Those studies meeting the following criteria were selected: (a) the relationship between sexism and cognitive, emotional, attitudinal, and behavioural variables involved in a heterosexual relationship was analysed, given that sexism implies an unequal relationship between men and women; (a1) sexism was considered to be analysed when traditional and new conceptions were used, including sexism, ambivalent sexism, hostile sexism, gender or traditional sexual role ideology, machismo, benevolent sexism, neosexism, and modern sexism; (b) the sample was made up by adolescents with an age range between 10-19 years, following the United Nations definition of adolescence (UNICEF, 2011). Likewise, the origin of the sample (for example, secondary schools, colleges, or first-year university courses and equivalent, depending on the education system of each country) was taken into account when it could be deduced or it was declared that the largest percentage of the sample was made up by adolescents with an age range of between 10-19 years; (c) they were research articles published in scientific journals and peer reviewed; (d) the languages of publication were English, Spanish, or Portuguese; (e) years of publication were from 2005 to 2018; and (f) they were not theoretical studies or reviews.

The literature search was conducted until June 2018 in the following primary databases (Perestelo-Pérez, 2013): PsycINFO, Psicodoc, PubMed, Scopus, and Web of Science. The following search strategy was used: << (sexism* OR sexist* OR neosexism* OR machism* OR “ideologia* de genero” OR “ideologia* de rol sexual” OR “gender ideolog*” OR “sexual role ideolog*” OR “ideologia* de gênero” OR “ideologia* de papéis sexuais”) AND (adolescent* OR teen*) >>, except for Psicodoc database, in which both the boolean operators (OR by O and AND by Y) and the truncation operator (* by $) were replaced, maintaining the same descriptors. These were inserted into the search fields for title, abstract, and keywords. A secondary search was also carried out by reviewing the bibliographic references of the articles that were included. Subsequently, taking into account the inclusion criteria, the following limitations were established: (a) year of publication: from 2005 to 2018; (b) language of publication: English, Spanish, and Portuguese; and (c) type of document/source publication: scientific journal/article, as the database would allow.

Abstracts of studies resulting from the search were analysed in order to verify compliance with the inclusion criteria, as well as the full text when the summary could not provide all required information.

Once the articles meeting the inclusion criteria were selected, the following information for each of them was extracted:

A total of 1,170 articles were obtained, of which 491 were duplicates and 659 were discarded for not meeting the inclusion criteria. The main reasons for exclusion were (by order of frequency): age range of participants, missing analysis of the relationship between sexism and a variable involved in the heterosexual relationship, and the definition of the sexism variable. Thus, 20 articles, which were accordingly reviewed, were selected.

According to the research methodology, 15 (75%) were ex post facto studies, two (10%) were experimental studies, two (10%) were instrumental studies, and one (5%) was a quasi-experimental study. Regarding the country of origin of the sample of adolescents, they were of Spanish origin in 13 studies (65%), they were American in four studies (20%), two studies (10%) shared samples from Bolivia and Ecuador and from Spain, Chile and Colombia, respectively, and Chinese in one study (5%). Regarding gender, 17 studies (85%) included women and men, two (10%) included only men, and one (5%) included only women.

On the other hand, versions and adaptations of five assessment instruments for sexism were used in the studies included: the Ambivalent Sexism Inventory (ASI; Glick & Fiske, 1996), the Attitudes toward Women Scale (AWS; Spence & Helmreich, 1972), the Neosexism Scale (Tougas et al., 1995), the Sex Role Egalitarian Scale (King & King, 1990), and the Sexism Detection Scale (DSA; Recio, Cuadrado, & Ramos, 2007).

Finally, with regard to the thematic areas of the cognitive, emotional, attitudinal, and behavioural variables involved in a relationship associated with sexism, 11 studies were found to include variables related to intimate partner violence, 4 studies were related to relationship experience, 3 studies related to sexual behaviour, and 1 study was concerned with each of the following areas: attraction to sexist partners, myths towards love, emotional dependence on the partner, and engagement quality. The number of studies per thematic area exceeds the 20 studies included since two studies (Pradas & Perles, 2012; Viejo, Ortega-Ruiz, & Sánchez, 2015) can be classified in more than one thematic area.

Table 1 shows the main features and results collected of the studies included in this theoretical study, organized according to the thematic areas.

The studies reviewed relating sexism with intimate partner violence in adolescent population show conflicting results. Thus, whereas in the study of Allen, Swan, and Raghavan (2009) women with more benevolent sexism report suffering less violence and benevolent sexist, men report committing less violence against women; other studies do not find associations with respect to benevolent sexism and the perpetration and victimization of sexual violence (Fernández-Fuertes, Carcedo, Orgaz, & Fuertes, 2018) or the perpetration of cyber-aggression (Rodríguez-Domínguez, Durán-Segura, & Martínez-Pecino, 2017). However, other studies show that both more traditional and benevolent sexism are related to a further justification of domestic violence (Garaigordobil & Aliri, 2013; Shen, Chiu, & Gao, 2012), greater acceptance of the use of physical violence against women by men in relationships (Lee, Begun, DePrince, & Chu, 2016; Pozo, Martos, & Alonso, 2010), and greater violence committed (Fernández-Fuertes et al., 2018; Pazos, Oliva, & Hernando, 2014; Reyes, Foshee, Niolon, Reidy, & Hall, 2016) and suffered (Anacona, Cruz, Jiménez, & Guajardo, 2017) in couples. In addition, more traditional sexism predicts greater use of violent strategies of conflict resolution within the couple (Pradas & Perles, 2012) and the perpetration of cyber-aggression towards women by men (Rodríguez-Domínguez et al., 2017). Regarding gender differences, there are also contradictory results. The research by Pazos et al. (2014) found that high levels of sexism are associated with greater sexual, relational, verbal-emotional, and physical violence threats in men, whereas it is only associated with greater relational violence in women. However, the study by Anacona et al. (2017) shows that in women only hostile sexism is associated with severe physical aggressions perpetrated, whereas benevolent sexism is associated with perpetrated mild physical violence, and in males hostile sexism is associated with mild psychological and physical aggressions, whereas benevolent sexism is linked to both perpetrated and suffered psychological aggression,. In addition, in other studies, such as Shen et al.’s (2012), there are no associations between traditional gender roles and perpetration and victimization of violence in women, whereas they predict sexual violence but not physical violence in men.

Regarding the relationship experience of adolescents and sexism, the results of the studies are also contradictory (De Lemus, Castillo, Moya, Padilla, & Ryan, 2008). In relation to women, a study found that greater relationship experience predicts higher levels of benevolent sexism (Montañés, Megías, De Lemus, & Moya, 2015), whereas another study shows that, although it is associated with higher levels of both types of sexism, only predicts higher levels of hostile sexism (De Lemus, Moya, & Glick, 2010). Finally, another study only found positive associations between relationship experience and benevolent sexism (Viejo et al., 2015). With regard to men, whereas a study found out that greater relationship experience predicts higher levels of both types of sexism (Montañés et al., 2015), another one indicates only prediction for benevolent sexism, and only prediction for hostile sexism when this relationship is mediated by age, only in adolescents from 12 to 14 years of age (De Lemus et al., 2010). On the contrary, Viejo et al. (2015) found that greater relationship experience is associated with greater hostile sexism.

With regards to sexual behaviour and sexism in adolescents, based on the results of the reviewed studies, there is a positive association between traditional gender ideology and beliefs about the use of contraceptives that pose a greater risk, such as lesser belief in easy access to contraceptive methods (Grose, Grabe, & Kohfeldt, 2014). Likewise, another study reveals that men with sexual experience (coital/non-coital) demonstrate greater sexist beliefs (hostile and benevolent), compared to those without sexual experience. This difference cannot be found in women (Ramiro-Sánchez, Ramiro, Bermúdez, & Buela-Casal, 2018b). Furthermore, results obtained by De Meyer et al. (2014) indicate that in sexually active adolescents traditional gender ideology is associated with a reduced use of contraceptives, more negative sexual experiences and more difficult communication among partners about sex, compared to those who have a more egalitarian gender ideology. In adolescents without sexual experience, traditional gender ideology is associated with more negative feelings towards sexual experiences, a greater belief that it is necessary to have sexual intercourses to maintain a relationship and more difficult communication among partners about sex, compared to those who have a more egalitarian gender ideology. With regard to gender differences, the study conducted by Ramiro-Sánchez et al. (2018b) demonstrates that in men greater benevolent sexism is associated with vaginal sex initiation at an earlier age and greater hostile sexism is linked to a lesser proportion of condom use. By contrast, benevolent sexism is only associated with a greater number of sexual partners in women.

The results regarding the relationship between sexism and attraction to sexist partners reveal that adolescents with higher benevolent sexist beliefs report greater attraction to partners who are benevolent sexists. In terms of gender, hostile sexist men are attracted to women with hostile sexist beliefs, whereas this is not found in women (Montañés, De Lemus, Moya, Bohner, & Megías, 2013).

Regarding the link between sexism and myths towards love, a positive association between benevolent sexism of adolescents and support of the idealized myth of love is shown in the study by Rodríguez et al. (2013), whereas hostile sexism is associated to a lesser extent. Regarding the myth of love-abuse link, hostile sexism is far more associated with it than benevolent sexism.

Regarding sexism and emotional dependence on the partner in the adolescent population, adolescents with higher levels of sexism show more emotional dependence on their partners (Pradas & Perles, 2012).

Regarding links between sexism and engagement quality in adolescents, stronger support for sexist attitudes is associated with poorer quality in relationships according to the study by Viejo et al. (2015), which means more conflicts, more power imbalance, and more transgressive behaviour within the couple.

The limitation of this theoretical study is the fact that the search is limited to the years 2005 to 2018, in order to specify a recent time interval, given the cultural changes that could affect studies on sexism conducted during a different time period (Connor, Glick, & Fiske, 2016). This also means that anything published from 2018 cannot be included. For this reason, as a future line of research, it would be advisable to carry out regular systematic reviews to enable updates on the matter.

Regarding the practical implications of this theoretical study, firstly, it is recommended that further research be carried out on how sexist attitudes influence the relationships of adolescents, using an experimental methodology, thus making it possible to draw causal conclusions. Secondly, the results of the studies in this review suggest the need to prevent and intervene in reducing sexist attitudes in adolescent populations. The aim of these interventions would be to prevent harmful forms of intimate interaction among adolescents based on the differentiation of roles (Fernández, Quiroga, Escorial, & Privado, 2016) and the inferiority of women compared to men, highlighting the importance of intervention in positive development in adolescent age (Eichas, Kurtines, Rinaldi, & Farr, 2018). In addition, it is recommended that the existing gender differences are taken into account in order to develop gender-specific interventions. In the same way, an important contribution is related to the clarification between the different conceptions of sexism, which will undoubtedly contribute to the correct use of the term at a practical level.

In this systematic review, it is concluded that maintaining sexist attitudes is related to harmful forms of intimate interaction among adolescents. Therefore, adolescents with sexist attitudes have more positive attitudes towards intimate partner violence (Anacona et al., 2017; Fernández-Fuertes et al., 2018; Garaigordobil & Aliri, 2013; Lee et al., 2016; Pazos et al., 2014; Pozo et al., 2010; Reyes et al., 2016; Rodríguez-Domínguez et al., 2017; Shen et al., 2012), greater risky sexual behaviours (De Meyer et al., 2014; Grose et al., 2014; Ramiro-Sánchez et al., 2018b), greater attraction to sexist partners (Montañés et al., 2013), greater support for the myth of idealized love and the myth of love-abuse link (Rodríguez et al., 2013), greater emotional dependence on the partner (Pradas & Perles, 2012), and poorer quality in relationships (Viejo et al., 2015).