Introduction

Researching teachers' attitudes towards homosexuality helps us to understand the extent to which teachers can engage in truly inclusive pedagogy. However, few studies have addressed this issue, especially because there are few measurement tools that capture teachers’ attitudes towards homosexuality. The literature highlights how teachers’ homophobic attitudes can lead them to avoid engaging with topics concerning sexuality and homosexuality in classrooms (Bhana, 2012; D’Urso et al., 2017; Scandurra et al., 2017). This is an important issue, given that many studies highlight how knowledge of homosexuality can help adolescents reduce their prejudices and form a vision of a more inclusive society (e.g., Petruccelli et al., 2015). Film and television pedagogy is one way in which teachers can engage or disengage with the topic of homosexuality. Film has been used as a pedagogical tool in universities and schools since at least the 1940s (e.g., May, 1946), and is in extensive use in well-equipped classrooms via interactive whiteboards, computers, and tablets (e.g. Cho & Johnson, 2020), and as part of students’ remote learning via digital technologies (Razak & Din, 2020). Given the widespread use of film and television pedagogy internationally and its potential to convey social norms regarding homosexuality to promote inclusive societies, our study focused on creating a tool for assessing teachers’ attitudes towards the representation of homosexuality in film and television. We describe the development and testing of this tool, and further explore the psychological disengagement mechanisms that regulate homophobic attitudes among teachers.

Prejudices and Pedagogy

In school contexts, teachers’ prejudices can impact how they communicate social norms to students through pedagogy (Chambers et al., 2013; Dessel, 2010; Vervaet et al., 2018). Prejudices are judgments that are expressed a priori, without firm evidence underpinning the basis of the judgment (Brandt & Reyna, 2010; Pereira et al., 2010). A prejudice creates subjective categories in people’s representations of society (e.g., an apparent social ‘group’), preventing inclusive attitudes towards social differences. Social prejudices can lead individuals to satisfy their needs for a positive and distinct social identity from the outgroup which, consequently, they perceive as being a threat to the identity of their ingroup (Tajfel, 1984). In other words, social actions and norms that are conceived as different from an individual’s self and social and cultural group can be considered as a threat to the individual’s personal and group identities. Therefore, prejudice helps, albeit sometimes in a deleterious way, an individual to forge a stronger sense of belonging to the ingroup and reduces their vulnerability of aligning with the outgroup (Petruccelli et al., 2015).

Teachers’ social prejudice can be an obstacle towards using inclusive pedagogies aimed at reducing prejudice in students, while at the same time influence students’ evaluations of social groups (Hwang & Evans, 2011 ; Pérez-Testor et al., 2010). Several studies observe a connection between homophobic prejudices and pedagogy, for example, teachers can be complicit in working against the rights of gay and lesbian people at school by not including representations of homosexuality in the curriculum, through religious speeches that portray heterosexual marriage as a moral standard, and by choosing the path of silence towards LGBT people by emphasizing the heteronormativity of society (Bhana, 2012; D’Urso et al., 2017; Msibi, 2012).

Homophobia Among Teachers

Homophobia has been defined as a complex construct that includes a set of distorted beliefs, strong prejudices, and negative attitudes, regarding homosexual people and their lifestyle, including their romantic relationships, parenting practices and opportunities, work roles, and social status (Anderson, 2018; Morin & Garfinkle, 1978). In Herek’s (2000) perspective, homophobia can be defined as a form of prejudice based on sexual orientation, that can manifest in verbal and non-verbal language made up of derogatory expressions and appellations and aversion to same-sex romantic and sexual relationships.

A common finding in research is that homophobia can negatively influence attitudes and beliefs of people who identify as being sexual minorities, for example weakening their efficacy to married or became parents (Petruccelli et al., 2015). Recent studies in the Italian context have highlighted teachers have held homophobic prejudices influenced by stereotyped, religious and sexist beliefs. The authors emphasised also the importance of analysing homophobic prejudice in teachers to improve their educational practices (Scandurra et al., 2017). Another Italian study concluded that teachers avoid teaching and discussing situations about homophobia and homophobic bullying, perhaps because they consider these issues too complex to manage, and that they would not confidently provide adequate support to students (D’Urso et al., 2017). From the student’s perspective, teachers have been described as not talking about homophobia in their school modules, and instead affirming the idea of a society based mainly on heterosexuality andon gender-related prejudices (man must be strong and dominant), therefore not considering other social dimensions (Butler & Astbury, 2008; Gorski et al., 2013; Msibi, 2012).

Failing to Engage with Homosexuality

Living in a society where social representations favour heterosexual relationships and support negative social representations of homosexuality and homophobic prejudices may lead teachers to disregard homosexual issues when teaching (Nappa et al., 2018). Social representations emerge out of socially shared beliefs, ideas and values that are widely spread in the cultural system. They manifest in social, educational, and political systems, and influence how people defend their cultural and political identities (Howarth et al., 2014). When social representations convey a negative image of a social group, they act as a gateway to prejudice in society. Furthermore, exposure to negative social representations impacts the well-being of gay and lesbian people (Jaspal, 2019). Social representations of sexual minorities can act as an obstacle or help towards building an inclusive school context (Phillips, 2012), depending on their content and how they are used by teachers and students.

The impact of teachers not engaging with homosexual issues in the classrooms can include students feeling that their homosexual identities are ignored and not valued (Dèttore et al., 2014). Research has found that teachers often avoid talking about topics concerning sexual minorities, to the point of not consciously recognizing episodes of homophobic victimization against persons belonging to sexual minorities or people who do not adhere to gender social roles (D'Urso et al., 2017), as well as being inhibited towards inclusive practices aimed at reducing homophobic attitudes (Baruch-Dominguez et al., 2016; Douglas, et al., 1999).

In Italy, the setting of the current study, teachers have expressed feeling powerless and having little self-efficacy in dealing with situations of homophobic bullying (Nappa et al., 2018), while others have been observed to consolidate homophobic attitudes through hetero-sexist language or attitudes during their lessons (Piedra et al., 2016; Rivers, 2011). Other studies have documented Italian teachers having negative feelings, hostility, and anger towards LGBT issues because they are unfamiliar with the LGBT community or because they have sexist prejudices related to conservatism and religion (Scandurra et al., 2017).

Use of Movies and Television in Teaching

There is a large amount of research on how teachers use films, television shows, and advertisements as pedagogical tools (Walker, 2018). These pedagogical choices have repercussions on students’ social and cultural norms because they are interpreted on an emotional and cognitive level by students (D'Urso et al., 2020; Kindall-Smith et al., 2011). Other lines of study have highlighted how images in advertisements and movies can change students’ attitudes towards specific issues by visually and narratively presenting an alternative social and cultural representation of the issue to the representation that the students originally hold (Gestos et al., 2018; Myers et al., 2019). However, teachers sometimes do not respond with inclusive pedagogy practices to reduce social stereotypes and prejudices (Poteat et al., 2019). Therefore, a new form of inclusive pedagogy using videos supporting the LGBT community, can give students the opportunity to know normal social dimensions, which, especially for the infant or adolescent, may be new, and which once known can reduce the prejudice towards gay people and lesbians and same-parent families. This follows the logic that increased contact with minority groups, especially among adolescents, can reduce prejudice towards them (Petruccelli et al., 2015; West & Hewstone, 2012).

Moral Disengagement Among Teachers

A further issue impacting teachers’ negative attitudes towards homosexuals and implement acts of homophobia is moral disengagement (Almeida et al., 2009). Moral disengagement is conceptualized as a process of cognitive reconstruing or reframing of destructive behaviour as being morally acceptable without changing the behaviour or the moral standards (Bandura et al., 1996). It allows people to act contrary to their moral beliefs while still upholding the experience of behaving morally (Obermann, 2011). Moral disengagement can also allow the person to continue to have negative attitudes since they will tend to justify them. In other words, a person can use moral disengagement mechanisms to justify their homophobic attitudes (Scarpati & Pina, 2017). Recent studies with young adults have found that moral disengagement is connected with homophobic attitudes (Camodeca et al., 2019; Maftei & Holman, 2020). Other studies conducted with adults, have found that moral disengagement strategies can underpin deviant actions as well as cognitive distortions, including beliefs about rape and violence (Carroll, 2009; Gini et al., 2014; Petruccelli et al., 2017). This evidence suggests that further study of the link between homophobic attitudes and moral disengagement is needed in order to understand the social-cognitive functioning of people’s actions, which includes teachers’ choices of pedagogy (Petruccelli et al., 2017). In the current study, we take the opportunity to explore potential connections between teachers’ homophobic attitudes and moral disengagement, to further deepen the analysis of our new measure.

The Current Study

The study is divided into two sections: the first section explores the psychometric qualities of a new measure for evaluating how teachers’ homophobic prejudices can be reflected in their attitudes towards social representations of homosexuality on television and in advertising. Specifically, we report on the results of tests of reliability and the validity of long and brief versions of the measure, including exploratory factor analysis, confirmatory factor analysis, and a concurrent validity test with a measure of latent and manifest prejudice. Then in the second section we explore how moral disengagement strategies might be connected with teachers’ negative attitudes towards the representation of homosexuality in film and television. This information should enrich the developing field of research on moral disengagement and homophobia in the context of teachers and pedagogy (Petruccelli et al. 2017).

Methods

Participants and Procedure

The study involved 241 primary and secondary school teachers from the North of Italy. Participants were recruited through an online survey, which was communicated to prospective participants using a snowball sampling method, where researchers advertised the survey using Facebook and WhatsApp with the link being passed on to other people through social networks. The majority of participants (60.2%, n = 145) self-identified as female. Participants’ ages ranged between 24 and 64 years (M = 45.01, SD: 10.19). All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The research was approved by the ethics’ committee of the Social and Forensic Psychology Academy of Rome. Data were collected between August 2020 and December 2020.

Instruments

Teacher Attitudes Towards the Representation of Homosexuality in Film and Television

A new instrument was designed to measure teachers’ attitudes towards the representation of homosexuality in film and television (the ‘TAHFT’ measure). Fifteen items were designed by the first author, with expert input from the director of the Academy of Social and Forensic Psychology and a member of the research ethics committee. The items covered a range of emotional and social aspects of teachers’ homophobic attitudes regarding film and television. Given the novelty of this topic, the items were not adapted from previous measures. Each item was measured using a 5-point Likert type scale (1 = strongly disagree, 5 = strongly agree). The item wording, measurement structure, and measurement validity are reported on in detail in the results section. The final set of 15 items was developed to examine a wide variety of aspects regarding the social and emotional dimensions of prejudices and stereotyped representations of the social context (Brown, 2011). We also decided to mix a set of simple and general questions with more complex questions to motivate participants to complete the measure.

Latent and Manifest Prejudice

The latent and manifest prejudice scale (Pettigrew & Meertens, 1995; Italian version by Arcuri & Boca, 1996) is a self-report questionnaire comprising 20 items that can be answered using a Likert scale ranging from 1 (absolutely disagree) to 6 (absolutely agree). This questionnaire, through two subscales (one relating to the latent form of prejudice and the other to the manifest one), investigates the possible preconceptions and distorted beliefs that participants may have in relation to foreigners and their presence in society (e.g., it would be annoying if a loved one or a relative of his got married to a non-EU person; Italians and foreign immigrants will never be able to feel at ease with each other, even if they become friends). Both subscales had good reliability indices (latent prejudice α = 0.85; manifest prejudice α = 0.89). We used these subscales for testing the convergent validity of the new measure.

Moral Disengagement

The Italian version of the Moral Disengagement Scale (Caprara et al., 1996) has 32 items evaluated on a 5-point Likert-type scale (1 = completely disagree, 5 = completely agree). The scale comprises eight subscales of moral disengagement (with four items in each subscale): moral justification, attribution of blame to victim, euphemistic labelling, advantageous comparison, displacement of responsibility, diffusion of responsibility, distortion of consequences, and dehumanization of victim). The Cronbach α of the subscales ranged from 0.79 to 0.87.

Analysis Plan

First, we conducted a correlational analysis of the 15 items of the TAHFT. We calculated the Cronbach’s alpha coefficient to estimate the reliability of the measure. Second, we tested the construct validity of the TAHFT using an Exploratory Factor Analysis (EFA) with principal axis factoring and direct oblimin rotation, and a confirmatory factor analysis (CFA) to check the validity of the factor structure. We used a correlation analysis of the latent and manifest prejudice scale and the mean values of the TAHFT, to test the convergent validity of the TAHFT. We also tested the validity of brief version of TAHFT, which is comprised of four items connected with emotional aspects of teachers’ attitudes and four items connected with social aspects of teachers’ attitudes. This brief version provides researchers with a leaner version of the TAHFT made of the core items measuring the latent construct. To test the brief version, we used an Exploratory Factor Analysis with principal axis factoring, Cronbach’s alpha coefficient for reliability, and a Pearson correlation coefficient among the long and short forms. Finally, to explore the role of moral disengagement strategies we carried out a linear regression model with the eight mechanisms of moral disengagement as independent variables and the mean value of the new scale as dependent variable. We also used gender as control variable. All analyses were conducted with SPSS version 26 except for the confirmatory factor analysis which was conducted in Mplus version 8.

Results

Preliminary Analysis

A correlation of the 15 items of the TAHFT identified between-item associations ranging from r = 0.31 to r = 0.72. This analysis revealed that one item (Movies and advertisements offer positive models of same-sex parenting families) had a very low association with the other items in the measure. We therefore removed this item from further analyses.

Reliability

A Cronbach’s alpha coefficient of 0.93 for the 14 items suggested excellent reliability. Item–total correlations were moderate to strong, ranging from 0.59 to 0.79 (see Table 1).

Table 1 Descriptive statistics of the questionnaire (N = 241)
Full size table

Construct Validity

In the EFA, an examination of the scree plot (Cattell, 1966), and percentage of variance accounted for, revealed the presence of one factor. The EFA showed a factor structure with one principal dimension (eigenvalue > 1; 8.15), with 58.23% of total variance explained. Bartlett’s Test of Sphericity was statistically significant, χ 2 = 2156.2 (91), p < 0.001, indicating significant correlations among the variables. The measure of Kaiser–Meyer–Olkin of sampling adequacy was 0.954. The CFA demonstrated that the new scales had a good fit to the data (CFI = 0.94; RMSEA = 0.08; X2 (91) = 2311.22; p < 0.001). The items loaded well onto the scale, with loadings ranging from 0.60 to 0.86 (see Table 2).

Table 2 Factor loadings of the TAHFT items (using EFA and CFA)
Full size table

Convergent Validity

The mean value of the TAHFT showed moderate positive and significant correlations with the dimensions of latent ethnic prejudice (r = 0.35; p < 0.001) and manifest ethnic prejudice (r = 0.39; p < 0.001).

Brief Version

An EFA of the eight-items version of the TAHFT also revealed a single factor (eigenvalue > 1; 4.80), that explained 60% of the total variance (see Table 3). Bartlett’s Test of Sphericity was statistically significant, χ2 = 1162.47 (28), p < 0.001, indicating significant correlations among the variables. The measure of Kaiser–Meyer–Olkin of sampling adequacy was 0.93. The short version had strong internal consistency (Cronbach's alpha coefficients of 0.92). The Pearson correlation coefficients between the long and short forms were 0.97.

Table 3 Factor loadings of the brief version of TAHFT
Full size table

Impact of Moral Disengagement on Teacher Attitudes Towards Homosexuality

The regression model with teacher attitudes as the dependent variable and moral disengagement dimensions as predictors was significant F(8,232) = 4.77 p < 0.001 and explained the 20% of the variance in teacher attitudes (Table 4). In this model, euphemistic labelling emerged as a significant strong predictor of negative attitudes towards the representation of homosexuality in film and television (β = 0.32, p < 0.001). Moreover, the dehumanization of victims was a significant predictor of negative attitudes (β = 0.20, p < 0.05).

Table 4 Moral disengagement strategies on attitudes toward representation of homosexuality
Full size table

Discussion

The primary aim of this study was to create and preliminarily validate a tool for assessing teachers' attitudes towards the representation of homosexuality in film and television. Subsequently, we wanted to verify if and which mechanisms of moral disengagement were predictive of teachers’ negative attitudes in the Italian context. The analysis suggests a good reliability index across all items. With concern to the main psychometric properties of the questionnaire, we found good internal consistency and a one- factor solution and a satisfactory index of construct validity. Furthermore, the analysis showed a very good reliability of scale (long and brief version). Although we designed the measure assuming that there would be two dimensions (emotional and social), this assumption was not supported by the exploratory and confirmatory factor analyses. This finding suggests that, in line with the literature (Scandurra et al., 2017), the social and emotional aspects of attitudes towards a specific topic (e.g., sexual minorities) can be cognitively merged and influence each other. Furthermore, through a ‘halo effect’, a person’s emotions can convey their perceptions of a specific event or social group (Zebrowitz & Franklin, 2014).

The positive and significant association between the mean value of the new measure and the other self-report measures used for the assessment of characteristics connected to this phenomenon (i.e., latent and manifest prejudice) offers support for the concurrent validity of the TAHFT. More research is needed to investigate how teachers’ ethnic prejudices could develop in association with their negative attitudes towards the representation of homosexuality in films and television. In line with the literature (Petruccelli et al., 2015), ethnic prejudice can support individual’s supporting negative emotions toward foreign ‘others’ and encourage individuals to perceive foreign others as a threat towards their own social values. This type of prejudice shares many mechanisms with negative attitudes toward homosexuality. Having a deep-seated prejudice that aims to preserve traditional values can also influence similar forms of negative and prejudicial attitudes in individuals.

The results also suggest that euphemistic labelling is one of the mechanisms of moral disengagement that supports negative attitudes towards homosexual issues among teachers. This result, in line with the literature (Bandura et al., 1996; Petruccelli et al., 2017), indicates that individuals seek to relieve themselves of their own responsibility in recognizing and challenging social inequalities and prejudice, by disguising their wrong action or conduct through respectable language that can implement negative attitudes towards the representation of homosexuality in film and television. Moreover, the attribution of blame is connected with negative attitudes among teachers. Following previous studies (Camodeca et al., 2019), reducing one's internal control, through the perception of the other party as being ‘wrong’ in their actions, can be functional to support teachers’ negative ideas towards sexual minority groups, and also towards their representation in film and television. In other words, if the stereotyped ‘other’ is considered to be different and therefore deserves not to enjoy inclusive attitudes, this serves to justify the thinker’s actions and beliefs and can result in strengthening those representations and making them enjoyable to hold (Bandura et al., 1996; Petruccelli et al., 2017).

The study, although it has innovative results, must be considered in the light of several limitations. A first limitation is in the study’s use of self-report questionnaires that can be subject to social desirability bias of participants. Furthermore, the geographic context and the size of the sample does not allow for generalizing the results obtained to a larger population. Therefore, future studies could verify the reliability of this scale in different geographical areas, and in different social groups. This future research could test the cross-cultural structure of the social and emotional attitudes towards the representation of homosexuality in film and television as well as differences in the socio-cognitive correlates (e.g., moral disengagement) of those attitudes, in different cultural and social groups.

Conclusion

This study presented and demonstrated the reliability and validity of a new measure: the Teacher Attitudes Towards Homophobia in Film and Television; and explored its associations with moral disengagement. The positive associations found between teachers’ negative attitudes and moral disengagement signal the importance of understanding people’s negative attitudes towards homosexual lives from a basis of cognitive mechanisms and distortions that empower the person to ignore moral standards. This type of moral disengagement can turn into prejudice, which is in turn used to justify personal negative attitudes. Given the importance of this cycle, understanding one’s own prejudices and attitudes might help teachers critically examine why they might not embrace teaching about homosexual lives in the classroom. Accordingly, the TAHFT measure could be used by teachers to assess their own prejudices, although this can be an activity that brings resistance and social desirability into play. However, even just reading the questionnaire can help us understand how much it is necessary to work on prejudices to think about a new inclusive pedagogy that uses new tools to promote human rights and social inclusion. Studying teachers' attitudes towards sexual minorities and the ways in which the media convey their inclusion is a basic step in thinking about inclusive teaching. If teachers position themselves positively towards all social and family configurations regardless of ethnicity and sexual orientation, they be able to help create a paradigmatic revolution that embraces the good ideals and purposes of inclusive schooling. However, on the contrary, if teachers succumb to the negative trappings of their own prejudices, including using cognitive mechanisms to support and ‘forgive’ those prejudices, implementing a change in schools is a difficult step to take. It is difficult for teachers to convey a genuine message of inclusion, if this message is flawed by their own negative emotions and beliefs towards minority groups.