College Students’ Stereotyped Beliefs

Abstract

Over recent years, socio-political discourse has been full of language aimed at reaching gender equality. This is a complex goal that should address the underlying bases of inequality—gender stereotypes that continue to legitimize unequal consideration and treatment. It is also a reality that universities are not exempt from. The objective of this study was to analyze university students’ stereotyped beliefs and look at the differences based on self-identified gender and branch of knowledge. The analysis looked at a sample of 3433 university students (67.9% women), aged between 17 and 56 (M = 18.95; SD = 2.35) and reported low rates of prevalence of stereotyped beliefs, with significantly higher means in men and in engineering students. The survival of gender stereotypes in a population who were born and raised in a legally egalitarian society points to the importance of education programs aimed at university teachers that would give them the capacity to incorporate a gender perspective in all disciplines, especially those disciplines reporting greater adherence to stereotyped beliefs.

1. Introduction

Gender equality is one of the great social concerns of the third millennium. Many Western societies are committed to the struggle to eradicate all forms of discrimination against women and girls. This commitment is manifest in the Sustainable Development Goals, which in 2015 defined the shared responsibility of the great social change towards gender equality (SDG 5) (United Nations, UN 2015). Spain cannot be left out of this collective task, especially in light of the empirical confirmation showing clear gender inequalities and the persistence of a sexist mentality in the country (Árbol and Entrena 2022). Along similar lines, the first Global Gender Gap Report, published in 2006, put Spain in the 11th position out of 128 countries (Hausmann et al. 2006), a position it no longer maintains. According to more recent data, the gender gap has grown, and Spain was in 22nd place in 2022, the same position it had in 2008 and 2009 (World Economic Forum 2022). That is not something we can ignore. Nonetheless, the challenge of achieving true gender equality cannot be seen just as a numerical objective; the complexity of the goal must be recognized (Phillips 2019). This complexity refers to the underlying situations that provide a foundation for inequality, sustained by the gender stereotypes that continue to legitimize unequal consideration and treatment of men and women.

Gender stereotypes refer to the set of “beliefs, images, or expectations people have about men and women” (Moya and Moya-Garófano 2021, p. 53). They are causal attributions about socially shared expectations of what it means to be a “man” or a “woman.” Men are associated with qualities such as dominance, dynamism, and competitiveness—instrumental characteristics—whereas women are associated more with the expressive and assumed to be sweet, tender, and sensitive (Ellemers 2018; García and Huertas 2001; González and Rodríguez 2020). In this way, men are guided towards the idea of agency (ambition, assertiveness, etc.) and women towards the idea of community (compassion, kindness, etc.) (Sczesny et al. 2019).

However, these harmful generalizations do not stop at addressing traits or descriptions of each gender; they also affect gender roles. In other words, they dictate the legitimacy of gender roles, behaviors, and occupations (e.g., doctor vs. nurse), affecting not only what beliefs men and women have, but also what beliefs they should have (Kollmayer et al. 2018). They go beyond being mere precepts (Chalabaev et al. 2013), as they are articulated in social discourse describing how men or women should behave. These powerful mechanisms contribute to women’s lack of visibility in various settings (Eizmendi and Peña 2023) and permeate various parts of our social reality, such as the media (Kitsa and Mudra 2020) and film (Dundes 2020; Streiff and Dundes 2017). This is not, therefore, a trivial topic in the construction of a fairer, more equitable society. In this regard, gender stereotyping has been shown to be a serious obstacle to achieving true gender equality (Lopez-Zafra and Garcia-Retamero 2021).

Accepting harmful social representations based on gender undoubtedly occurs from an early age. The empirical evidence indicates stereotyped beliefs in infancy (King et al. 2021) and adolescence (Villanueva and Grau 2019), which is a confirmed reality despite the possible “blindness” to discriminatory situations because they are so normalized (García-Pérez et al. 2011). But what happens in the university setting? Are students who reach the university level less prone to hold to this dogma? It seems that gender inequality in terms of roles and differential traits is not treated lightly in the academic setting.

Studies in recent years focusing on gender stereotypes have shown that university students exhibit gender stereotypes but do so from a more ambivalent position. A meta-analysis of opinion polls between 1946 and 2018 by Eagly et al. (2020) indicated how gender stereotypes on a general level have changed over time. International research about stereotyped beliefs in university students show that they still think that men are by nature more aggressive than women, who they assign a more nurturing role to (Cubillas et al. 2016). The scientific literature has identified dogmatic beliefs in the young population that lead to women being less likely to enter disciplines such as science, technology, engineering, and mathematics (Chauke 2022; Fisher et al. 2020); this fact motivates the inclusion of branch of knowledge as a predictor of interest.

In the Spanish context, the results from Castillo and Montes (2014) indicate that traditional feminine stereotypes such as dependence and complacency and masculine stereotypes such as self-efficacy and competence are no longer current in university students. However, there is still causal attribution that links women to traits such as sweetness, submission, sensitivity, and delicateness, whereas men are associated with bravery, insensitivity, stability, and possession (Castillo and Montes 2014; Rodríguez and Mancinas 2016). The evidence does not just involve stereotyped beliefs about traits but also includes differential attributions related to roles, such as boys being more able to lead (Pacheco et al. 2014).

More recently, the pattern of results has continued. University students continue to exhibit beliefs rooted in sexist gender stereotypes, through which female students are valued for their dependency and fragility (León and Aizpurúa 2020). In fact, both male and female students indicate the three core aspects of gender stereotyping of women: (a) more responsibility for housework; (b) childcare; and (c) more pressure to fit a certain body image (Merma et al. 2022). A transtemporal analysis (2006–2016) by Lopez-Zafra and Garcia-Retamero (2021) sheds light on some of the changes in characteristics attributed differently to men and women. They identified that men felt that they were increasing their feminine characteristics, whereas women were only increasing stereotyped beliefs in relation to cognitive (e.g., being good with numbers or solving problems) and physical elements (e.g., being stronger or more muscular).

These limiting beliefs affect everyone. Even though there are limited studies on gender stereotypes and non-binary people, it has been revealed that gender stereotypes stigmatize (trans)gender people (Howansky et al. 2019).

To discover the archetypal reality that still permeates academia, the aim of the present study is to examine the stereotyped beliefs of university students. There are three specific objectives:

• Evaluate the incidence of stereotyped beliefs in students.
• Determine whether there are statistically significant differences related to gender traits and roles based on students’ self-identified gender.
• Determine whether there are statistically significant differences related to gender traits and roles based on the students’ branch of knowledge.

2. Methods

2.1. Participants

After selection using random intentional sampling stratified by scientific-academic area, a total of 3433 students at the University of Santiago de Compostela (USC-Spain) (67.9% women, 30.8% men, 1.3% non-binary), aged between 17 and 56 years old (M = 18.95; SD = 2.35), participated voluntarily in the study. They were predominantly studying in social and legal sciences (53.7%), followed by science (18.6%), health sciences (13.4%), arts and humanities (13.1%), and engineering and architecture (0.6%). Consequently, 18% of the reference population (19,060 students) is estimated to have participated.

2.2. Instrument

The Gender Stereotypes Scale created by Cubillas et al. (2016) was the basis for creating a Spanish version of the instrument. It is made up of a series of statements that explore young people’s positions on gender stereotypes. The scale was constructed considering the theoretical framework around gender studies and related research (Rocha and Díaz 2005; Castro and Casique 2010). The items were assessed for comprehensibility by a panel of experts drawn from teachers at the University of Santiago de Compostela, and the scale was applied with their indications in mind. The items were scored on a Likert-type scale with responses from (1) completely disagree to (5) completely agree. Each student was placed on a gradient, from a conservative position of notable differences in the conception of what it is to be a man or a woman and the roles for each to a more egalitarian position, which views men and women as having similar rights and abilities.

The final questionnaire was the result of the following steps:

• Bibliographic review by the research team, triangulated with external validation performed by five expert judges using an assessment form, who were also asked to assess the items considering criteria of validity, location, intelligibility, and unambiguousness.
• Pilot administration to 280 students to validate the morphosyntactic suitability of the items and the questionnaire’s fit to the starting construct.
• Analysis of the instrument, reported in the present study.

The final scale was called the Spanish Gender Stereotypes Scale (GSS-E). It is worth noting that we chose the designation “scale” rather than “questionnaire” because, as Bolaños-Medina and González-Ruiz (2012, p. 717) put it, it deals with a “set of declarations that measure the subjects’ levels of agreement or disagreement with them,” and, in addition, the questions have graduated responses (Morales 2011).

The final version of the instrument is made up of 18 items (α = 0.84) grouped into two dimensions: Dimension 1—Stereotypes linked to gender traits (α = 0.714—Items: 5, 6, 8, 11, 14, 15, 16, 17, 18; e.g., Men represent intelligence and protection) and Dimension 2—Stereotypes linked to gender roles (α = 0.778—Items: 1, 2, 3, 4, 7, 9, 10, 12, 13; e.g., The most important thing for a woman is to be a mother). Reliability was assessed by calculating the internal consistency using Cronbach’s alpha. Given that the measures of reliability for both dimensions and the overall scale were above 0.70, the scale is understood to have an excellent index of internal consistency (Oviedo and Campo-Arias 2005).

Following the review and analysis, the questionnaire was structured as follows: (1) objective of the questionnaire, data protection, instructions, and thanks; (2) data about the person completing the questionnaire; (3) the scale items about the beliefs around gender stereotypes; and (4) acknowledgements.

2.3. Procedure

Once approval was obtained from the Ethics Committee at the University of Santiago de Compostela (code of ethical approval of research: USC 36/2022), contact was made with the participating departments. Teachers were asked to collaborate, and various researchers administered the scale to the students in their classrooms in single sessions during the 2021–22 academic year. The researchers were present during administration to provide any necessary help and to ensure that the scale was completed correctly. Prior to this, there was a signed agreement completed to follow the ethical directives of the American Psychological Association with regard to consent, confidentiality of responses, and anonymity of the participants.

The students were Informed of the study objectives and were told that participation was voluntary and confidential—to avoid, as far as possible, biases due to social desirability—and they were asked to answer as honestly as possible. Completing the questionnaire took around 10 min, in a suitable environment (free from significant distraction), which allowed them to concentrate. To reduce the temptation to lie in responses, the instrument did not contain any questions that might lead to subjects’ identities being determined, and responses were given an ID code that preserved respondents’ anonymity.

The study applied a transversal study design. This type of design allows for the description of a population at a given point in time. In addition, it also allows relationships between variables to be established, along with differences between different segments of the population (León and Montero 2003).

2.4. Data Analysis

This study followed a quantitative, empirical-analytical approach, with a non-experimental, exploratory design. The first step in analyzing the data was to perform a descriptive analysis to assess its quality and distribution. Then, Cronbach’s Alpha was calculated to show suitable internal consistency of the scale and the dimensions making it up. Subsequently, the latent variables associated with the two factors making up the scale were constructed from the observed variables (items), followed by descriptive analysis of the newly constructed variables and analysis of variance (ANOVA) to meet the objectives of the study. Following that, Scheffé post hoc test was applied. The effect size was determined using partial eta squared (ηp² and interpreted following Cohen’s (1988) criteria, which indicate a small effect size when ηp² = 0.01 (d = 0.20), moderate when ηp² = 0.059 (d = 0.50), and large when ηp² = 0.138 (d = 0.80). In the ANOVA, the independent variables were gender and the branch of knowledge that the student was studying in, and the dependent variables were the two dimensions noted above (D1: Stereotyped beliefs associated with gender roles; D2: Stereotyped beliefs associated with gender traits). When the difference of means analysis was performed, Levene’s statistic was used to assess whether the principle of homoscedasticity was complied with. In cases where the significance prevented verification of a case, robust Brown-Forsythe (F*) tests were used, followed by post hoc Games-Howell tests.

3. Results

3.1. Descriptive Statistics of the Study Variables

Table 1. Descriptive statistics of the items in the Gender Stereotypes Scale and the constructed variables.

	Min.	Max.	M	SD
1. Childcare is more of a woman’s responsibility than a man’s.	1	5	1.14	0.54
2. Children obey their fathers not their mothers.	1	5	1.42	0.81
3. The most important thing for a woman is to be a mother.	1	5	1.21	0.61
4. Children are better taught by a mother than by a father.	1	5	1.37	0.78
5. Women represent love and weakness.	1	5	1.26	0.65
6. Being a man is better than being a woman.	1	5	1.28	0.78
7. If a man makes enough money, the woman doesn’t have to work.	1	5	1.20	0.64
8. For men, the priority in life is to have a house and a car	1	5	1.53	0.93
9. Men are better suited to jobs that need mental agility (e.g., engineers, etc.)	1	5	1.19	0.59
10. The man should always be the boss at home.	1	5	1.08	0.40
11. Men represent intelligence and protection.	1	5	1.15	0.50
12. Women are better suited to jobs where they provide care, treatment, and service (e.g., psychologists, nurses, teachers, etc.).	1	5	1.36	0.79
13. Women cannot perform the same activities as men.	1	5	1.25	0.69
14. Men are unfaithful by nature.	1	5	1.31	0.77
15. By nature, men need sex more than women.	1	5	1.47	0.89
16. Men shouldn’t show their weaknesses or their feelings.	1	5	1.13	0.54
17. Women are emotionally stronger than men.	1	5	1.76	1.02
18. Men are more aggressive than women.	1	5	2.31	1.18
F1_GenderRoles	1.00	4.78	1.24	0.39
F2_GenderTraits	1.00	4.00	1.46	0.45

Note. M = mean; SD = standard deviation.

shows the results of the descriptive analysis of the observed variables (items) and constructed variables (dimensions) in order to be able to assess the prevalence of these stereotyped beliefs in the students (objective 1). The mean scores were generally low or very low in all of the variables in the study. The highest mean score was in item 18 (Men are more aggressive than women) (M = 2.31; SD = 1.18), which is also a moderately low level for this reported belief. The lowest mean score was in item 10 (The man should be the boss at home) (M = 1.08; SD = 0.40), indicating that 91.8% of the sample completely disagreed with this statement.

With regard to the constructed variables, low mean scores were reported for the two types of stereotyped beliefs: those related to gender traits (M = 1.46, SD = 0.45), and those related to gender roles (M = 1.24, SD = 0.39). It is worth noting that even though these scores indicate low levels of these stereotyped beliefs reported in university students, there was a higher level of these ideas when they alluded to people’s (physical or personality) traits (F2) than when they indicated differences in gender roles (behavior or occupations) (F1).

3.2. Gender Differences in Stereotyped Beliefs

In response to the second specific objective of this study, we performed an analysis of robust tests of equality of means (using, specifically, the Brown-Forsythe test), as the principle of homoscedasticity was not met. The dependent variables were the two types of stereotyped beliefs, and the independent variable was the student’s gender. The results showed statistically significant differences in the prevalence of stereotyped beliefs about gender roles (F*_{(2, 615)} = 138.70, p < 0.001; ηp² = 0.080, moderate effect size) and gender traits (F*_{(2, 157)} = 56.63, p < 0.001; ηp² = 0.044, small effect size). Following post-hoc Games-Howell analysis, the results indicated that male students exhibited more stereotyped beliefs about gender traits (M = 1.61; SD = 0.54) than their female counterparts (M = 1.40; SD = 0.39). No significant differences were found for non-binary students (M = 1.44; SD = 0.52). The results of post-hoc testing for the variable related to stereotyped beliefs about gender roles between the groups gave a similar pattern. Female students (M = 1.17; SD = 0.28) and non-binary students (M = 1.14; SD = 0.25) had significantly lower mean scores than male students (M = 1.42; SD = 0.54) (Figure 1).

3.3. Differences in Stereotyped Beliefs Based on Branch of Knowledge

The Brown-Forsythe test was again used to analyze the results of the differences in means, with the two dimensions related to students’ stereotyped beliefs (traits-roles) as dependent variables, and the branch of knowledge of the course they were studying as the independent variable. There were statistically significant differences according to branch of knowledge in Dimension 1 (stereotyped beliefs about gender roles) (F*_{(4, 40)} = 4.523, p = 0.004; ηp² = 0.008, small effect size) and in Dimension 2 (stereotyped beliefs about gender traits (F*_{(4, 51)} = 3.204, p < 0.05; ηp² = 0.006, small effect size). Following robust post-hoc tests to determine which groups demonstrated which differences in Dimension 1, students doing engineering and architecture were found to have a significantly higher mean score (M = 1.59; SD = 0.74) than students in other branches of knowledge. There were also substantial differences between the social and legal sciences branch and the arts and humanities branch, with the latter having a lower mean score (M = 1.17; SD = 0.29) (

Table 2. Descriptive statistics of the dependent variables by branch of knowledge.

	Mean	SD	Minimum	Maximum
D1_Roles
Science (G1)	1.23	0.37	1.00	3.67
Social and legal sciences (G2)	1.26	0.42	1.00	4.78
Health sciences (G3)	1.23	0.36	1.00	3.44
Engineering and architecture (G4)	1.59	0.74	1.00	3.67
Arts and humanities (G5)	1.17	0.29	1.00	3.22
D2_Traits
Science (G1)	1.43	0.42	1.00	3.44
Social and legal sciences (G2)	1.47	0.47	1.00	4.00
Health sciences (G3)	1.47	0.45	1.00	3.33
Engineering and architecture (G4)	1.88	0.80	1.00	3.44
Arts and humanities (G5)	1.42	0.42	1.00	3.56

Note. Post-hoc test: Significant differences in D1 between G4 and the other groups and between G2 and G5. In D2, there were statistically significant differences between G4 and the other groups.

).

The results of the post-hoc testing for the second dimension (stereotyped beliefs about gender roles) indicated significant differences between the mean scores from engineering and architecture students (M = 1.88; SD = 0.80) and the scores from students in other branches of knowledge, with engineering and architecture students again scoring higher in having stereotyped beliefs about gender traits (Table 2 and Figure 2).

4. Discussion

This study examined the components of gender stereotypes in university students, looking at differences in relation to self-identified gender and the branch of knowledge of the course the students were enrolled in. The results indicated that generally low scores are reported, especially related to men’s roles in private spaces and their childcare responsibilities. These results bear witness to the change in stereotyped beliefs in Spain, at least among young people. Our results are in line with Social Role Theory, which indicates the changing nature of gender stereotypes in line with the social situation of men and women (Eagly 1987). This paradigm explains the changes that have affected gender relations, especially with regard to women’s roles (Eagly et al. 2020). In this regard, the 2014 study by Castillo and Montes—also with Spanish university students—indicated beliefs associating productive space with men and private space with women. Today, our results confirm that stereotypes about masculine roles are not immutable (Moya and Moya-Garófano 2021), and the stereotypes that have undergone the most change are those under the heading “family and children” (Priyashantha et al. 2021).

Spanish society has seen substantial change in the last five years in democratic policies, particularly with regard to gender equality, and it is this that has affected the gender stereotypes detected in the participants, according to recent studies focusing on the university student population (Aydin Ozkan and Kucukkelepce 2019). However, if we look at two important variables in the study of gender roles, the change does not seem so clear. Firstly, the statistics show that in Spain, women continue to shoulder more of the burden of childcare (INE 2022). It seems to be the youngest who are more accepting of coparenting and co-responsibility, a result in line with the feminist values that find acceptance in this age group (León and Aizpurúa 2020).

In addition, changes to gender stereotypes are not produced in all of the dimensions; nor do they have a uniform impact. Despite low expressed scores, the higher scoring items in our results (the more widely held beliefs) included those suggesting that men “needed” sex more than women and those identifying women as emotionally stronger than men. This is in line with the original study (Cubillas et al. 2016). Our study also shows the continuation of stereotypes referring to the physical characteristics men and women should have (Moya and Moya-Garófano 2021), but also to those about sexuality and personality traits. There are variations in gender stereotypes, but descriptive stereotypes (traits) have not changed as much as prescriptive stereotypes (roles outside the home). Perhaps men’s apparent movement towards feminine attributes is not as rapid as hoped (Lopez-Zafra and Garcia-Retamero 2021), and they remain wedded to a hegemonic model of masculinity (Lund et al. 2019) that seeks to give men the active role in sexuality (Pérez 2020).

In terms of differences by gender, our data show that men had more stereotyped beliefs than female and non-binary students, with the magnitude of these differences being moderate or small. This is not a novel result in the literature. National-level (Fernández et al. 2022; Madolell et al. 2020) and international studies have confirmed this idea. In fact, the evidence indicates that female university students are more sensitive to questions of gender (Khan et al. 2018). It is reasonable to think that female and non-binary students do less to propagate stereotypes, as they are the groups that suffer most from the prejudice and stigma that these beliefs produce. In effect, the differences we found between women and non-binary people indicate more stereotyped beliefs in the latter. In this regard, as far as we are aware, these are the first results to identify gender differences that include non-binary people in Spain.

We also found small but significant differences according to the branch of knowledge of the degree programs the students were enrolled in at the university under study, which are worth highlighting. Engineering and architecture students had more stereotyped beliefs, while arts and humanities students had the lowest level. One explanation for this may lie in the evident predominance of one gender or other in certain university courses. Courses within engineering and architecture are overwhelmingly taken by men. This imbalance is clear in the data from the Spanish Ministry for Universities (2023), indicating that in the 2021–22 academic year, three times fewer women than men applied to complete engineering qualifications. This subtle disproportion in favor of men has been the norm throughout the history of Spanish universities, meaning it is no surprise that this is the knowledge area where there are more stereotyped beliefs, particularly as it is male students who more commonly hold these beliefs.

In contrast, arts and humanities participants have the least stereotyped beliefs. These are courses that are predominantly taken by women, which would explain this result. However, comparing our results with the literature is more of a challenge, because in Spain, the available studies only compare some of the courses our study covered or group them in other knowledge structures (Esteban and Fernández 2017; Madolell et al. 2020). For example, in our study, students in the social and legal sciences did not have the lowest scores in either dimension. In fact, there were significant differences compared to arts and humanities students in the gender roles dimension. This may be because of the range and diversity of courses that fall under the social and legal sciences branch. Perhaps students who are studying for degrees in law or economics are not as “exposed” to educational stimuli that promote the removal of gender stereotypes as students in education sciences. In this regard, the results from Madolell et al. (2020) indicate that students taking courses in health and education have less stereotyped beliefs than students studying law or business administration.

Despite the empirical contribution to the admittedly scarce literature about the topics covered by this study, this study does have some limitations that should be borne in mind. The sample was limited to a single public university, and despite being configured to be representative of that institution, it does not cover other cultural contexts in Spain, which limits how far the results may be generalized. Future work needs to involve universities that reflect the cultural diversity of Spain, particularly those with different ownership models (private and public). Another issue to bear in mind when interpreting the results is that of social desirability bias, because it is still possible that respondents are unwilling to report their beliefs, or that they may be unwilling to admit to themselves that they hold these stereotypes. Likewise, we intend to expand the sample in the area of engineering and architecture in order to avoid potential biases derived from the reduced participation of students in this field and to provide a more solid and consistent base for the differences identified. In addition, given the complexity of the topic, it would be extremely interesting to complement the results with data from a qualitative study. That would allow for a deeper, more comprehensive picture of the phenomenon.

5. Conclusions

The results of this study shed light on the deconstruction of a restrictive, limited model that places women in an unequal position compared to their male counterparts. Our data are promising in this regard. However, we must not forget that gender stereotypes continue to permeate various areas of Spanish social life (Moya and Moya-Garófano 2021). If we consider that the discourse on gender equality has value in and of itself for the academic space (Silander et al. 2022), we can accept that the university setting is a space that is isolated from social reality. However, academic culture itself tends to perpetuate sexism, ignoring the reasons, often unconscious and alienated, behind these sexist practices (Espinoza and Albornoz 2023; Troncoso et al. 2019). This is why we must continue working towards true equal education that will allow us to work for the dismantling of gender stereotypes and hence the creation of a fairer, more equal society, trying to reach, as recommended, SDG 5, which refers to achieving gender equality and empowering all women and girls (Odera and Mulusa 2020). There is a clear need for education programs aimed at university teachers with the objective of specifically incorporating a gender perspective in all qualifications. This is particularly the case in the areas that reported greater levels of stereotyped beliefs. It is presumed necessary to re-evaluate the learning paths of professors, through approaches that allow us to identify and reflect on the new challenges, demands, and proposals referred to in the competencies of teaching excellence in higher education (Estévez et al. 2023). It continues to be important to include teaching strategies that encourage gender equality—including at the university level—so that we do not lose the progress that has been made up to now.