Brazilianity of System Justification for Gender Relations: Evidence of ValidityABSTRACT
The Gender-Specific System Justification Scale (GSSJS) was adapted to Brazil through two studies. 639 participants (study 1) answered the adapted version of the scale, a sociodemographic questionnaire, and measures to collect evidence of convergent-discriminant and incremental validity, with the GSSJS being administered again to 670 participants (study 2), aiming to replicate the results previously found. The items presented adequate discriminatory power, and better evidence of validity and precision was found for a single-factor structure with six items. The results indicated the replication of the one-factor structure of the original scale, but in a short version. Limitations regarding the content of the GSSJS, considering gender inequality and the Brazilian context, are discussed.
KEYWORDS:
system justification; cross-cultural test adaptation; test validity; test reliability; gender
RESUMO
A Escala de Justificação do Sistema para Relações de Gênero (EJSRG) foi adaptada para o Brasil por meio de dois estudos. 639 participantes (estudo 1) responderam à versão adaptada da EJSRG, a um questionário sociodemográfico e a medidas para coleta de evidências de validade convergente-discriminante e incremental, sendo a EJSRG novamente administrada para 670 participantes (estudo 2), visando replicar os resultados previamente encontrados. Os itens apresentaram adequado poder discriminativo e foram encontradas melhores evidências de validade e precisão para uma estrutura unifatorial com seis itens. Os resultados indicaram a replicação da estrutura unifatorial da escala original, porém em uma versão reduzida. Limitações acerca do conteúdo da EJSRG considerando a desigualdade de gênero e o contexto brasileiro são discutidas.
PALAVRAS-CHAVE:
justificação do sistema; adaptação transcultural de medida; validade da medida; precisão da medida; gênero
Gender inequality reflects the asymmetric power relations between men and women in society (Lima, 2020), resulting in social, political, and economic asymmetries (Adichie, 2017). This form of inequality is also theoretically discussed through the understanding of hierarchical systems based on gender, which are historically upheld through structural patriarchy (Farias et al., 2023; Biermann & Farias, 2021).
More extreme elements associated with gender inequality can be observed across all continents, such as violence against women (Acosta et al., 2018; Dillon et al., 2013). Beyond such explicit manifestations, more subtle everyday situations reflect this disparity, indicating the acceptance and naturalization of this form of inequality. These manifestations occur in various global contexts and encompass areas such as education, health and survival, labor market, division of roles, empowerment and political representation, and economic participation and access to opportunities (World Economic Forum [WEF], 2024).
Based on these areas, a report by the World Economic Forum (2024) highlighted that, considering the pace of progress toward gender equality, Latin America shows improvement in the path toward parity. However, it is emphasized that only 74.2% of gender parity has been achieved. Specifically regarding Brazil, the country improved its position compared to 2022, moving from 94th to 70th place in a ranking of 146 countries on gender equality (WEF, 2024).
Additionally, a study conducted in the United States by Lean In Organization and McKinsey & Company (2024) indicates that women are underrepresented across different sectors of the labor market when it comes to leadership positions, with significantly lower advancement in corporate environments compared to men, regardless of race. In Brazil, according to the transparency report panel from the Ministry of Labor and Employment and the Ministry of Women (2024), wage disparities are present both at entry-level positions and in higher-ranking roles. Black women are paid less than Black men, and this same pattern is reproduced between women and men more broadly. Furthermore, in more hostile expressions of gender inequality, domestic violence increased in 2023 compared to the previous year, with nearly a 10% rise in reported assaults (Brazilian Public Security Directory, 2024). These data reflect the stable process of naturalizing gender inequality in the country, highlighting the need to draw greater attention to this phenomenon in the Brazilian context.
In this sense, it is important to note that understanding the social world and its interactions does not occur in a neutral or impartial manner, but is aligned with personal motivations (Jost et al., 2003). Therefore, this acceptance and naturalization reflect social and psychological processes that lead individuals to perceive a set of attitudes, behaviors, and social structures as fair and aligned with the normative standards of society (Costa‐Lopes et al., 2013). It is thus essential to better understand how these processes contribute to the maintenance of this form of inequality in Brazil.
System Justification Theory
In light of the above, System Justification Theory (Jost & Banaji, 1994) emerges as a relevant theoretical framework. This theory addresses the motivational dimension associated with individuals’ tendency to defend and justify the societal status quo (Jost, 2020), representing a psychological process through which social arrangements and systems are legitimized (Jost & Banaji, 1994). Such justification operates as an underlying mechanism of institutional legitimization of an unequal system, endorsed by individuals from both high- and low-status groups (Jost & Hamilton, 2008). This defense leads the status quo to be rationalized and legitimized as fair, natural, and orderly (Gatica et al., 2017; Jost, 2020), which is directly linked to the perpetuation of inequalities.
This system-justifying process serves as a defense mechanism that minimizes potential distress arising from cognitive dissonance and contributes to individuals’ conformity to this context (Jost & Hamilton, 2008). Specifically concerning marginalized groups, unlike those belonging to advantaged social groups who benefit directly from the status quo, disadvantaged individuals may feel the need to minimize the discomfort of living in a harmful context. As a result, they may become inclined to defend social systems-even unequal ones-as a way to reduce cognitive dissonance (Jost et al., 2003).
In regard to gender issues, Gender-Specific System Justification can be understood as a system justification theory for gender relations. This refers to the process of interpreting beliefs about existing gender relations as fair and natural, thereby justifying the system in which these relations are embedded. In this way, the status quo, though unequal, is perceived as legitimate (Jost & Kay, 2005). Moreover, this system justification may be further reinforced by sexist beliefs that support the acceptance of gender differences as legitimate (Connelly & Heesacker, 2012).
Measuring System Justification and Gender Relations
In line with the theoretical framework of System Justification Theory (Jost & Banaji, 1994), various instruments have been used to assess this phenomenon, resulting in a sometimes heterogeneous yet theoretically coherent body of research. The General System Justification Scale (Kay & Jost, 2003) has been used to measure this construct in the context of gender inequality, though often in combination with other instruments that directly address gender issues, such as the Ambivalent Sexism Inventory (Prusaczyk & Hodson, 2020; Connelly & Heesacker, 2012).
Different tools are employed to evaluate system justification in research on gender inequality, even when not directly operationalized from the theory or when drawing on related constructs such as neoliberal beliefs and belief in a just world (Fitz et al., 2012), among others (Chapleau & Oswald, 2014; Girerd & Bonnot, 2020; Chayinska et al., 2021). Despite these examples, there is greater parsimony and precision in using the Gender-Specific System Justification Scale (GSSJS) (Jost & Kay, 2005) compared to using multiple unrelated measures. Studies that have specifically employed the GSSJS or variations thereof often did not associate them with other measures addressing sexist beliefs, gender roles, stereotypes, or gender discrimination (Iatridis & Stergiou, 2016; Kosakowska-Berezecka et al., 2016; Kray et al., 2017; Saguy & Szekeres, 2018).
Therefore, the complexity of investigating gender-based system justification is best captured when studies employ the GSSJS (Jost & Kay, 2005). However, despite its widespread use in international research, this instrument has not yet been adapted for the Brazilian context, highlighting the need to make this tool available for a better understanding of the phenomenon in Brazil and to reduce interpretative gaps between the vast empirical knowledge produced in different cultural settings using this scale. Furthermore, as previously illustrated by the statistical data, the Brazilian context does not mirror that of the United States, presenting not only epidemiological but also cultural differences. Thus, adapting the scale to Brazil also entails a critical evaluation of the instrument’s adequacy in its current form.
In this regard, considering the pervasive presence of gender inequality today and its processes of acceptance and naturalization, it becomes essential to have appropriate tools to understand how this legitimization process is structured and how it may contribute to the maintenance of such inequality in the Brazilian context. Therefore, the present study aimed to adapt the GSSJS for use in Brazil and to gather evidence of the measure’s validity and reliability. Initially, procedures for translation and cultural adaptation to Brazil were conducted, followed by an investigation of the scale’s factorial structure through exploratory and confirmatory analysis. To this end, the study examined the GSSJS’s associations with other measures that assess theoretically related or unrelated constructs (convergent-discriminant validity), as well as its potential for incremental validity when controlling for other relevant constructs in relation to a theoretically significant phenomenon. Finally, Study 2 aimed to compare the best-fitting factorial structure identified in Study 1.
Study 1 - Exploratory Factor Analysis of the Gender-Specific System Justification Scale in the Brazilian Context
This study was conducted in order to adapt the GSSJS for the Brazilian context. In addition, it sought to investigate the most appropriate factorial structure for the measure in an exploratory manner and collect evidence of validity and reliability.
Method
Sample
The study included 639 participants aged between 18 and 79 years (M = 31.5; SD = 10.9), of whom 75% were women. Among them, 41.2% were single and 31.3% were married or in a stable union, 80.2% were heterosexual, and 47.5% self-identified as White, while 42.5% self-identified as Mixed-race (i.e., Pardo). Additional sample characteristics stood out, 45.3% of the sample identified as Catholic, 40.8% as middle class, 20.2% as lower-middle class, and 27.8% as upper-middle class. Moreover, 63.9% of participants reported having completed an undergraduate degree. Specifically in relation to political orientation, the majority identified as left-wing (27%) or centrist (22.8%). Finally, regarding the geographic distribution of respondents, 48.3% lived in Ceará and 33.6% in Piauí, while the remainder of the sample was distributed across 18 other Brazilian Federative Units from all five regions. For better visualization, see Table 1 with the summary of sociodemographic information for Study 1.
Instruments
The people who agreed to participate were asked to respond to a questionnaire with different measures and a sociodemographic questionnaire. Data collection was conducted as part of a larger study with data from various instruments collected in the second half of 2020; however, the present study used the following instruments:
Gender-Specific System Justification Scale (GSSJS) - originally created by Jost and Kay (2005), based on the general system justification instrument developed by Kay and Jost (2003), but directed specifically at gender inequality. The scale is operationalized with eight items focusing on gender relations, for example: “In general, relations between men and women are fair.” Finally, it is noteworthy that items are answered using a nine-point Likert scale, ranging from 1 (Strongly disagree) to 9 (Strongly agree). Following adaptation guidelines (Pasquali, 2010), this version was translated independently from English to Brazilian-Portuguese by three PhD researchers in social psychology. The three translations were then evaluated in terms of content and merged into a single version, which was then backtranslated from Brazilian-Portuguese to English. This back-translated version was compared with the original version of the scale to ensure the original proposed meaning of the items, but in a way appropriate for the Brazilian context. Finally, semantic fidelity to the ideas expressed in the original items was maintained. Ultimately, the final Brazilian-Portuguese version was consolidated, ensuring that both the items and instructions were comprehensible and suitable for the data collection context.
Short Version of the Ambivalent Sexism Inventory (ASI) - originally developed by Glick and Fiske (1996) and validated for the Brazilian context by Formiga, Gouveia, and Santos (2002) in its long version, and with the short version validated for Brazil by Souza and colleagues (2024, in preparation). The measure comprises 12 items, equally divided into two factors analogous to the two theoretical dimensions: Hostile Sexism (α = 0.86; including items such as “Feminists make unreasonable demands of men”) and Benevolent Sexism (α = 0.81; including items such as “Women should be cherished and protected by men”). The response scale is a five-point Likert scale with endpoints: 1 = Strongly Disagree and 5 = Strongly Agree.
Effort Subscale of the Effort-Reward Imbalance in Domestic and Family Work Scale - originally developed by Siegrist et al. (2009) and later transculturally validated for Brazil by Vasconcellos and colleagues (2016). The full measure presents three dimensions: Effort, Reward, and Overcommitment. Specifically in the present data collection, only the Effort subscale was used (α = 0.93) to assess Domestic Overload (DO), which comprises eight items focusing on quantitative workload, containing statements such as “Many times I feel like I have to do ‘a thousand things’ at the same time in domestic and family work.” The response scale is a five-point Likert scale with endpoints: 1 = Strongly Disagree and 5 = Strongly Agree.
Satisfaction With Life Scale (SWLS) - originally proposed by Pavot and Diener (1993) and having shown good psychometric adequacy in the Brazilian context (Gouveia et al., 2003). The measure comprises five items in a unidimensional structure (α = 0.86), including items such as “In most ways my life is close to my ideal.” The instrument is answered using a Likert scale ranging from 1 (Strongly Disagree) to 7 (Strongly Agree).
Ethical and Data Collection Procedures
Data were collected via an online questionnaire hosted on the SurveyMonkey platform and disseminated through the general population's social media, such as WhatsApp and Instagram. Participants were informed that their participation was voluntary and anonymous and were presented with the Free and Informed Consent Form (FIC) and were informed that participation was conditional upon consent. The FIC was made available, and those who selected “Yes, I agree to participate in the study” were asked to complete it.
Data collection followed Brazilian National Health Council Resolutions No. 466/2012 and No. 510/2016 regarding human research, as well as Circular Letter No. 2/2021/CONEP/SECNS/MS. Additionally, data collection was approved by the Research Ethics Committee of the University of Fortaleza (Approval No. 4.088.481).
Data Analysis
Data were analyzed via Jamovi version 2.3. Descriptive and central tendency statistics, exploratory factor analysis (EFA), internal consistency analysis, and composite reliability (Raykov, 1997) were conducted. For factor extraction criteria, eigenvalues greater than 1 (Kaiser), scree plot (Cattell), and parallel analysis (Horn) were used. Bartlett’s test of sphericity and the Kaiser-Meyer-Olkin (KMO) index (≥ .70) were used to evaluate data adequacy (Hair et al., 2019). Additionally, a MANOVA and Mann-Whitney tests were conducted to assess item discriminative power. Finally, Pearson’s r correlations and multiple linear regression analyses were conducted to gather validity evidence.
Results
Initially, the data matrix was examined to evaluate its suitability for factor analysis based on KMO (0.80) and Bartlett’s test [χ²(28) = 1326; p < 0.001]. Complementarily, item discriminative power was assessed. The sample was split into low (< 3.75) or high (> 3.75) gender-specific system justification groups according to the median. A MANOVA revealed that the items demonstrated satisfactory discriminative power according to Pillai’s trace (0.594) [F(8, 631) = 115; p < 0.001]. Mann-Whitney tests were then conducted (Table 2), identifying significant differences between items for the criterion groups. However, two of the eight items presented significant differences at p < .01, but marginally higher than the others (p < .001). This significant difference may indicate greater difficulty in detecting relevant differences, representing fewer sensitive items, resulting in less pronounced differences compared to the other items (Hair et al., 2019).
Moreover, based on point-biserial correlations to identify effect size from the Mann-Whitney test, moderate to strong effect sizes were found (rpb = 0.46 to 0.57), except for the aforementioned two items (0.12 and 0.13), a result considered during exploratory analysis to determine whether these items should indeed be retained or removed.
Given matrix adequacy and significant discriminative power, an exploratory factor analysis was conducted using maximum likelihood extraction without rotation or fixing the number of factors to extract. The analysis indicated the possibility of extracting one factor based on eigenvalues > 1 (Kaiser), four factors based on the scree plot (Cattell), and three factors based on parallel analysis (Horn).
However, items loaded on different factors in the multifactor model, with a nonsignificant model test (p = 0.113). Given that the original scale presupposes a unidimensional structure (per Kaiser), it was decided to conduct another exploratory factor analysis. The same extraction method was used, this time with oblimin rotation and fixing just one factor, as suggested by the eigenvalues. Results indicated that two items exhibited factor loadings below 0.30 and that Cronbach’s alpha for the scale increased when these items were excluded (Table 2).
The new analysis conducted with the remaining six items demonstrated satisfactory factor loadings and evidence of internal validity, as well as the appropriateness of a unidimensional six-item structure, RMSEA = 0.08; 95% CI [0.06; 0.11]; TLI = 0.95; [χ²(9) = 48; p < 0.001]. Additionally, the six-item structure accounted for 44.7% of the explained variance (+10.2% increment compared to the eight-item factorial structure), and Cronbach’s alpha was 0.82 with a composite reliability of 0.83, indicating better fit compared to the original structure. Although RMSEA is at the upper limit of acceptability, taken together with the suitable TLI (≥ 0.95) and the understanding that simpler models with fewer items may inflate RMSEA values (Hu & Bentler, 1999), the indicators are considered satisfactory.
Pearson’s r correlations were also carried out to assess item-total and item-item associations. Items correlated positively and significantly (p < 0.001) with the total scale score, ranging from coefficients of 0.62 to 0.79. All items also showed significant positive correlations among themselves (Figure 1).
Regarding convergent-discriminant validity evidence, it is noteworthy that the scale showed associations with different constructs in the expected directions and magnitudes (Table 3). The results revealed significant and moderate positive correlations with Benevolent Sexism (r = 0.50, p < 0.001) and Hostile Sexism (r = 0.63, p < 0.001), indicating an association with the theme of gender relations, without, however, representing overlap between measures and constructs. Thus, the analysis identified convergent validity evidence between the new version of the GSSJS and both dimensions of Sexism, Hostile and Benevolent.
Additionally, in relation to discriminant validity, a significant positive association was found between the GSSJS-6 and Life Satisfaction (r = 0.26, p < 0.001), and a significant negative association with Domestic Overload (r = −0.19, p < 0.001). Given these results, the low correlations of the GSSJS-6 with these two constructs provided evidence of discriminant validity, indicating that distinct phenomena are operationalized.
Finally, incremental validity evidence of the GSSJS-6 as a predictor of Life Satisfaction was analyzed. When considering participants’ Domestic Overload, political orientation, and gender, 2.94% of the variance in Life Satisfaction was explained. It is worth noting that this relationship was considered, taking into account participants’ levels of Benevolent Sexism, given its relevance in the literature on gender relations. However, when the GSSJS-6 was added to the model, a significant increment was observed, accounting for 10.32% of the variance. Additionally, it is noteworthy that the inclusion of this variable rendered the other predictors nonsignificant while the GSSJS-6 remained significant even when controlling for the other predictors (Table 4). Thus, it can be considered that the new version of the scale has a high incremental explanatory potential, even when including constructs with low to moderate associations with the outcome of interest.
Results for convergent validity support previous evidence indicating that Hostile Sexism, through overt prejudice, and Benevolent Sexism, more associated with paternalistic gender-based roles, directly contribute to constructions related to the maintenance of sexist ideologies (Bareket & Fiske, 2023). Moreover, the positive association of the measure with life satisfaction also aligns with previous findings referring to the role of system justification in reducing dissonances that can cause discomfort (Jost et al., 2003). These results provide empirical evidence of the consistency and reliability of the adapted scale for measuring individual differences in gender-specific system justification.
Study 2 - Analysis of the System Justification Scale for Gender Relations in the Brazilian Context: A Replication of the Unidimensional Structure
This study aimed to investigate the adequacy of the previously established unidimensional structure to replicate the results. The objective was to compare the best-fitting factor structure found in Study 1, composed of six items, with the original unidimensional structure composed of eight items.
Method
Sample
The sample consisted of 670 participants aged between 18 and 81 years (M = 32.89; SD = 9.91), of whom 77.6% were women. Regarding marital status, 42.4% were married or in a stable union, and 32.2% were single. Furthermore, 83.4% identified as heterosexual, 43.3% self-identified as Mixed-race (Pardo), and 46.7% as white. Other characteristics stood out: 45.5% identified as Catholic, 45.2% as middle class, and 47.5% as lower-middle class. In terms of education, 41.5% of the participants reported having completed postgraduate studies (Master’s/Ph.D.), and 27.3% had completed an undergraduate degree. Politically, most of the sample identified as left-wing (33.1%), but there was good distribution among center-left (20.7%), center (14.9%), center-right (9.6%), and right-wing (16.4%). For further details, see Table 5 for a summary of the sociodemographic variables.
Instruments
Participants who agreed to participate were asked to respond to a questionnaire containing the adapted version of the Gender-Specific System Justification Scale (GSSJS). The full eight items from the original version were collected, given the objective of this study to compare item compositions based on the originally proposed factorial structure and the one identified in the exploratory analysis of Study 1. Additionally, participants completed a sociodemographic questionnaire.
It is important to highlight that the Gender-Specific System Justification Scale (Jost & Kay, 2005; Kay & Jost, 2003) is rated on a 9-point Likert scale ranging from 1 (Strongly Disagree) to 9 (Strongly Agree). Items 3 and 7 were reverse-coded during analysis.
Ethical and Data Collection Procedures
The same ethical and data collection procedures described in Study 1 were followed for the present study. The ethical assumptions that guided this study were based on Resolutions No. 466/2012 and No. 510/2016 of the Brazilian National Health Council, which establish ethical guidelines for research involving human beings, as well as on Circular Letter No. 2/2021/CONEP/SECNS/MS of the National Research Ethics Committee. Furthermore, the study was approved by the Research Ethics Committee of the University of Fortaleza (Approval No. 5.298.110).
Data Analysis
Structural Equation Modeling (SEM) was conducted using the Brazilian version of the GSSJS to evaluate the unidimensional structure of the scale and test the possible replication of the results from the previous study, which found a better fit for the six-item version. The estimation method used was Robust Maximum Likelihood.
Initially, model fit was evaluated using the chi-square test and the chi-square/degrees of freedom ratio (χ²/df), with values between 1 and 3 considered satisfactory (Damásio, 2013; Nascimento et al., 2016). Additional fit indices included the Comparative Fit Index (CFI), Goodness of Fit Index (GFI), and Tucker-Lewis Index (TLI), with values close to 1 considered more adequate (Tabachnick & Fidell, 2007; Marôco, 2010; Byrne, 2010). Root Mean Square Error of Approximation (RMSEA) and Standardized Root Mean Square Residual (SRMR) were also used to evaluate model fit, with values closer to 0 considered more acceptable (Nascimento et al., 2016).
Factor loadings were assessed using standardized Beta values and their respective confidence intervals. Item communality was evaluated using R² values. Internal consistency was assessed using Cronbach’s alpha, McDonald’s omega, and composite reliability (Raykov, 1997). The unidimensional structure with the original 8 items and the 6-item structure identified in Study 1 were tested. Data were analyzed using Jamovi version 2.3, and descriptive and central tendency statistics were computed to characterize the sample.
Results
To evaluate the adequacy of the previously proposed unidimensional structure, SEM was conducted comparing the unidimensional structure identified in Study 1 (6 items) with the original 8-item structure.
Model 1 (SEM 1) specified a single factor with the eight original items (GSSJS-8). Results indicated: χ² (20) = 116.2; p < 0.001; χ²/df = 5.81; TLI = 0.87; CFI = 0.91; GFI = 0.96; RMSEA = 0.083; 95% CI [0.07; 0.10]; SRMR = 0.054. Factor loadings ranged from 0.14 to 0.72. Internal consistency was acceptable: Cronbach’s alpha = 0.71, McDonald’s omega = 0.72, and composite reliability = 0.74 (Table 6).
Overall, the structure showed acceptable model fit; however, the χ²/df ratio exceeded the acceptable range, suggesting caution in interpreting the model fit. Additionally, items 5 (0.14) and 7 (0.26) had very low factor loadings, replicating findings from Study 1 that identified these items as inadequate for the model. Given these results, although the unidimensional model showed generally acceptable fit indices (except for the χ²/df ratio), we opted to test the unidimensional structure again, excluding items 5 and 7.
Model 2 (SEM 2) specified a unidimensional structure with 6 items (GSSJS-6), as found in Study 1. Results were: χ² (9) = 54.3; p < 0.001; χ²/df = 6.03; TLI = 0.92; CFI = 0.95; GFI = 0.97; RMSEA = 0.083; 95% CI [0.06; 0.11]; SRMR = 0.042. Factor loadings ranged from 0.44 to 0.72 (Table 4).
Internal consistency indices for this model were: Cronbach’s alpha = 0.75, McDonald’s omega = 0.76, and composite reliability = 0.77. While the RMSEA remained stable between the two measurement models, the TLI, CFI, GFI, and SRMR values all showed better fit and adequacy for the 6-item unidimensional structure. Although the chi-square test remained significant, the χ²/df ratio again indicated some degree of misfit between the theoretical model and the data. Nonetheless, internal consistency indicators were stronger than in the 8-item version, and factor loadings were within an acceptable range.
Discussion
The results enabled the preliminary collection of evidence regarding the validity and reliability of the unidimensional factor structure of the reduced version for the Brazilian context. The analyses indicated that two items from the original scale structure did not demonstrate discriminative power regarding the core focus of the phenomenon operationalized by the instrument, namely items 5 (“Most policies related to gender issues and the sexual division of labor serve a greater good”) and 7 (“Discrimination against women in society is getting worse every year”). It is noteworthy that the content presented in these specific items, although performing well in their originally proposed cultural context-the United States-may not be well understood or may not adequately capture the phenomenon within the Brazilian context. Specifically, regarding item 5, two distinct ideas are included, making it difficult to interpret which information should be the focus of the participants’ responses. Additionally, the expression “a greater good” may not have been properly understood due to its abstract nature. As for item 7, there may be difficulty in evaluating this item, as participants may lack sufficient knowledge on the topic to feel comfortable expressing an opinion. Finally, although both items broadly address gender issues, they may lack optimal semantic alignment with the Brazilian context, as well as more direct references to the specific dimension of gender. For example, they mention discrimination but do not specify its type, and they refer to gender issues without naming them directly. Furthermore, the Brazilian context is not equivalent to that of the United States regarding gender inequality, showing not only epidemiological but also cultural and structural differences in the extent of such inequality (WEF, 2024), which may influence participants’ understanding when these issues are not presented more directly. Given this inadequacy and the fact that these items consistently failed to perform adequately across different analyses in both studies, they were excluded from the final version of the scale to be made available for the Brazilian context.
Additionally, the unidimensional structure composed of six items in the new version of the GSSJS demonstrated adequate psychometric parameters for its update. It is worth noting that the parameters identified are consistent with what was observed in the original U.S. version (Jost & Kay, 2005), with even better internal consistency compared to the American version (α = 0.65).
Finally, the identification of convergent and discriminant validity evidence indicated that the scale aligns with the theory of system justification as applied to gender relations, while not overlapping with other highly relevant constructs in this field, such as Hostile and Benevolent Sexism (Glick & Fiske, 1996; Bareket & Fiske, 2023).
Regarding incremental validity evidence of the GSSJS-6, even when considering variables that were not strongly associated with the scale, it was still possible to observe a significant explanatory potential of this variable compared to others. Despite these promising initial results, it is considered that the replication of the identified factorial structure contributed additional validity evidence to the scale in a new study, providing further support for the use of the GSSJS-6.
The results provided support for evidence of validity for the measure, replicating the factorial structure obtained in the original study of the instrument, albeit in a reduced version. However, it is crucial to acknowledge that the GSSJS may present some limitations in the Brazilian context regarding the comprehension of its content. The item reduction had a positive impact on the scale; however, the remaining items may not adequately capture relevant nuances in attitudes legitimizing this form of inequality.
The objectives were considered to have been satisfactorily achieved, providing a psychometric scale preliminarily adequate for use in the Brazilian context. Nevertheless, considering the inadequacy of some fit and data adequacy indices, it is critically understood that the scale content may not be fully aligned with Brazil’s cultural and social complexities, or with the structural factors underlying gender issues. Therefore, the interpretation of results when using the scale, with the composition and structure presented here, must be conducted with caution.
Finally, it is understood that making this scale available facilitates comparative studies on this construct internationally, allowing for more specific and accurate investigations of the processes involved in legitimizing gender inequality in Brazil through a culturally contextualized measure. Additionally, the scale enables empirical studies on gender inequality and its justification in Brazil, complementing research in this area that has thus far relied more heavily on the Ambivalent Sexism Inventory.
Regarding theoretical contributions, the validation of this measure strengthens the study of system justification in the country, demonstrating the adequacy of the operationalization of this phenomenon and enabling an initial step in discussions about the generalizability of studies focusing on this construct in the Brazilian context. In line with previous findings in the literature, during the process of adaptation and collection of validity evidence, the understanding of this construct was reinforced through local specificities, such as the association between system justification and life satisfaction (Napier et al., 2010), and its convergence with sexist beliefs (Chayinska et al., 2021). Furthermore, by facilitating cross-cultural comparisons, the adaptation of the scale to the Brazilian context contributes to identifying convergent and divergent evidence across countries, enhancing both theoretical and empirical understanding of the maintenance of gender inequality and its legitimation process. For the theoretical framework of system justification, considering that different sociocultural contexts may influence system justification differently (Napier et al., 2010), the availability of instruments sensitive to these particularities directly influences the accumulation of evidence in the field. This contribution broadens the theory by demonstrating the application of the concept adapted to a context different from that of the United States, minimizing misinterpretations in the literature when applied to Brazil.
It is important to highlight, however, that despite the overall adequacy of the scale's structure, some model fit indices with unsatisfactory values underscore the need for review and refinement of its content. It becomes necessary to evaluate the items by incorporating both theoretical and contextual elements specific to the country, as well as more targeted gender-related content, in order to ensure better precision indices and a more robust analysis of the legitimation processes of gender relations in Brazil.
Moreover, some limitations were identified in the present study, such as the absence of a Content Validity Index during the process of adapting the scale’s items for preliminary suitability prior to data collection. Additionally, no comparison between distinct samples was carried out in Study 2, which could have facilitated a comparison of fit and adequacy indices to assess whether the inadequacy of certain parameters was an artifact due to sample characteristics or related to the structure of the measure. Lastly, both studies included predominantly female samples, which should be considered when interpreting the results. Specifically concerning incremental validity, although studies have pointed to the positive association between life satisfaction and system justification (Napier et al., 2010), future studies should aim to replicate the incremental validity of the adapted scale with more evenly distributed samples in terms of gender or with samples predominantly composed of men to assess the stability of the measure.
Nonetheless, these limitations are not considered to have compromised the achievement of the study’s proposed objectives, as relevant information about the scale in its format and content, as made available in other countries, was provided. Moreover, the variability in sociodemographic characteristics between the two studies in this research helped to mitigate the potential negative impact of using a single sample in the confirmatory factor structure study.
For future directions, it is recommended to develop items encompassing different spheres in which gender inequality is manifested within the context of gender relations in Brazil, to propose an expanded version of the scale focused on justifying this prevailing system. Additionally, it is suggested that the new scale be developed transculturally between Brazil and other countries to align the measurement of phenomena across different contexts. Finally, it is recommended that an expanded version of the scale seek to carefully avoid overlapping with elements derived from sexism and gender role measures, thus becoming an increasingly relevant and suitable tool for addressing the complex phenomenon of gender inequality, its legitimation, and possible ramifications.
References
-
Acosta, D. F.; Gomes, V. L. D. O.; Oliveira, D. C. D.; Marques, S. C.; Fonseca, A. D. D. (2018). Representações sociais de enfermeiras acerca da violência doméstica contra a mulher: estudo com abordagem estrutural. Revista Gaúcha de Enfermagem, 39. https://doi.org/10.1590/1983-1447.2018.61308
» https://doi.org/10.1590/1983-1447.2018.61308 - Adichie, C. N. (2014). Sejamos todos feministas. Editora Companhia das Letras.
-
Bareket, O., & Fiske, S. T. (2023). A systematic review of the ambivalent sexism literature: Hostile sexism protects men’s power; benevolent sexism guards traditional gender roles. Psychological bulletin, 149, 637-698. https://doi.org/10.1037/bul0000400
» https://doi.org/10.1037/bul0000400 -
Biermann, M.C., Farias, M.G. (2021). Patriarchy and Feminist Perspectives. In: Shackelford, T.K., Weekes-Shackelford, V.A. (eds) Encyclopedia of Evolutionary Psychological Science. Springer, Cham. https://doi.org/10.1007/978-3-319-19650-3_1823
» https://doi.org/10.1007/978-3-319-19650-3_1823 - Byrne, B. M. (2010). Structural equation modeling with AMOS: basic concepts, applications, and programming. Routledge, Taylor and Francis Group.
-
Chapleau, K. M., & Oswald, D. L. (2014). A system justification view of sexual violence: Legitimizing gender inequality and reduced moral outrage are connected to greater rape myth acceptance. Journal of Trauma & Dissociation, 15(2), 204-218. https://doi.org/10.1080/15299732.2014.867573
» https://doi.org/10.1080/15299732.2014.867573 -
Chayinska, M., Uluğ, Ö. M., Solak, N., Kanık, B., & Çuvaş, B. (2021). Obstacles to Birth Surname Retention Upon Marriage: How Do Hostile Sexism and System Justification Predict Support for Marital Surname Change Among Women?. Frontiers in Psychology, 12, 702553. https://doi.org/10.3389/fpsyg.2021.702553
» https://doi.org/10.3389/fpsyg.2021.702553 -
National Research Ethics Committee. (2021). OFÍCIO CIRCULAR Nº 2/2021/CONEP/SECNS/MS. Recuperado em 10 de dezembro de 2021, de Recuperado em 10 de dezembro de 2021, de http://conselho.saude.gov.br/images/Oficio_Circular_2_24fev2021.pdf
» http://conselho.saude.gov.br/images/Oficio_Circular_2_24fev2021.pdf -
Connelly, K., & Heesacker, M. (2012). Why is benevolent sexism appealing? Associations with system justification and life satisfaction. Psychology of Women Quarterly, 36(4), 432-443. https://doi.org/10.1177/0361684312456369
» https://doi.org/10.1177/0361684312456369 -
Brazilian National Health Council. (2012). Resolução nº 466, de 12 de dezembro de 2012. Recuperado em 10 de dezembro de 2021, de Recuperado em 10 de dezembro de 2021, de https://conselho.saude.gov.br/resolucoes/2012/Reso466.pdf
» https://conselho.saude.gov.br/resolucoes/2012/Reso466.pdf -
Brazilian National Health Council. (2016). Resolução nº 510, de 07 de abril de 2016. Recuperado em 10 de dezembro de 2021, de Recuperado em 10 de dezembro de 2021, de https://conselho.saude.gov.br/resolucoes/2016/Reso510.pdf
» https://conselho.saude.gov.br/resolucoes/2016/Reso510.pdf -
Costa‐Lopes, R., Dovidio, J. F., Pereira, C. R., & Jost, J. T. (2013). Social psychological perspectives on the legitimation of social inequality: Past, present, and future. European Journal of Social Psychology, 43(4), 229-237. https://doi.org/10.1002/ejsp.1966
» https://doi.org/10.1002/ejsp.1966 -
Damásio, B. F. (2013). Contribuições da Análise Fatorial Confirmatória Multigrupo (AFCMG) na avaliação de invariância de instrumentos psicométricos. Psico USF, 18(2), 211-220. https://doi.org/10.1590/s1413-82712013000200005
» https://doi.org/10.1590/s1413-82712013000200005 -
Dillon, G.; Hussain, R.; Loxton, D.; Rahman, S. (2013). Mental and physical health and intimate partner violence against women: A review of the literature. International journal of family medicine, 2013. https://doi.org/10.1155/2013/313909
» https://doi.org/10.1155/2013/313909 -
Farias, M.G., Biermann, M.C., de Melo Maia, L.F., de Oliveira Meneses, G. (2023). Structural Patriarchy and Male Dominance Hierarchies. In: Shackelford, T.K. (eds) Encyclopedia of Domestic Violence. Springer, Cham. https://doi.org/10.1007/978-3-030-85493-5_2152-1
» https://doi.org/10.1007/978-3-030-85493-5_2152-1 -
Formiga, N. S., Golveia, V. V., & Santos, M. N. D. (2002). Inventário de sexismo ambivalente: sua adaptação e relação com o gênero. Psicologia em estudo, 7(1), 103-111. https://doi.org/10.1590/s1413-73722002000100013
» https://doi.org/10.1590/s1413-73722002000100013 - Brazilian Public Security Directory. (2024). Anuário Brasileiro de Segurança Pública 2024 (Ano 18). São Paulo: Fórum Brasileiro de Segurança Pública. ISSN 1983-7364.
-
World Economic Forum. (2022). Global Gender Gap Report 2022 - Insight Report. Recuperado em 17 de dezembro de 2023, de Recuperado em 17 de dezembro de 2023, de https://www3.weforum.org/docs/WEF_GGGR_2022.pdf
» https://www3.weforum.org/docs/WEF_GGGR_2022.pdf -
World Economic Forum. (2024). Global Gender Gap Report 2024 - Insight Report. Recuperado em 09 de dezembro de 2024, de Recuperado em 09 de dezembro de 2024, de https://www3.weforum.org/docs/WEF_GGGR_2024.pdf
» https://www3.weforum.org/docs/WEF_GGGR_2024.pdf -
Gatica, L., Martini, J. P., Dreizik, M., & Imhoff, D. (2017). Psychosocial and psycho-political predictors of social inequality justification. Revista de Psicología, 35(1), 279-310. https://doi.org/10.18800/psico.201701.010
» https://doi.org/10.18800/psico.201701.010 -
Girerd, L., & Bonnot, V. (2020). Neoliberalism: An ideological barrier to feminist identification and collective action. Social Justice Research, 33(1), 81-109. https://doi.org/10.1007/s11211-020-00347-8
» https://doi.org/10.1007/s11211-020-00347-8 -
Glick, P., & Fiske, S. T. (1996). The ambivalent sexism inventory: Differentiating hostile and benevolent sexism. Journal of personality and social psychology, 70(3), 491. https://doi.org/10.1037/0022-3514.70.3.491
» https://doi.org/10.1037/0022-3514.70.3.491 -
Gouveia, V. V., Chaves, S. S. D. S., Oliveira, I. C. P. D., Dias, M. R., Gouveia, R. S., & Andrade, P. R. D. (2003). A utilização do QSG-12 na população geral: estudo de sua validade de construto. Psicologia: teoria e pesquisa, 19, 241-248. https://doi.org/10.1590/s0102-37722003000300006
» https://doi.org/10.1590/s0102-37722003000300006 - Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2019). Multivariate Data Analysis (8th ed.). Cengage Learning.
-
Hu, L.T. and Bentler, P.M. (1999). Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus New Alternatives. Structural Equation Modeling, 6 (1), 1-55. https://doi.org/10.1080/10705519909540118
» https://doi.org/10.1080/10705519909540118 -
Iatridis, T., & Stergiou, N. (2016). Complementary gender stereotypes and system justification: the moderating role of essentialist lay theories for group differences. Revista de Psicología Social, 31(1), 29-58. https://doi.org/10.1080/02134748.2015.1101312
» https://doi.org/10.1080/02134748.2015.1101312 - Jost, J. T. (2020). A Theory of System Justification. Harvard University Press.
-
Jost, J. T., & Banaji, M. R. (1994). The role of stereotyping in system‐justification and the production of false consciousness. British journal of social psychology, 33(1), 1-27. https://doi.org/10.1111/j.2044-8309.1994.tb01008.x
» https://doi.org/10.1111/j.2044-8309.1994.tb01008.x - Jost, J. T., & Hamilton, D. L. (2008). Stereotypes in our culture. In Dovidio, J. F., Glick, P., & Rudman, L. A. (Eds.). (2008). On the nature of prejudice: Fifty years after Allport. John Wiley & Sons. 208-224.
-
Jost, J. T., & Kay, A. C. (2005). Exposure to benevolent sexism and complementary gender stereotypes: consequences for specific and diffuse forms of system justification. Journal of personality and social psychology, 88(3), 498. https://doi.org/10.1037/0022-3514.88.3.498
» https://doi.org/10.1037/0022-3514.88.3.498 -
Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F. J. (2003). Political Conservatism as Motivated Social Cognition. Psychological Bulletin, 129, 339-375. https://doi.org/10.1037/0033-2909.129.3.339
» https://doi.org/10.1037/0033-2909.129.3.339 -
Jost, J. T., Pelham, B. W., Sheldon, O., & Ni Sullivan, B. (2003). Social inequality and the reduction of ideological dissonance on behalf of the system: Evidence of enhanced system justification among the disadvantaged. European Journal of Social Psychology, 33(1), 13-36. https://doi.org/10.1002/ejsp.127
» https://doi.org/10.1002/ejsp.127 -
Kay, A. C., & Jost, J. T. (2003). Complementary justice: Effects of “poor but happy” and “poor but honest” stereotype exemplars on system justification and implicit activation of the justice motive. Journal of personality and social psychology, 85(5), 823-837. https://doi.org/10.1037/0022-3514.85.5.823
» https://doi.org/10.1037/0022-3514.85.5.823 -
Kosakowska-Berezecka, N., Besta, T., Adamska, K., Jaśkiewicz, M., Jurek, P., & Vandello, J. A. (2016). If my masculinity is threatened I won’t support gender equality? The role of agentic self-stereotyping in restoration of manhood and perception of gender relations. Psychology of men & masculinity, 17(3), 274. https://doi.org/10.1037/men0000016
» https://doi.org/10.1037/men0000016 -
Kray, L. J., Howland, L., Russell, A. G., & Jackman, L. M. (2017). The effects of implicit gender role theories on gender system justification: Fixed beliefs strengthen masculinity to preserve the status quo. Journal of personality and social psychology, 112(1), 98. https://doi.org/10.1037/pspp0000124
» https://doi.org/10.1037/pspp0000124 - Lima, M. E. O. (2020). Psicologia Social do Preconceito e do Racismo. Blucher Open Access. 143 p.
- Marôco, J. (2010). Análise de equações estruturais: Fundamentos teóricos, software e aplicações. ReportNumber. 410 p.
-
McKinsey & Company & Lean In. (2024). Women in the workplace 2024. Recuperado em 09 de dezembro de 2024, de Recuperado em 09 de dezembro de 2024, de https://www.flipsnack.com/78C9ACFF8D6/women-in-the-workplace-2024/full-view.html
» https://www.flipsnack.com/78C9ACFF8D6/women-in-the-workplace-2024/full-view.html -
Ministry of Labor and Employment and the Ministry of Women (2024). Painel do Relatório de Transparência salarial. Recuperado em 09 de dezembro de 2024, de Recuperado em 09 de dezembro de 2024, de https://app.powerbi.com/view?r=eyJrIjoiMjNhY2IyN2UtMjllMy00MjQxLTg2YzYtNjcyYzk0YjE4MjQxIiwidCI6IjNlYzkyOTY5LTVhNTEtNGYxOC04YWM5LWVmOThmYmFmYTk3OCJ9
» https://app.powerbi.com/view?r=eyJrIjoiMjNhY2IyN2UtMjllMy00MjQxLTg2YzYtNjcyYzk0YjE4MjQxIiwidCI6IjNlYzkyOTY5LTVhNTEtNGYxOC04YWM5LWVmOThmYmFmYTk3OCJ9 -
Nascimento, T. G., Pimentel, C. E. & Adaid-Castro, B. G. (2016). Escala de Atitudes frente à Arma de Fogo (EAFAF): Evidências de Sua Adequação Psicométrica. Psicologia: teoria e pesquisa, 32(1), 239-248. https://dx.doi. org/10.1590/0102-3772201602187239248
» https://dx.doi. org/10.1590/0102-3772201602187239248 - Pasquali, L. (2010). Instrumentação psicológica: fundamentos e práticas. Artmed Editora.
-
Pavot, W. & Diener, E. (1993). Review of the satisfaction with life scale. Psychological Assessment, 5, 164-172. https://doi.org/10.1037/1040-3590.5.2.164
» https://doi.org/10.1037/1040-3590.5.2.164 -
Pinheiro, L.; Medeiros, M.; Costa, J.; Barbosa, A. H. (2023). Gênero é o que importa: determinantes do trabalho doméstico não remunerado no Brasil. Ipea, 44 p., http://dx.doi.org/10.38116/td2920-port
» http://dx.doi.org/10.38116/td2920-port -
Prusaczyk, E., & Hodson, G. (2020). “To the moon, Alice”: Cavalier humor beliefs and women's reactions to aggressive and belittling sexist jokes. Journal of Experimental Social Psychology, 88, 103973. https://doi.org/10.1016/j.jesp.2020.103973
» https://doi.org/10.1016/j.jesp.2020.103973 -
Raykov, T. (1997). Estimation of composite reliability for congeneric measures. Applied Psychological Measurement, 21(2), 173-184. https://doi.org/10.1177/01466216970212006
» https://doi.org/10.1177/01466216970212006 -
Saguy, T. & Szekeres, H. (2018). Changing minds via collective action: Exposure to the 2017 Women’s March predicts a decrease in (some) men’s gender system justification over time. Group Processes & Intergroup Relations, 21(5), 678-689. https://doi.org/10.1177/1368430217750475
» https://doi.org/10.1177/1368430217750475 -
Siegrist J, Wege N, Pühlhofer F, Wahrendorf M. (2009). A short generic measure of work stress in the era of globalization: effort-reward imbalance. Int Arch Occup Environ Health, 82(8):1005-13. https://doi.org/10.1007/s00420-008-0384-3
» https://doi.org/10.1007/s00420-008-0384-3 - Tabachnick, B. G., Fidell, L. S., & Ullman, J. B. (2007). Using multivariate statistics (Vol. 5, pp. 481-498). Pearson.
-
Vasconcellos, I. R. R. D., Griep, R. H., Portela, L., Alves, M. G. D. M., & Rotenberg, L. (2016). Adaptação transcultural para o Português brasileiro e confiabilidade da escala esforço-recompensa no trabalho doméstico. Revista de Saúde Pública, 50. https://doi.org/10.1590/S1518-8787.2016050006138
» https://doi.org/10.1590/S1518-8787.2016050006138
Research data is available on request from the corresponding author.
Note: ***p<0.001
