Social Invisibility and Socio Cultural Construction of Gender in Historical Narratives of Chilean High School Students

This research seeks to evaluate the degree of inclusion of the gender perspective and the promotion of education in and for equality in the historical narratives of students in Chilean Secondary Education (n = 105). The study focuses on the analysis of the discursive-narrative mechanisms employed by the students and, in particular, of their representations of gender relations. To this end, the place given to men and women in past and present societies is analysed in the narratives generated by the students (n = 780). A mixed methodology is applied, combining quantitative (descriptive and inferential) and qualitative analyses of the manifest content. The results obtained report the persistence of stereotypical, exclusionary and androcentric perspectives, evidenced in hegemonic gender attributions and in the maintenance of the sex/gender system. These data confirm the need to implement specific teacher training programmes aimed at the acquisition of critical competences and the effective inclusion of the gender perspective in history education.


Introduction
Key insights for understanding the patriarchal system can be found in gender studies and queer theory. Authors such as Butler (2001Butler ( , 2006 and Spivak (2012) criticise the transmitted concept of woman and being a woman. These authors agree that "woman" as a concept and symbol is an incoherent social construction, which responds to historical contexts and, above all, to the power structures inherited from patriarchy and heteronormative conceptions. In this sense, social structures, ethnicities, sexuality and identities have traditionally been dominated and reproduced by male hegemonies and elites. Spivak (2012) agrees with Scott (2008) that the problem of women's absence and marginalisation is not only solved by their inclusion in narratives that propose them as protagonists, because this would mean maintaining the status quo of gender inequality and the subsidiary logic that governs patriarchal structures. The way forward must be to understand how these patriarchal categories have been constructed, as well as the hierarchies that have positioned men's history above women's actions, narratives and stories. From this perspective, Butler (2001) suggests the deconstruction of traditional concepts such as man, woman, mother or gender, in order to establish new categories that respond to social and pedagogical realities in contexts of diversity. It is necessary to construct new categories that open spaces for difference, dialogue, inclusion and participation from a social justice perspective (Fraser, 2008).

Gender in history education
Different research has addressed the teaching of history from a gender perspective in the Ibero-American region (Marolla, 2019a;Ortega-S|nchez & Pagès, 2018;Pagès & Sant, 2012). Similarly, from the scientific field of history education, the inclusion of the gender perspective and, in particular, of women and their historical experience, has received increasing interest in the whole of its research concerns. Recent works such as those by Ortega-S|nchez and Pagès (2020), Ortega-S|nchez and Barba Alonso (2021) and Marolla et al. (2021) have in common their concern for the However, authors such as Bickmore (2008), Aguilar (2013) and Crocco (2018) consider that school can be a space to generate change and promote social justice. The inclusion of women and their historical experiences in the teaching of social sciences must take place from reflective and critical perspectives (García Luque, 2013;Rodríguez Martínez, 2014). In this sense, if this inclusion is approached from a critical perspective, girls could identify with a history that positions them as protagonists and agents of change (Crocco, 2018;Marolla, 2019aMarolla, , 2919b, and empower themselves in the struggle for gender social justice (Díaz de Greñu & Anguita, 2017;Díez Bedmar, 2015).
The permanence of hegemonic gender structures is explained by the naturalisation and normalisation of its meanings (Tomé & Rambla, 2001). Gender and femininity are constructions that come from a counter-position to the characteristics attributed to men (Butler, 1999). Studies agree that the exclusion of women and their historical experiences in dominant narratives refers to a social production constructed from relations of power, domination and subordination (Foucault, 2008).
In this context, this research seeks, on the one hand, to evaluate the level of inclusion of the gender perspective in the historical narratives of the third and final year of the Chilean baccalaureate. On the other hand, it aims to analyse the specific representations of unequal gender relations in these narratives.
Considering narrative competence as an inherent part of the development of inclusive social historical thinking (Peck, & Seixas, 2008;Sant et al., 2014), this research aims to assess the degree of inclusion of the gender perspective and the promotion of education in and for equality in the historical narratives of Chilean secondary school students. The study focuses its interest on the analysis of specific representations of gender relations, based on the following research questions: What is the place of men and women in past and present societies in the narratives generated by Chilean secondary school students? How do students' historical narratives contribute to promoting social participation and intervention in social problems such as unequal gender relations?

Research design
The study is developed in ex post facto non-experimental research designs, in which the independent variables are not manipulated or cannot be intervened in because they have already occurred (Hern|ndez et al., 2010). Cross-sectional quantitative and qualitative methodological principles (Denzin & Lincoln, 2018) of manifest content were applied in order to obtain a description of the study variables and to make interpretative inferences at a specific point in time. The ideas of Cohen et al. (2007) in the same vein and the principles of critical theory (Apple, 2014;McLaren & Kincheloe, 2015) were followed as a theoretical-reflexive axis around gender inequality.

Participants
The study involved two schools and four classroom groups in the city of Antofagasta, Antofagasta Region in Chile. The contextual difference between the two schools lay only in its type of funding: public sector (school 1) and private sector (school 2). Both schools belong to a middle socio-economic profile. The first school had one classroom group and the second had three (Table 1). For the selection of participants, accidental or convenience sampling was used, depending on the possibilities of access of the research team to the field of study and the degree of appropriateness to the research objectives. The participants were enrolled in the junior (16-17 years) and senior (17-18 years) years of Chilean Secondary Education. In the first school, one senior class-group participated, and in the second school, two first-year and one senior class-group participated. After receiving the teachers' authorisation and signed informed consents, a total of 112 students participated, of whom 105 (72 girls, 31 boys and 2 non-responders) returned a historical discourse that responded to the proposed research objectives.
In order to select the schools, intentional criteria were used, seeking to ensure that they were in line with the objectives and the possibilities of the research team to access the field of study (Bisquerra, 2004;Rodríguez et al., 1999). In this sense, the criteria of Simons (2011) and Stake (2007) were considered both in the selection of schools and participants. The total sample of participants was not based on probabilistic criteria of representativeness, but on its ability to explain potential differences, and to relate and understand the information that students could provide from different perspectives.

Instrument
A questionnaire was administered with a single open-ended question: "What do you know about the history of Chile?" following the method of Létourneau and Caritey (2008). To organise the data obtained and proceed with the narrative analysis, we used the categorical scale, variables and levels of progression for narrative analysis of , structured in three explanatory categories: historiographical approach, historical protagonists and protagonist spheres of social action. However, the different geographical context in which the instrument was applied made it necessary to make adaptations, and to create new variables and progressions in order to respond to the objectives of the study. To adapt the variables applied in this research, the empirical research model related to inductive grounded theory was used. This model has been designed and applied in the research of Ortega-S|nchez and Pagès (2018), Ortega-S|nchez (2017), Gonz|lez-Monfort et al. (2015) and Sant et al. (2014).
Assuming the empirical possibility of achieving a normal behaviour or distribution in random variables of an ordinal nature, the reliability of the instrument was assessed using Cronbach's alpha. In this first analysis, in which all the original variables of the Ortega-S|nchez et al. (2020) scale were included, an unsatisfactory result was obtained (α < 0.5). This result can be attributed to the fact that the original variables were applied to narratives provided by Spanish students.
Therefore, the reference scale was adapted by eliminating variables 6, 9 and 10 (unanswered in the Chilean context), and those that had a lower correlation with the total score of the instrument. The final variables were as follows: V1 . For measurement, three levels of narrative approximation were established: level 1 (0 to 3 conceptual or nominal recurrences), level 2 (4 to 6 conceptual or nominal recurrences); level 3 (7 or more conceptual or nominal recurrences) ( Table 2). The narrative appears starring countries, social groups and/or politicalinstitutional entities humanized in proto-structures of apparent male domination. Collective anonymous 1-3 The narrative is articulated by groups of collective anonymous individuals, who act as the central protagonists of the stories.
Leading women 1-3 The narrative shows a tendency to prioritize figures of female historical protagonists, builders of the social-historical story.
2 Economic 1-3 The social action of the selected historical protagonists takes place in local or global scenarios around economic interests and/or decision-making. C1: Category 1 (historical protagonists); C2: Category 2 (leading scope of social action). V: Variable. NPr: Ordinal progression levels.
To check the accuracy, stability, internal consistency, correlation between variables and ability to measure the same construct avoiding random errors in measurement, a new calculation of Cronbach's alpha was carried out. Therefore, the study variables were modified and recoded to adjust them to the Chilean context before obtaining the final result. The analysis obtained an adequate level (α = 0.8).
In this new theoretical-inductive definition of the variables and categories, compliance with the measurement criteria (completeness, exclusivity, precision, relevance or categorical adequacy to the research objectives, and homogeneity or categorical responsiveness to a single classification principle) was ensured. Its feasibility was also assessed after pilot application on a sample from the same documentary population (n = 23).
Additionally, to determine the content validity of the instrument, we relied on the evaluation of four experts in social science teaching, as well as other researchers who have applied a similar instrument (

Procedure
To obtain the narratives, the method of Létourneau and Caritey (2008) was used and, for the data dump and narrative analysis, the categorical scale, variables and levels of progression of narrative analysis, designed and applied by , modified for adaptation to this research, was used. Data collection took place between July and September 2019 in the city of Antofagasta, Antofagasta Region of Chile, seeking to ensure that the students' responses were generated in the everyday context of the classroom.
The students were given two blank sheets of paper to write their answers. In addition, no writing instructions were given, in order to foster an atmosphere of trust and freedom in the construction of their texts.

Data analysis
The content was analysed through a constant comparison of data (Libarkin & Kurdziel, 2002), a thematic analysis based on the identification of response patterns (Denzin & Lincoln, 2018) and the association of narrative extracts obtained with the categories of analysis. This qualitative analysis allowed us to understand, from the discursive expressions, aspects of discursive and chronological organisation, and narrative relevance.
Aspects such as reliability, credibility and transferability of the results were guaranteed by the research team through discussion, reflection and adjustments to the categories worked on (Simons, 2011). The content of the recording units was validated by confirming its possible assignment to a single variable and category of analysis, the existence of a clear explanatory capacity of the categories for each of the study variables that made them up (content validity), and the degree of semantic similarity between the coding units and its categorical classification (semantic validity). Likewise, the empirical categories of the study were confirmed in accordance with the research objectives. In addition to checking the appropriateness of the proposed categorical system (validity), the consistency of the measurement procedure (reliability) was also checked. Procedural reliability was ensured by obtaining a high degree of inter-coder agreement (research team coder 1 and external researcher coder 2) in relation to the criteria of stability (absence of ambiguity in the assignment of recording units to a variable and category) and reproducibility (inter-researcher agreement in narrative coding) (Krippendorff, 1990). The reliability of coding and categorisation (Muñoz & Montoro, 2005) was also confirmed by applying Scott's π index (π = 0.87).
In order to obtain the saturation of the data, a count of structuring keywords was applied, through the registration of its absolute and relative frequency of appearance, a procedure applied in other studies in the educational field (Conde & Armas, 2019). Each variable grouped key concepts and names according to affinity and historical-narrative meaning.
The documentary corpus (n = 112) was identified by alphanumeric codes (Ccenter nº_ Nstudent nº_ Aclass group nº). The narrative extracts (n = 780) were assigned to one of six emergent variables: political-military historiographical focus, leading men, quasi-characters, female characters, economics and other topics.
The R Project v.3.6.1 software was used for its organisation, coding and quantitative analysis; on the other hand, the ATLAS.ti v. 7.5.4 software for qualitative data was used for the calculation of frequencies and percentages of key words.
A quantitative descriptive analysis was performed (mode and median), and non-parametric statistics were used, as the data did not meet the hypotheses of normality (Kolmogorov-Smirnov test) and homoscedasticity (Levene test). The Kruskal-Wallis test was applied in order to identify statistically significant differences between classroom groups and the Mann-Whitney U test to test for potential differences according to gender and school. Finally, the Chi-square test of independence was applied to test for the existence of an association between variables and Spearman's correlation coefficient to calculate the strength of correlation. The significance level established was p < 0.05, with a confidence level of 95% and an error of 5%.
Both participants and teachers were informed of the aims of the research, guaranteeing anonymity in both the collection and processing of the responses provided. Informed consent was given to students and teachers explaining in detail the aims of the study, in accordance with the ethical protocols defined by the Helsinki declaration. Thus, the confidentiality and anonymity of all participating parties was guaranteed (Creswell, 2014;Stake, 2007).

Results
According to the recurrences identified in the narratives of 39 students, the majority mention leading men' (Mo = 3), meaning that male characters were cited on more than seven occasions (37% of the total). Significant statistical differences were identified as a function of classroom-group (p = 0.023), with classroom 3 being the group with the most recurrences at level 3 (f = 16). There were no significant differences in this variable according to centre (p = 0.216) ( Table 3). An important finding is obtained in the contrast according to the gender of the students. In this case, statistically significant differences were found in the variable 'leading men', with girls responding most frequently at level 3 (42%) ( Table 4). The variable V2 (political-military characters) reaches a medium level of narrative approximation (Mo = 2). It was also found that there are significant statistical differences of 6% (p = 0.059) depending on the classroom-group in this variable (Table 5). Applied the Chi-square test of independence, was found that anonymous-collective characters (V4) and politicalmilitary characters (V2) are statistically independent (p = 0.598). Similarly, the variable 'leading men' (V1) is statistically independent of the variable 'political-military characters' (V2) (p = 0.921).
In relation to the variable 'quasi-characters', it was found that there are no significant differences according to classroom group (p = 0.342) and school (p = 0.608). Also, a low frequency of quasi-characters is observed with a value Mo = 1 (45%), which means that students mentioned quasi-characters up to three times (Table 6). On the other hand, statistically significant differences were found for this variable according to sex (p = 0.027) ( Table  7). Mentions of the anonymous-collective characters had a level 2 in the frequencies recorded, which means that the characters were mentioned 4 to 6 times in the narratives. There were no significant statistical differences according to classroom-group (p = 0.132) or school (p = 0.377) in this variable (Table 8).

Correlation between variables
The highly correlated variables were V1 (leading men) and V4 (anonymous-collective characters) with a Spearman's correlation coefficient of 0.89. This result indicates that a high presence of leading men is related to a majority discourse with anonymous-collective characters (Table 9).
In relation to quasi-characters, 165 mentions were counted, which is equivalent to 21% of the total number of narrative extracts. It should be noted that all were written using the masculine gender variation of the Spanish language, i.e., indígena (f0 = 60; 8%), español (f0 = 31; 4%) and mapuche (f0 = 27; 3%). It is also important to note that 21 women were included in the narrative extracts, representing 3% of the total. Among these mentions, Michelle Bachelet (f0 = 12; 2%), Inés de Su|rez (f0 = 3; 0.4%) and Gabriela Mistral (f0 = 3; 0.1%) stand out, among others mentioned only once (Table  11). The narratives are developed in three structural areas: political-military, leading men and quasi-characters, which is equivalent to a frequency (f0) of 729 mentions and 93% of the total number of narrative extracts. These values mark a differential contrast with the place occupied by women in the narrative plot: 2% of the total described, a percentage mostly assigned to a prototype of powerful woman with masculine characteristics.

Qualitative analysis of narrative excerpts
In accordance with a traditional historiographical approach, the student C1N2_A2 writes a text chronologically coherent with events and male leaders from different stages of Chilean history: [...] In colonial times, Chile was part of the Viceroyalty of Peru, which was discovered by the Spaniard Diego de Almagro and conquered by Pedro de Valdivia. In 1810, Chile managed to have its first Council of Ministers, so that in 1816 independence was officially declared, after going through multiple civil wars and changes in the constitution in the 19th century [...] Pedro Aguirre Cerda was elected, who made improvements mainly in education. After him, presidents such as Arturo Alessandri, Ramón Frei, Juan Antonio Ríos followed until the notorious coup d'état. This begins with Salvador Allende winning the presidential Also with political-military characteristics and an androcentric focus is the narrative of C1N3_A2 on two different chronological stages: on the one hand, the process of conquest and colonisation and, on the other, the coup d'état in Chile: [...] For the formation of the Chilean State, the country has gone through great processes and changes throughout its history. Initially, the country was inhabited by indigenous people until the arrival of the Spaniards, who took sovereignty over the territory. Then, due to social and economic issues in the country, a group of people, among them Bernardo O'Higgins, decided to fight for independence, like other Latin American countries. Another milestone, in this case in modern Chilean history, was the coup d'état that took place in 1973. After Salvador Allende came to power [...], the military general Augusto Pinochet decided, together with a military junta, to attack the Palacio de la Moneda (Presidential Palace) and regain power by force [...].
Along these lines is the narrative of C2N67_A3: "Chile was a territory 'discovered' by Diego de Almagro, founded and conquered by Pedro de Valdivia, both sent by the Spanish Crown [...]". The abundance of politico-military narratives involving men as protagonists is thus evident.
The quasi-characters are described using the masculine gender variation of the Spanish language: "[...] It was discovered by Ferdinand Magellan. Before its discovery there lived many indigenous tribes, among which the Mapuches stand out for their constant struggle against the Spaniards. The Spanish massacred hundreds of indigenous people [...]" (C2N84_A3). It can be seen that the context and themes, in which these characters are included, are related to politicalmilitary events, giving continuity to traditional historiographical perspectives.
Evidence of this perspective is C2N82_A3's narration of non-recurrent aspects of traditional history and the perception of quasi-characters from a male perspective: "[...] the history of Chile begins with the arrival of the Spaniards. Before this arrival, the Chilean territory was inhabited by indigenous people from Onas to Diguitas, and from Yaganes to Araucanos, among others" (C2N82_A3).
Mentions of women and their history are brief and scarce and, on the whole, are not important in the narratives. Moreover, they are included in political-military narratives, in which, generally, it is the men in power who centralise the historical plots. This shows that the figure of president Bachelet does not have significant relevance in the narratives, in any case, directed at her role as a "powerful woman" in contrast to other government leaders and the aforementioned dictator. Therefore, her inclusion in the narrative is neither due to her relevance as a woman nor to her actions, which could contribute to the struggle for gender equality and social justice.
In this line, despite the brevity of his/her comments, C2N50_A2 declares not to know dates or stages, but bases his/her narration on characters with power, including some women not related to any historical phenomenon or other characters: [...] I don't remember any dates, maybe just names and some facts, like O'Higgins who was the first President of Chile, José Miguel Carrera [...], Manuel Blanco Encalada, Allende, Augusto Pinochet, Gabriela Mistral, who won the Novel Prize, Pablo Neruda, Violeta Parra, Nicanor Plaza, Michelle Bachelet (first woman president) and Sebasti|n Piñera (C2N50_A2).
The range of people mentioned is remarkable. Although reference is made to the figure of Michelle Bachelet in her role as president, her relevance is overshadowed by a long list of men with political and military power. In the same context, Gabriela Mistral is identified as a poet and writer, and Violeta Parra as a singer-songwriter. Although a break with more traditional historiographical accounts is generated, these narratives do not describe or analyse their actions or voices, placing the protagonism of the texts on different powerful male figures.
In this sense, C2N71_A3 describes some political and military events in Chilean history. In his/her narrative, in a disjointed manner and without a clear chronological logic, the poet Gabriela Mistral is mentioned: "[...] Arturo Prat was part of the War of the Pacific. Gabriela Mistral and Pablo Neruda won the Nobel Prize for Literature [...]". Consequently, the absence of a central female role in the construction of historical and cultural knowledge can be identified.
However, there are some narratives that are very interesting in its descriptions and analysis. These narratives reflect on the construction of history, generating questioning and critical thinking about the proposed topic. As an example, we can mention what C2N61_A3 states: "Indigenous people have been marginalised from history. Besides, I only know about political issues, is that all history is about? (C2N61_A3).
In addition, the comments in C2N106_A4 refer to contemporary social and political movements and citizen participation: [...] [After] many political changes and social revolutions and economic crises, such as social issues, middle class movements, the feminist movement, the stock market crash, etc., [...] people want to have a voice in government [...] (C2N106_A4).
The characteristics of these minority narratives mark an important difference and contrast with the rest of the accounts. They are proof that critical thinking is present in some narratives and of the different views on history, its construction and its teaching, as the following text shows: "the Incas, instead of attacking the Diaguita, lived and learned from them. Besides, it is important to know that Pablo Neruda won a prize and raped a woman [...]" (C2N102_A4).

Discussion
It is possible to assert that narratives are dominated by the leadership of men who hold political, military and economic power. When the narrative does not have a male protagonist, the story is collectivised, which means that leadership is given to groups such as "the Spanish", "the indigenous" or "the workers", who do not represent women and their historical experience. Therefore, women as individual protagonists and their possible representation through quasicharacters or proto-structures are excluded from the historical construction (Crocco, 2008(Crocco, , 2010Talburt & Steinberg, 2005). These results confirm the continuity and consolidation of positivist curricular approaches, the technical and linear conception in perceived historical learning, and its implications in the initial training of history and social sciences teachers (Moreira et al., 2019;. Considering Rüsen's (2004) levels of complexity of critical historical consciousness, the narrative discourses and the characters/protagonists selected by the students therefore conform to traditional and/or exemplary narratives, and assume a general deterministic, stereotyped and reproductive character. It confirms, in fact, the permanence of narratives based on a coherent and selective relationship of facts, events and great characters/protagonists of the past (exemplary model), where women and the female experience remain in dependent and subsidiary positions of male leadership or quasi-characters, selected as structuring axes of the stories.
Specifically, the narratives tend to contain details, facts and male characters linked to the nation's victories, to the construction of a homogeneous political identity and to the attachment to traditional national sentiments (also homogeneous). These results confirm that the version of history that is taught and transmitted to pupils is still linked to a traditional positivist approach, without any relation to the social problems of the present . This approach moves away from the development of temporal and empathetic skills, critical thinking and the promotion of social participation (Santisteban, 2010;Seixas & Morton, 2013;Wineburg, 2001). Indeed, the version of the story told by the participants does not consider that, through their learning, empowerment on gendered social issues and participation as global citizens in the fight against inequality is possible ( Barton & Levstik, 2004).
The women mentioned in the narratives do not play a central role. On the contrary, they are often accompanied by men. Therefore, women are included as subordinates in a predominantly male narrative (Britzman, 2002). Following Spivak (2012), Marolla (2019aMarolla ( , 2019b and Ortega-S|nchez and Pagès (2018), women present in school historical narratives function as supports for official male narratives. In this way, these narratives are representations of heteropatriarchal structures that promote the subordination of women, as well as their actions and collective history (Fern|ndez Valencia, 2015).
According to Marolla (2020) and Stanley (2010), the women mentioned in the historical narratives are present because they implicitly embody some masculine characteristic in their actions and expressions (Marolla, 2019a). Arguably, the inclusion of these women does not represent a change in sexist and stereotypical structures in favour of male hegemony.
Along these lines, Sant et al. (2014), Ortega-S|nchez and and Marolla (2019b) state that the perpetuation of classical historiographical traditions in history education causes students to assume that the history of women, as well as that of non-elites, is not relevant in the construction of historical knowledge and its processes. This circumstance contributes to the fact that students do not feel part of history. It can be inferred, therefore, that the history taught is being transmitted by a model of the teacher as a "storyteller" (Evans, 1990;VanSledright, 2011).
The consistency and validity of the design, methodological procedure and instrument applied in this research (pioneering in the American regional space) have been demonstrated in similar narrative analyses from the field of children's literature (Shahnaz et al., 2020), whose results are consistent with those presented in this research. They demonstrate the absence or stereotypical presence of women in the construction of both social-historical and literary narratives. Consequently, reflective teaching practices are necessary (Pagès, 1999), which help to rethink the teaching of history in order to pluralise the construction of personal and social identities, and to promote social diversity. In this sense, it is required that women and their voices in history are not included because of a hegemonic male assimilation or because of their anecdotal or additive appearance, but because of their relevance in the integrated articulation of the historical narrative.

Conclusions
The maintenance of dichotomous gender identities and its permanence in historical school narratives informs the need to "subvert the dominant heteronormative system and to challenge the formation of self-identical subjects towards a consideration of open and necessarily indeterminate identities (...)" (Molet & Bernad, 2015, p. 128). Consequently, the need for teacher training in the social sciences aimed at the critical analysis of curricula, discourses and teaching practices can be affirmed.
This analysis should lead to the challenge of overcoming androcentric and hegemonic structures in the transmitted historical narratives. The goal that should drive teaching practices is to position social justice as an intrinsic value in any educational process. As stated by Crocco (2010Crocco ( , 2018, Vavrus (2009) and Barton (2002), it is essential to question existing structures and to propose teaching processes where women and their history, as well as gender diversity, promote critical thinking and empowerment of students to address social-political challenges around inequalities.
In this sense, Colley (2019) considers that teachers need to be able to: 1) Address students' representations of gender and feminism; 2) Encourage students to identify and describe systems of power; 3) Analyse the impact that gender identities and their understanding of institutionalised oppression have on their historical analyses, in order to work towards and achieve gender equality in the teaching and learning processes in history education.
The resistance of androcentric perspectives in school curricula and textbooks therefore requires working on the narrative skills of students and teachers in training (Ortega-S|nchez & Pagès, 2018), with the aim of highlighting those actors and social spheres traditionally relegated to invisibility. The problematisation of social representations of gender, based on the deconstruction of stereotypes, prejudice and the relations of domination in which they are embedded, requires unveiling the social and historical nature of inequality.
Inequality, recognisable in the deconstruction of specific gender relations, is evident in the attribution of certain feminine behaviours and stereotypes that limit the complexity and autonomous development of identity diversity (Ortega-S|nchez, 2017). In the educational sphere, it is therefore necessary to "counter-socialise" or "counternormalise" identity through critical reflection and social action. From this perspective, we agree with Kumashiro (2000) in stating that "teaching is not a representational act, an unproblematic transmission of knowledge about the world to the student, but is a performative act, constituting reality as it names it" (p. 46).

Recommendations
The inclusion of women and their historical experience has to be accompanied by a discursive change, in order to contribute to the construction of a comprehensive historical knowledge. The purpose should not only be to include women and their history in an educational programme to overcome male models, but also to strengthen critical thinking about social and gender inequalities, promoting the deconstruction of prejudices and stereotypes about nondominant diversities. We agree with Ortega-S|nchez et al., (2020), Marolla (2020), Spivak (2012) and Scott (2008) that women's inclusion must come about through the subversion of the hegemonic and patriarchal structures on which historical education has been based. In this context, students should be provided with tools so that, by applying critical thinking skills, they can understand how history and its protagonists have been constructed, narrated and transmitted.
According to Schmeichel (2015), gender studies and feminist epistemology, and practices could provide a solid theoretical basis for the inclusion of women in the curriculum. As proposed by McIntosh (2005) and Woyshner (2002), the process of historical deconstruction should lead to work on issues such as bodies, sexuality, mentality, daily life and private life, among others. This process should promote historical competences such as empathy, identity, socialpolitical empowerment and civic participation in favour of the fight against social injustice, stereotypes and genderrelated biases (Woyshner & Schocker, 2015). As historiography has already shown, it is essential that the topics that are worked on in history education show students a diversity of themes and areas, which can contribute to generating, in the end, spaces and competences of empathy, participation and empowerment.

Limitations
In the analysis and interpretation of the results obtained in this study, the following limitations should be considered: 1) The cross-sectional nature of the research design, whose space-time delimitation requires its extension, in a comparative manner, to the rest of the countries in the Ibero-American regional area. It is necessary to apply longitudinal designs in which it is possible to implement and evaluate inclusive educational interventions. 2) The school historiographic trends identified are based on the limitations of the non-probabilistic sampling selected (convenience or accidental sampling). Consequently, it would be advisable to consolidate its results by carrying out a probability sampling, capable of reaching representative samples of the student population to which they belong. 3) The narratives written by students constitute a large corpus of self-reported data. This circumstance could lead to biases related to students' selective memory when writing their texts and to social desirability in the argumentative orientation of their narratives.