitive and emotional development of individuals at different ages, based on the Piagetian model (see Aboud, 2005). Studies on this issue have shown that individuals first develop the ability to categorize (e.g., colors, food); this is followed by the concept of the self (i.e., personal identity), social identity (i.e., group membership), and finally empathy and the ability to take perspective (Berger, 2012). Through this cognitive progression, boys and girls learn the concepts, traits, and behaviors that are socially accepted for the various social groups (e.g., men and women). In short, according to this approach, social attitudes are mediated by self-identification processes (Nesdale, Kiesner, Durkin, Griffiths, & Ekberg, 2007) and the cognitive development of individuals (Aboud, 2008).

The second approach is that of theories based on social learning. According to them, the development of prejudice and stereotypes is strongly influenced by ­variables that are external to individuals (e.g., Bussey & Bandura, 1999). The social learning theory posits that children learn discriminatory behaviors by modeling, that is, by observing other people who are important to them and imitating their behavior (Bandura, 1986). Models are more or less imitated depending on how attractive (i.e., pleasant, powerful, or involving higher social status) or familiar they are. From this approach, parental figures are considered particularly relevant as socializing agents of their children (e.g., Bussey & Bandura, 1999; Leaper, 2002; Leaper and Friedman, 2006; McHale, Crouter, & Whiteman, 2003). Although the influence of parents is important, there is not always a direct relationship between their prejudice and that of their children (Enesco & Guerrero, 2011). Other contextual factors or socialization figures (e.g., school, peer groups, the media) may have a strong impact on the development of prejudice in children. According to the social learning theory, learning can take place even if the child does not imitate the behavior immediately. The child also learns the characteristics of the situation, recognizes the victim of discrimination, and understands the consequences of the behavior for the model. Although the consequences for the model (i.e., punishment or reward) are particularly important, only rewards seem to lead to the long-term maintenance of behavior (Bussey and Bandura, 1999).

Despite the differences between the focus of the cognitive and social learning approaches, they both acknowledge the importance of both internal and external variables (cf. Aboud, 2008; Bussey & Bandura, 1999). Today, any psychosocial analysis of prejudice requires explanatory models in which both types of variables are considered. Bigler and Liben (2007) have recently proposed an inclusive theory that considers both internal (i.e., cognitive and motivational) and external (i.e., social) variables to explain the development of intergroup prejudice and stereotypes. Developmental intergroup theory (DIT; Bigler and Liben) proposes a model based on three processes: determining the relevance of certain personal characteristics, categorizing people based on such relevant characteristics, and developing stereotypes and prejudices about such social groups. These three processes involve the cognitive skills of individuals and are also influenced by the messages conveyed by society and the situations observed by children in their environment. In short, current theories suggest that both environmental factors and internal or cognitive factors are important to understand the process through which stereotypes and prejudices develop during childhood (Aboud, 2008; Bigler & Liben, 2007; Bussey & Bandura, 1999).

Gender Prejudice

Gender prejudice , also known as sexist ideology, is one of the main pillars that support the patriarchal system and inequalities between men and women (Glick & Fiske, 2001; Pratto & Walker, 2004). Accordingly, the study of sexism as a legitimizing ideology of gender inequalities has led to important theoretical developments and empirical evidence. Over the last two decades, many studies have been conducted on “new forms” of sexism (e.g., Glick & Fiske, 1996; Swim, Aikin, Hall, & Hunter, 1995; Tougas, Brown, Beaton, & Joly, 1995). Such forms are new in that they were not explored by classic literature on prejudice and have only recently started to receive interest. Furthermore, “new” does not mean that they evolved from more traditional forms. Traditionally, sexism has been conceptualized as hostility of men toward women (Cameron, 1977). However, this concept does not explain the “positive” characteristics of the feminine stereotype (Eagly & Mladinic, 1993) or the negative effects of behaviors that are apparently benevolent and flattering to women (Jackman, 1994; Major & Vick, 2005; Moya, Glick, Expósito, de Lemus, & Hart, 2007; Vescio, Gervais, Snyder, & Hoover, 2005) .

One of the theoretical proposals that has had the strongest influence on the analysis of these “new forms” of sexism is ambivalent sexism theory (AST), proposed by Glick and Fiske (1996, 2001). According to these authors, two expressions of gender prejudice coexist: hostile sexism (HS) and benevolent sexism (BS). HS refers to the classic concept of prejudice. It is defined as combative hostile ideology toward women based on the belief that men should have more power than women (i.e., dominative paternalism) given that the characteristics of men are more valuable than those of women (i.e., competitive gender differentiation). It is also based on the idea that women are dangerous because of their sexuality and men should control women’s attempts to usurp their power using their sexuality, among other strategies (i.e., heterosexual hostility) . By contrast, BS is a more subtle sexist ideology based on protective paternalism. It is based on the belief that men should protect women because they are sweet and fragile and therefore depend on them. It highlights positive stereotypical characteristics of women as wonderful, sensitive, and kind creatures (e.g., Eagly & Mladinic, 1993). Yet, it assumes that men and women have different characteristics (i.e., complementary gender differentiation) and that the characteristics of women are more appropriate for certain roles (e.g., domestic and care-related activities) that exclude them from the public sphere. Finally, BS highlights that men need women as intimate partners, as heterosexual intimate relations are essential to achieve true happiness (i.e., heterosexual intimacy).

Both HS and BS function as legitimizing ideologies that complement each other to justify and maintain gender inequality in many different countries and cultures, including Spain (Glick et al., 2000, 2004). HS is targeted as a punishment for women who transgress established gender roles, while BS acts as a reward (e.g., protection, idealization) for women who behave according to traditional gender roles .

Similarly to research conducted in other countries, studies in Spain have revealed that scores obtained in HS and BS scales are correlated with other measures of gender ideologyFootnote 1. Specifically, both HS and BS have been found to be correlated with traditional sexism measured with the Escala de Ideología de Género (Gender Ideology Scale; Moya, Expósito, & Padilla, 2006; Expósito, Moya, & Glick, 1998; \(r =.46,p<.01\) for HS and \(r =.28,p<.01\) for BS), neosexism—a new form of negative attitudes toward women focused on the work environment—(\(r =.39,p<.01\) for HS and \(r =.17,p<.01\) for BS—only men; Expósito et al., 1998; Moya & Expósito, 2001), and rape myths (\(r = -.45,p<.01\) for HS and \(r = -.29,p<.01\) for BS; Frese, Moya, & Megías, 2000). Such correlations were always higher with HS than with BS, although it is interesting to note the existence of correlations with BS. Importantly, ambivalent sexism (AS) has not been found to be correlated with measures of social desirability (Moya & Expósito, 2008) .

The concept of AS assumes that HS and BS must be positively correlated. That is, individuals who endorse HS will also tend to endorse BS and vice versa. This idea has been confirmed by the studies conducted (e.g., Glick & Fiske, 1996; Glick et al., 2000). Research carried out by Glick et al. (2000) with participants from 19 countries revealed that Spain was the country in which the highest correlations were found between HS and BS: \(r =.49,p<.01\) in males and \(r =.64,p<.01\) in females. Data obtained by Moya and Expósito (2008) in a study conducted with 2833 Spanish males (with a mean age of 32.7 years) and 2400 Spanish females (with a mean age of 28.3 years) confirmed the moderately high correlations found between both types of sexism: \(r =.53\) in males, \(p<.01\) and \(r =.58,p<.01\) in females.

With regard to the relationship between HS and BS and sociodemographic characteristics, studies conducted with Spanish samples have shown clear differences between the scores of males and those of females. In the samples mentioned above, Moya and Expósito (2008) found that males scored significantly higher than females both in HS (2.82 versus 1.96) and in BS (2.68 versus 2.37), although differences were greater in HS than in BS; similarly to the results of studies performed in other countries, females rejected HS more than BS, while males showed the opposite pattern. This has been observed even at very early ages. For example, Lameiras and Rodríguez (2002) explored a sample of 406 Spanish students attending Educación Secundaria Obligatoria (compulsory secondary education—ages 12–16). In their study, mean scores of girls were 2.00 in HS and 2.70 in BS while those of boys were 3.20 and 2.82 in HS and BS, respectively .

Other studies conducted in Spain have shown the possible consequences of both BS and HS regarding gender discrimination (e.g., Durán, Moya, & Megías, 2011; Expósito, Herrera, Moya, & Glick, 2010; Moya et al., 2007). Overall, they have revealed that the sexist ideology attributed to a social actor (usually a man) influences individuals’ reactions toward behaviors performed by such actors, particularly when they are imposing or violent. For instance, Moya et al. (2007) conducted three studies in which they explored women’s reactions to ostensibly protective restrictions (i.e., BS). Overall, results showed that only benevolently sexist women accepted a protectively justified (hypothetical) prohibition but mainly when imposed by a husband (not a coworker). These authors concluded that fusing benevolence with dominance and protective paternalism can lead women (especially those who are high in BS) to accept different types of restrictions (e.g., the prohibition of driving on a long trip or opposition to an internship that involved interviewing criminals). Expósito et al. (2010) found that women’s BS (but not HS) predicted viewing a husband as more threatened by his wife’s promotion at work and more likely to perpetrate violence toward her. They concluded that women high in BS may embrace traditional roles in relationships partly to avoid antagonizing male partners, thus maintaining the status quo .

In a different line of research conducted in Spain, Durán et al. (2011) explored how the sexist ideology attributed to men who performed abusive and violent behaviors toward women influenced people’s reactions to such behaviors. Specifically, Durán et al. (2011) portrayed a hypothetical marital vignette in which the husband forced his wife to have sex. Results showed that participants (i.e., women and men) who learned that the husband was high in BS (versus those who received no information about the husband’s sexist ideology) ranked sexual marital rights (for him) and duties (for her) more highly, and regarded forced sex as rape to a lesser extent. The higher participants’ BS scores were, the stronger these effects were. In another study (Durán, Moya, Megías, & Viki, 2010, Study 1), Spanish high school students read about a rape committed by a boyfriend or husband who was described either as benevolently sexist or not. Participants’ BS scores predicted greater victim blame when the rapist was described as a husband (but not a boyfriend) who held benevolently sexist attitudes.

Ambivalent Sexism: Empirical Evidence of Internal and External Factors Involved in its Development

A peculiar feature of sexism compared to other types of prejudice (e.g., religious or ethnic prejudice) is that contact between the in-group and the out-group naturally increases with age, as a consequence of heterosexual intimacy. In the gender relations that take place during childhood, prejudice toward the out-group and self-imposed segregation at times of recreation and leisure typically prevail (Martin and Ruble, 2004; Powlishta, 2003; Powlishta, Serbin, Doyle, & White, 1994; Serbin, Connor, Burchardt, & Citron, 1979; Subirats & Brullet, 1988). Yet, during adolescence , although the stereotypes and prejudices developed during childhood persist, heterosexual individuals presumably start to feel strongly attracted to people of the other gender (Maccoby 1998; Underwood & Rosen, 2009). Based on the AST (Glick & Fiske, 1996), Glick and Hilt (2000) proposed a theoretical model of the development of AS. Along the lines of current theoretical proposals on the development of stereotypes and prejudice, this model considers both internal (e.g., motivational) factors of individuals and external factors to individuals (e.g., social influence) to explain the development of AS. Below, we provide an analysis of the empirical evidence that illustrates this development of gender relations, with a special focus on the evidence obtained in Spain.

Gender Prejudice During Childhood

During childhood, categorization and intergroup hostility as a result of social comparison usually prevail in gender relations (Glick & Hilt, 2000; Maccoby, 1990, 1998). Several studies have consistently revealed the existence of openly negative attitudes toward the other gender and voluntary segregation between boys and girls during early childhood (Martin & Ruble, 2004; Powlishta, 1995). Among the possible causes of this “voluntary segregation” Glick and Hilt (2000) highlight both cultural aspects (e.g., the way families and society in general pay attention to gender in the clothing chosen for boys and girls, the way of talking to boys and girls, etc.; sexism transmitted through the media, toy advertisements, etc.) and motivational aspects (i.e., the desire to identify with a group and differentiate oneself from other groups; Tajfel, 1981) .

The need for identification is one of the basic motivations during childhood (e.g., Kohlberg, 1966; Maccoby, 1998), and gender is one of the most accessible and salient categories to establish a categorization and social comparison between groups (Martin & Ruble, 2004). The distinction between genders emerges during early childhood. Various studies have shown that most boys and girls distinguish and use gender labels between the ages of 18 and 30 months (Campbell, Shirley, & ­Caygill, 2002; Poulin-Dubois, Serbin, & Derbyshire, 1998; see Martin & Ruble, 2010) and self-categorize according to such labels (Stennes, Burch, Sen, & Bauer, 2005; Thompson, 1975). Moreover, identification of gender categories is related to a higher preference for elements that are typically associated with one of the two groups, such as toys (e.g., truck, doll, Zosuls, Ruble, Tamis-LeMonda, Shrout, Bornstein, & Greulich, 2009), and colors (e.g., pink, blue, Karniol, 2011; see Navarro, Martínez, Yubero, & Larrañaga, 2014, for a replication in Spain). From the age of two and a half or three years, boys and girls learn gender stereotypes and also their relationships with gender status differences (Martin, 2000). For example, when 3-year-old boys and girls are asked to identify the gender of a character that shows anger (an emotion typically associated with ways of exerting and showing power and status in patriarchal models), they choose the masculine gender in most cases (Leinbach & Fagot, 1993, cited by Fagot, Rodgers, & Leinback, 2000) .

Researchers have found evidence of openly negative intergroup relations between boys and girls from childhood (Martin & Ruble, 2004; Powlishta, 1995); for example, 3-year-old boys and girls are reluctant to interact with peers of the other gender (Serbin, Connor, Burchardt, & Citron, 1979). Self-imposed segregation according to gender tends to increase throughout childhood (see Maccoby, 1998, 2002). Around the age of 5, individuals evolve toward competitive gender differentiation (e.g., Martin, Ruble, & Szkrybalo, 2002), according to which both boys and girls interact almost exclusively with members of their gender in-group, showing a clear preference for in-group members (i.e., in-group favoritism) and rejection of the members of other groups (i.e., out-group rejection/denigration). Gender differentiation increases with the development of gender constancy, that is, the realization that gender is a stable category over time throughout different contexts or situations (Kohlberg, 1966; Stangor & Ruble, 1987). Gender constancy helps consolidate individuals’ knowledge of which behaviors are appropriate for each gender (Lutz & Ruble, 1995). For instance, Spanish girls score higher in empathy and positive-cooperative conflict-resolution strategies than boys since childhood, and these differences increase with age (Garaigordobil & Maganto, 2011), whereas boys use more aggressive conflict-resolution strategies (Garaigordobil & Maganto, 2011) and already show more direct forms of aggression at primary school (Albadalejo-Blázquez, Ferrer-Cascales, Reig-Ferrer, & Fernández-Pascual, 2013). In short, groups formed during childhood have the same characteristics and play the same role as the social groups of adults; in other words, group membership provides distinctiveness, social identity, a feeling of belonging and mutual support for both boys and girls (see Brewer, 2007) .

However, variables that are external to individuals (i.e., situational variables) also strongly influence the emergence of conflictive relationships between groups. The evidence available shows that boys and girls are socialized differently (cf. Bussey & Bandura, 1999). Therefore they assume a series of social norms and patterns of behavior that are adequate according to the specific context and their own gender. Boys and girls who are exposed to environments in which gender categories are often used are more likely to use such labels to organize their social world (Bigler & Liben, 2007). According to the developmental intergroup theory, developed by Bigler and Liben, gender-based categorization becomes socially salient through explicit cues such as clothing or language. Furthermore, social influences occur both explicitly (e.g., lyrics of songs or rhymes) and implicitly (e.g., nonverbal behavior of adults, observation of interactions between genders and certain social roles in society). In most schools, for example, role distribution according to gender is clear. According to data obtained in the Spanish region of Andalusia, female teachers predominate at initial stages of education (92 % of female teachers in preprimary education) while the percentage of male teachers increases in higher courses (51 % male teachers in secondary education). In addition, organizational and managerial positions are usually held mostly by men (65 % in primary education and 79 % in secondary education; Junta de Andalucía 2005). Furthermore, several studies exploring the transmission of different models of behavior for boys and girls at school have found differences in the interactions between teachers and boys and girls. Both male and female teachers devote more time and pay greater attention to boys than girls (Delamont, 1984; Subirats, 1986), give more feedback to boys than girls on their work (Freixas & Luque, 1998), and give more praise and educational support to boys than girls (Spender & Sarah, 1993).

Another socializing agent of major importance is family. There is broad empirical evidence of the influence of parents on the development of sexist attitudes in their children. Regarding the development of attitudes toward gender roles, a meta-analysis that included 43 empirical studies concluded that parents and children significantly share beliefs on gender roles (Tenenbaum & Leaper, 2002) . Specifically, studies show that mothers have a significant importance in transmitting traditional roles to daughters (e.g., Eccles, Jacobs, & Harold, 1990; Ex & Janssens, 1998; Kulik, 2004; Moen, Ercickson, & Dempster-McClain, 1997; Smith & Self, 1980). In the Spanish population, Montañés, de Lemus, Bohner, Megías, Moya and Garcia-Retamero (2012) found evidence of the transmission of benevolently sexist beliefs from mothers to their adolescent daughters. The BS of both mothers and daughters was even found to be a negative predictor of the academic performance of daughters mediated by their motivation to get an academic degree. The authors of that study did not have access to father-daughter dyads to explore the influence of the sexism of fathers on the sexism of their daughters. Another study with a Spanish sample of 2867 participants (764 mothers, 648 fathers, 768 adolescent daughters and 687 adolescent sons) analyzed the intergenerational connection between the sexism of both parents and that of their sons and daughters (Garaigordobil & Aliri, 2011). The study revealed positive correlations between mothers’ sexism (HS, BS, AS) and their daughters’ sexism (HS, BS, AS) and the BS of their sons; the authors also found positive correlations between fathers’ sexism (BS, AS) and their sons’ sexism (HS, BS, AS, neosexism) but did not find any relationships between the sexism of fathers and that of their daughters (Garaigordobil & Aliri, 2011). The authors of the study suggest that in the future it would be interesting to explore the mechanisms or processes that may explain this relationship (i.e., imitation, deliberate transmission of ideology or other sources of influence such as religion, politics, or sending one’s children to certain schools or enrolling them in certain activities; Garaigordobil & Aliri, 2011) .

Another important variable that should be considered to understand the development of gender stereotypes and prejudice is the prevailing culture and its manifestations through materials that children are exposed to: for example, traditional stories (Colomer, 1994; Turin, 1995) and games or toys that are differentiated according to gender (Martínez Reina & Vélez Cea, 2006). In Spain, an analysis of 20 classic children’s stories published at the end of the 1980s (i.e., Almodóvar, 1987) revealed that the number of male protagonists was higher than that of female protagonists and that in 90 % of cases females were depicted as subordinate to males, even when they were queens or princesses. In addition, 80 % of the females depicted in these stories were in charge of household tasks, while 75 % of the intellectual activities described (referring either to professional or conflict resolution activities) were performed by males (Pérez-Grau, 2006). In fact, the latest report on the campaign of toy advertisements conducted in 2010 by the Observatorio Andaluz de Publicidad No Sexista (the Andalusian observatory of non-sexist advertising, an advisory body of the Spanish regional government of Andalusia) reported that 63.49 % of the advertisements for toys and games analyzed that year in Spain contained sexist treatment. A high percentage of such advertisements (85 %) promoted models that consolidated traditional patterns for each gender (e.g., toys related to household tasks for girls) and 17 % of advertisements promoted beauty standards that were considered as a synonym of success and were always targeted at girls (Observatorio Andaluz de la Publicidad No Sexista, 2010). In short, boys and girls are constantly exposed to models associated with gender through stories, games (including video games), films, and television (see the review conducted by Bussey & Bandura, 1999).

From an early age, the behavioral manifestations of boys and girls indicate that both groups assume social norms and patterns of behavior perceived as adequate according to the specific context and their own gender. According to Pellegrini and Long (2003), during the time that both gender groups spend separately during childhood, boys develop the necessary skills to maintain their dominance and status, using physical violence in games. For example, groups of boys tend to organize themselves very quickly according to a hierarchy, while hierarchies are less marked in relationships between girls (Savin-Williams, 1980). Further, boys appropriate themselves of larger territories in games (Thorne, 1986) and are more likely to interrupt girls’ activities (Subirats & Brullet, 1988). Overall, boys’ games are usually exclusive for boys; they tend to refer to heroes and adventures and imply danger and aggressiveness, without developing romantic ideals (Flannery and Watson 1993). However, these “scripts” or patterns of behavior can later become behaviors of protection and paternalism of boys toward girls during adolescence , and boys often act as strong and brave “knights” once other interpersonal motivations come into play (Glick & Hilt, 2000). By contrast, girls are socialized to be passive and adhere to traditional roles (Rudman & Glick, 2008). From an early age they learn scripts based on fairytales that encourage them to become “princesses” highlighting the importance of physical appearance and the goal of finding a “Prince Charming” who will take care of them and protect them (see Rudman & Glick, 2008; Walkerdine, 1984). From the age of 4, girls prefer romantic fairytales while boys prefer adventure stories (Collins-Standley, Gan, Yu, & Zillman, 1996). The romantic idealization of males as knights in shining armor who take care of and “rescue” helpless females, internalized during childhood (e.g., Prince Charming, knight errant, protector, hero) is activated by adult women at an implicit level, which implies that such associations (e.g., men-savior) are strongly learned and rooted in the memory of women (Rudman & Heppen, 2003). In short, during childhood, relationships between girls and boys are characterized by “gender segregation” due to internal (i.e., cognitive) and external (i.e., socialization) factors (Maccoby, 1998). In fact, games shared by boys and girls tend to be limited to those initiated by teachers or other adults (Fabes, Martin, & Hanish, 2003). At this stage, the behavior of boys and girls reflects competitive gender differentiation and the domination of boys in games and spaces, which is characteristic of hostile prejudice.

Gender Prejudice During Adolescence: Ambivalent Prejudice

The gender segregation that characterizes childhood decreases at the beginning of adolescence, when individuals start to interact more with and show greater interest in peers of the other gender (Cairns, Leung, & Cairns, 1995; Pellegrini, 1994) . According to Glick and Hilt (2000), adolescence is the key period in the development of the foundations of AS that will later prevail during adulthood; these authors consider that this is due to biological factors (e.g., puberty) and social factors (e.g., expectations, social norms). During childhood , affiliation motivation is materialized into in-group identification and social comparison with the out-group; during adolescence, however, this motivation is transformed into a desire to generate a positive affiliation with people of the other group. During adolescence, individuals start to have a strong curiosity and interest in getting to know the other group; this is partly due to the motivation of interdependence, understood as the need to share a relationship with a person of the other gender for reproductive and affective reasons (see Rudman & Glick, 2008, Chap. 9) . The search for heterosexual intimacy leads adolescents to develop attitudes and beliefs that allow them to approach members of the other gender group, who had been mostly ignored or even rejected until then. The emergence of heterosexual romantic impulses during adolescence interacts with the gender differentiation and power differences developed since childhood (Glick & Hilt, 2000).

In the process of developing new interactions of a romantic nature or to seek intimacy between the genders, adolescents tend to use gender clichés, stereotypes and scripts learned during their childhood and observed in their immediate environment through the media, games and other modeling processes mentioned above (cf. Bussey & Bandura, 1999; Louis, Stork-Brett, & Barlow, 2013). The analysis of the content of magazines targeted at adolescent girls shows that the main topics they cover are relationships, dates and tips to be attractive for boys (Pierce, 1990). In Spain, a comprehensive analysis of the content of magazines for adolescents conducted by the Instituto Asturiano de la Mujer (2005), the Women’s Institute of the Spanish region of Asturias, revealed that such magazines promote the roles of “princess” for girls and “Prince Charming” for boys. An analysis of the contents of the covers of magazines for adolescents in Spain revealed that magazines targeted at girls use stereotypical models on their covers, imposing a model of ideal woman that requires physical and aesthetic perfection and does not correspond to reality (Blanco-García & Leoz, 2010). These influences are particularly significant during adolescence, given that gender role expectations increase in both female and male adolescents (Hill & Lynch, 1983; O’Sullivan, Graber, & Brooks-Gunn, 2001) .

According to the explanatory model developed by Glick and Hilt (2000), during the first stages of adolescence (i.e., pre-adolescence), boys and girls start to interact with members of the other gender and to develop gender subtypes (e.g., “tomboys”—girls that are not very feminine; “girls,” “sissies”—effeminate boys, Six & Eckes, 1991). According to Glick and Hilt, boys start to rate stereotypically feminine traits very favorably and even to idealize them (e.g., warmth, sensitivity) and to develop benevolent attitudes toward certain subtypes of girls with whom they would like to have intimate relationships; by contrast, they negatively assess and direct their HS toward girls that pose a threat to male domination and that are perceived as competitors (e.g., those who do not have stereotypically feminine traits, excel at school or are determined to have a career). As girls start to have romantic relationships with boys, they tend to act as “objects of love,” assuming the benevolence into which they were socialized and conferring great importance to their ability to attract partners of the other gender (Martin, Luke, & Verduzco-Baker, 2007) . On the other hand, the “chivalry” of boys is important at the beginning of intimate relationships: they take the initiative and actively try to seduce the girl; girls, by contrast, remain passive, try to attract the boy and in most cases decide who they want to have sexual relations with and how far they want to go in such relations (Rose & Frieze, 1993). Benevolent attitudes, which are congruent with the romantic scripts of girls’ childhood (Feiring, 1996; Holland & Einsenhart, 1990; Rudman & Glick, 2008), are seen as more socially desirable than hostile attitudes; adolescent girls feel attracted to boys defined as benevolently sexist the more experience such girls have in intimate relationships (Montañés, de Lemus, Bohner, Moya, & Megías, 2013) or the more accessible their experience in intimate relationships is (Montañés, Megías, de Lemus, & Moya, in press). These data confirm that BS has a reinforcing effect for adolescent girls, efficiently and insidiously maintaining traditional gender roles .

The changes in gender relations during adolescence increase the complexity of sexist attitudes and do not imply a decrease of prejudice but rather a change in the way it is expressed (Furman & Wehner, 1997; Glick & Hilt, 2000; Maccoby, 1990). According to Glick and Hilt, intergroup hostility does not disappear but instead becomes conditional (e.g., boys who used to say they hated girls may now only feel degraded by some specific types of females such as feminists; see Glick, Diebold, Bailey-Werner, & Zhu, 1997). In line with the predictions of the model proposed by Glick and Hilt, de Lemus, Moya and Glick (2010) corroborated in Spanish samples that, despite the overall decreasing trend of sexism with age (Lameiras & Rodríguez, 2002, 2004; de Lemus, Castillo, Moya, Padilla, & Ryan, 2008), the increase in early experiences of intimate relationships between adolescents predicts an increase in AS in individuals of both sexes. Specifically, after statistically controlling for the effect of age, they observed that experience in romantic relationships predicted higher BS in boys in general and higher HS only in younger boys (ages 12–14). As for girls, their experience in romantic relationships was correlated with higher scores in HS but not in BS. More recent studies with an experimental design showed that the accessibility of intimate relationships increased both HS and BS in boys and BS in girls (Montañés et al. in press) . This confirms the influence of intimate relationships on the sexism of adolescents of both genders. This may indicate that adolescents activate sexist ideology, particularly in the intimate relationship environment, as a form of control to ensure their partners adjust to the traditional gender roles and stereotypes they have learned from early childhood .

Results of the above-mentioned studies (de Lemus et al., 2010; Montañés et al., in press) corroborate the proposal made by Glick and Hilt (2000). Specifically, boys assume benevolent attitudes as they start to become interested in having relationships with girls, perhaps as a way to explain their attraction to them without contradicting their prior hostility during childhood. BS makes it possible to reconcile affection for female partners with traditional roles. Moreover, from an instrumental point of view, assuming benevolently sexist beliefs may increase boys’ chances of success in trying to initiate intimate relationships, given that adolescent girls rate profiles of benevolently sexist boys as the most attractive (Montañés et al., 2013) .

In short, there is little empirical evidence available so far on the subject and results need to be corroborated experimentally by longitudinal studies; yet, the findings obtained so far in Spain suggest that having experience in romantic relationships (i.e., having had previous dates) or thinking about them (i.e., making them accessible) may initially not decrease but rather increase gender prejudice (de Lemus et al., 2010; Montañés et al., in press) or increase girls’ ratings of potential benevolently sexist heterosexual partners compared to non-sexist ones (Montañés et al., 2013).

Conclusions

As proposed by the main current theoretical approaches of social and developmental psychology, both internal factors (i.e., cognitive development, motivation) and external factors (i.e., social influences) are keys to understand the development and maintenance of prejudice during childhood and later during adolescence . Specifically, gender prejudice differs from other intergroup conflicts in that males and females are segregated in various spheres of public life but have intimate and private relationships in other spheres (e.g., Glick & Fiske, 1996; Pratto & Walker, 2004). Considering this peculiarity of gender relations, we reviewed the theory of development of AS (Glick & Hilt, 2000), providing empirical evidence of its validity, mainly from Spanish studies. This theory postulates that culture is mostly responsible of the origins of gender prejudice during childhood; yet, motivational forces (e.g., in-group favoritism, out-group rejection) may be responsible for the strength of this prejudice during childhood (Brown, 1995). During this stage, relationships between the genders are characterized by hostility and segregation (Maccoby, 1998, 2002); during adolescence, however, prejudice evolves toward more ambivalent forms as a consequence of heterosexual dependence (Glick & Hilt, 2000; Rudman & Glick, 2008).

As interactions between boys and girls start to increase during adolescence, open hostility between groups decreases (e.g., Connolly, Craig, Goldberg, & Pepler, 2004). In this regard, intimate heterosexual contact facilitated by the growing romantic interest in members of the other gender can be seen as a way of reducing prejudice understood as intergroup hostility. However, more subtle forms of prejudice emerge and are particularly efficient at maintaining social inequality (e.g., Jackman, 1994; Major & Vick, 2005; Vescio et al., 2005; Moya et al., 2007). The initiation of heterosexual romantic relationships during adolescence promotes the maintenance of sexism, transforming it into more complex ideological structures based on ambivalence (i.e., including both positive–paternalistic–attitudes and negative–hostile–attitudes toward women). Thus, although romantic relationships may bring together both gender groups, this in fact tends to perpetuate rather than eliminate status inequalities (de Lemus et al., 2010; Montañés et al., in press).

The evidence presented in this chapter was obtained in the framework of developmental and social psychology. It highlights the importance of studying sexism from childhood and particularly of focusing on adolescence, considering motivational factors associated with the start of romantic relationships. Gaining greater knowledge of the mechanisms that lead to the acceptance of such sexist beliefs will help to develop effective interventions to reduce the impact of sexist beliefs and replace them with other beliefs based on equality between women and men (cf. de Lemus, Navarro, Velásquez, Ryan, & Megías, in press).