Navigating organizational diversity: addressing transphobia among human resources professionals

Abstract

Purpose

This study aims to analyze the relationship between attitudes towards organizational diversity and towards trans people, correlating them with the levels of transphobia among HR professionals. Drawing upon the Intergroup Contact Theory, we also seek to understand whether the results may be influenced by the contact variable.

Design/methodology/approach

Data were collected through the administration of a questionnaire to a sample of 184 HR professionals. The questionnaire was comprised of self-constructed questions and previously validated scales, including: Attitudes Toward Diversity Scale (ATDS) (Montei et al., 1996), Attitudes Toward Transgendered Individuals (ATTI) Scale (Walch et al., 2012) and Genderism and Transphobia Scale (GTS) (Hill and Willoughby, 2005).

Findings

The results do not show high levels of negative attitudes towards organizational diversity or intolerance towards trans people, being these variables correlated, with higher levels of intolerance in the portion of the sample without contact with trans people.

Originality/value

This article contributes to the increase of knowledge in the field of HR Development Policies focusing on the thematic of Diversity and Inclusion, which is a very current and important topic in organizations today. Also, the relevance and contribution of the study point to the need for Human Resources Management Practices that take on positive discrimination and/or encourage the adoption of affirmative actions by organizations.

Propósito

Este estudio tiene como objetivo analizar la relación entre las actitudes hacia la diversidad organizacional y hacia las personas trans, correlacionándolas con los niveles de transfobia entre los profesionales de recursos humanos. Basándonos en la Teoría del Contacto Intergrupal, también buscamos comprender si los resultados pueden estar influenciados por la variable de contacto.

Diseño/metodología/enfoque

Se recopilaron datos a través de la administración de un cuestionario a una muestra de 184 profesionales de recursos humanos. El cuestionario estaba compuesto por preguntas autoconstruidas y escalas previamente validadas, que incluían: Attitudes Toward Diversity Scale - ATDS (Montei et al., 1996), Attitudes Toward Transgendered Individuals (ATTI) Scale (Walch et al., 2012) y Genderism and Transphobia Scale - GTS (Hill and Willoughby, 2005).

Hallazgos

Los resultados no muestran altos niveles de actitudes negativas hacia la diversidad organizacional o intolerancia hacia las personas trans, siendo estas variables correlacionadas, con niveles más altos de intolerancia en la porción de la muestra sin contacto con personas transgénero.

Originalidad/valor

Este artículo contribuye al aumento del conocimiento en el campo de las políticas de desarrollo de recursos humanos, centrándose en la temática de Diversidad e Inclusión, la cual es un tema muy actual e importante en las organizaciones hoy en día. Además, la relevancia y contribución del estudio señalan la necesidad de prácticas de gestión de recursos humanos que asuman la discriminación positiva y/o fomenten la adopción de acciones afirmativas por parte de las organizaciones.

Citation

Baptista, J., Costa, D. and Gonçalves, S.P. (2024), "Navigating organizational diversity: addressing transphobia among human resources professionals", Academia Revista Latinoamericana de Administración, Vol. 37 No. 2, pp. 287-300. https://doi.org/10.1108/ARLA-09-2023-0154

Introduction

There is a complex relationship between gender and professional opportunities, with evidence suggesting that performance is evaluated based on gender (Schilt, 2006). The labor market is traditionally characterized as a heteronormative space, where both men and women are expected to conform to the construction, expression and maintenance of binary gender categories, reinforcing power inequality between a cisgender majority and a transgender minority (Butler, 2006; Thanem and Wallenberg, 2016). In this sense, assuming that the human capital is equal, it is important to understand the existence of mechanisms that reproduce gender stereotypes and their effects in the workplace. This leads to organizations not integrating trans people in the same way as cis people, given that literature suggests trans people are at a higher risk of unemployment or living under higher levels of poverty compared to cis people (Baptista et al., 2023a; Conron et al., 2012).

In a global context, it is estimated that around 25 million people do not identify with the gender assigned at birth (Winter et al., 2016), and in a European context, this number is estimated to range between 30 thousand and 1.5 million people (Amnesty International, 2014). The term trans describes all individuals who, in some way, experience a degree of incongruence between their gender identity and the sex assigned at birth (Winter et al., 2016). In turn, the term cis refers to all individuals who do not identify as trans and/or have a gender experience congruent with the gender identified at birth (Bauerband et al., 2019).

It is recognized that the fact that gender is still perceived as binary in numerous aspects of Western societies leads to many people being unaware of the existence of trans people and, consequently, not having any kind of contact with them. Various empirical studies have examined how contact with trans individuals leads to greater support for their rights (e.g. King et al., 2009) and reduced levels of prejudice and discrimination (e.g. Willoughby et al., 2010). This theorization has been termed as the Intergroup Contact Theory, initially developed by Gordon Allport, suggesting that intergroup contact can reduce prejudice, especially under the following conditions: (1) equal group status within the situation; (2) common goals; (3) intergroup cooperation; (4) authority support and (5) potential for friendship in the contact situation (Allport, 1954; Pettigrew, 1998).

Diversity can be defined as a variety of attributes that are inherently linked to each individual's uniqueness and recognition (Fleury, 2000). As a research topic, it is a highly emerging subject, with the first studies having emerged in the 1960s in the USA during the movements against discrimination, civil rights and equal access to work (Ayega and Muathe, 2018). LGBTQIA + diversity remains a segregated component within diversity management, encompassing gender diversity, a peripheral subcategory that further invisibilizes the experiences and concerns involving trans people (Ozturk and Tatli, 2016). In an organizational context, diversity can be defined as the range of differences among individuals within an organization, reflecting a changing workforce (Kim, 2006). The working population holds both positive and negative attitudes toward diversity, with negative attitudes being multifaceted, including prejudice and rejection of various initiatives aimed at promoting organizational diversity (De Meuse and Hostager, 2001).

Allport (1954) defines prejudice as a negative attitude toward other individuals based on their membership in a particular social group. In the organizational context, it can be described as a set of irrational beliefs that members of certain groups have less abilities, motivation or are responsible for a greater number of workplace issues. Prejudice has adverse effects both on a personal and organizational level, where targets of prejudice may exhibit lower levels of job satisfaction, organizational commitment, health and productivity (Anglim et al., 2019; Di Marco et al., 2016).

At the European level, the professional context reports the highest rates of discrimination against trans people, not only in job seeking processes but also in the workplace, as well as in unemployment rates (Baptista et al., 2023a, b; Beauregard et al., 2021). This discrimination is multidimensional and systemic, with higher levels of unemployment and poverty compared to cis people; institutional, as trans people work in organizations without antidiscrimination policies; and interpersonal, facing discrimination and/or harassment in the workplace (Baptista et al., 2023a; Nadal et al., 2014). The repercussions in the organizational environment can result in limited presence in workplaces, absenteeism, reduced productivity, low wages or unemployment, which can impact the well-being, health and safety of trans people (Davidson, 2016; Nadal et al., 2014).

Transphobia is the term assigned to the type of discrimination and prejudice specifically linked to gender identity, encompassing discomfort, fear, hatred, disgust and unfair treatment against all individuals who express non-normative gender identities and expressions (Hill, 2002; Hill and Willoughby, 2005). Associated with this concept is also the term genderism, which defines the discriminatory experiences of individuals who are perceived as belonging to a gender opposite to the one they identify with (Browne, 2004).

If the concept of diversity encompasses minority groups beyond the heterosexual and cis male (Baggio, 2017), trans individuals fall within that concept. In a professional context, in comparison to cis individuals, trans individuals are at a greater risk of facing barriers or disadvantages in the process of entering the job market (Waite, 2020). However, despite the responsibility of HR professionals with Recruitment and Selection roles to perform their tasks without considering gender identity or other personal life data that does not pertain to the job role, this is not always the case (Dias and Bernardineli, 2016).

In this sense, these professionals should not hold attitudes that guide their behavior patterns regarding organizational diversity and, specifically, gender diversity and trans people, that support or promote transphobia and/or genderism. Hence, the first hypothesis (H1) is proposed:

H1.

Negative attitudes towards organizational diversity have a positive correlation with negative attitudes towards trans people, as well as the levels of transphobia and/or genderism in a sample of HR professionals with Recruitment and Selection roles.

Additionally, if in accordance with the Intergroup Contact Theory, contact can effectively reduce prejudice (Allport, 1954; Pettigrew and Tropp, 2006), it is relevant to explore whether attitudes towards trans people and the corresponding levels of transphobia and/or genderism could be influenced by contact. Thus, the second hypothesis (H2) is proposed:

H2.

Attitudes towards trans people and the levels of transphobia and/or genderism of HR professionals with Recruitment and Selection roles are negatively related to contact.

Methodology

The present study aims to understand the extent to which attitudes towards organizational diversity are related to attitudes towards trans people, and the relationship of these variables with the levels of transphobia and/or genderism in a sample of HR professionals with Recruitment and Selection roles. Furthermore, using the Intergroup Contact Theory (Allport, 1954; Pettigrew, 1998), the study seeks to explore whether attitudes towards trans people and the corresponding levels of transphobia and/or genderism in the sample can be justified by contact with LGBTQIA + individuals, specifically trans individuals.

To test the research hypotheses, data were collected through the administration of a questionnaire comprised of self-constructed questions and previously validated scales, including: Attitudes Toward Diversity Scale (ATDS) (Montei et al., 1996), with a Cronbach's Alpha of α = 0.90; Attitudes Toward Transgendered Individuals (ATTI) Scale (Walch et al., 2012), with a Cronbach's Alpha of α = 0.96; and Genderism and Transphobia Scale (GTS) (Hill and Willoughby, 2005), with a Cronbach's Alpha of α = 0.95. Data analysis was conducted using SPSS Statistics 28 (IBM Corp., 2021), SPSS AMOS 26 (Arbuckle, 2019), Monte Carlo PCA for Parallel Analysis (Watkins, 2000) and Mplus 8.7 (Muthén, 2021).

The ATDS (α = 0.90) aims to measure attitudes toward diversity, where a high score reflects negative attitudes towards organizational diversity (Montei et al., 1996). The ATTI Scale (α = 0.96) is designed to assess stigma towards trans people, measuring cognitive evaluations and emotional reactions towards them, with higher scores indicating lower tolerance and acceptance of trans individuals (Walch et al., 2012). Lastly, the GTS (α = 0.95) aims to measure discrimination against trans people, where higher scores suggest higher levels of prejudice against them (Hill and Willoughby, 2005).

Additionally, apart from self-constructed questions and sociodemographic questions, questions were included to validate the participants' attention, given the total length of the questionnaire (with 97 questions). In this regard, two questions were included from the Organizational Citizenship Behaviors (OCB) Scale, namely: “To validate your attention, please select the number 2 as your answer” and “To validate your attention, please select the number 6 as your answer” (Podsakoff et al., 1990).

Data collection began on May 18, 2022, and concluded on July 7, 2022, upon reaching a significant number of responses, corresponding to a minimum value of five responses per item on the scale with the biggest dimension (Hair et al., 2006). Access to the sample was facilitated through messaging and sharing, with the LinkedIn being widely used to efficiently reach this type of professionals. Ethical procedures were ensured, and the confidentiality and anonymity of each participant were upheld, in accordance with the General Data Protection Regulation 2016/679 of the European Union, effective since May 25, 2018, in Portugal. Participation was voluntary and unpaid. The study also received a favorable opinion from the Ethics Committee of ISCSP (School of Social and Political Sciences) – University of Lisbon in January 2022, registered as decision CE-01-2022.

The sample consists of 184 HR professionals with Recruitment and Selection roles, with an average age of 31.97 years (SD = 7.851). The only inclusion criteria, aside from the specific professional area, were to be 18 years of age or older and work in Portugal. The sample includes 133 cis women (72.3%), 46 cis men (25%), one non-binary person (0.5%), two individuals who selected the option “other” (1.1%) and two individuals who selected the option “prefer not to disclose” (1.1%). In terms of educational level, the sample comprises four individuals (2.2%) with high school education, one person (0.5%) with post high-school education, three individuals (1.6%) with a technical professional higher education diploma, 96 individuals (52.2%) with a bachelor's degree and 80 individuals (43.5%) with a master's degree.

Results

To enhance the robustness of the used instruments, Confirmatory Factor Analysis (CFA) was conducted for each one, with appropriate fit indices considered to be Comparative Fit Index (CFI) values > 0.90 and Root Mean Square Error of Approximation (RMSEA) and Standardized Root Mean Square Residual (SRMR) values < 0.08 (Bentler and Bonett, 1980; Byrne, 2012; Hu and Bentler, 1999). The standard values in the literature considered in this study pertain to models with excellent fit, and values slightly below these should not be automatically excluded. Instead, a combination of values should be considered, rather than excluded based solely on one indicator falling slightly below the excellent threshold (Marsh et al., 2004). Firstly, to analyze the three-dimensional factor structure of the ATDS (Table 1), as proposed by Montei et al. (1996), multiple CFAs were conducted using the statistical software Mplus, version 8.7 (Muthén, 2021).

Due to the inadequacy of the various tested models, we chose to construct an Exploratory Structural Equation Modeling (ESEM). This type of model integrates both Confirmatory Factor Analyses (CFA) and Exploratory Factor Analyses (EFA), allowing not only the establishment of relationships between dimensions in the model but also the calculation of complex structures of variance error and the creation of bifactorial models, accessing the presence of cross-loadings (Zyl and Klooster, 2022). The execution of the ESEM (Table 2) confirmed the presence of a three-factor structure in a bi-factor configuration (with a general factor and three first-order factors), with acceptable model fit indices – although the CFI values were slightly lower than desired (χ2 = 559.528; RMSEA = 0.06; CFI = 0.87; SRMR = 0.05), affirming the construct validity of the instrument, in accordance with the recommendations of Marôco (2014).

Regarding the ATTI Scale, a Confirmatory Factor Analysis (CFA) was conducted (Table 3) to test the single-factor solution initially suggested by Walch et al. (2012). The obtained fit indices fell short of the desired levels, particularly concerning the value of the SRMR (0.09).

The inadequacy of the model suggests that a single-factor solution may not be the most appropriate. Consequently, an Exploratory Factor Analysis (EFA) was conducted to explore the presence of a more suitable factor structure. The 20 items of the ATTI Scale underwent a Principal Component Analysis (PCA) using SPSS Statistics 28 (IBM Corp., 2021). The results of the PCA revealed the presence of three components with eigenvalues exceeding 1 (ranging from 1.14 to 9.20), explaining 45.98%, 8.11% and 5.87% of the variance, respectively. However, an examination of the scree plot indicated a steep drop between the first and second components, which was reinforced by Parallel Analysis conducted using Monte Carlo PCA for Parallel Analysis (Watkins, 2000). This analysis suggested only three components with eigenvalues exceeding the corresponding criterion values for a random data matrix of the same size (i.e. 20 variables x 184 participants).

In this sense, extracting two components appeared to be the most appropriate, as the two-factor solution explained a total of 54.08% of the variance. Subsequently, a varimax (orthogonal) rotation was performed to aid the interpretation of the two components. This structure contrasts with the single-factor structure proposed by Walch et al. (2012). In this newly proposed two-factor structure, the first component (items 2, 3, 4, 6, 7, 8, 9, 11, 15, 18, 19 and 20) appears to measure the social acceptance of trans people, while the second component (items 1, 5, 10, 12, 13, 14, 16 and 17) appears to measure recognition and social contact with trans people (Table 4). Furthermore, the presence of some cross-loadings between the two factors suggested the existence of a general factor. As a result, two possible factorial models were analyzed: (1) a model with two correlated factors and 20 items; and (2) a bifactor model, where each item loads on its respective first-order factor and simultaneously on a general factor. The bifactor model demonstrated the best fit when compared to the other models, confirming the construct validity of the instrument in accordance with the recommendations of Marôco (2014). This factorial structure exhibits appropriate loadings on the factors.

Finally, the two-dimensional structure of the GTS, initially proposed by Hill and Willoughby (2005), underwent a Confirmatory Factor Analysis – CFA (Table 5). The fit indices indicated the suitability of the initial proposed model, confirming the construct validity of the scale in accordance with the recommendations of Marôco (2014).

Thus, the factorial solutions of the three applied instruments (ATDS, ATTI Scale and GTS) proved to be acceptable, according to the recommendations of Marôco (2014). Subsequently, it was relevant to analyze the results obtained from each of the instruments, as they exhibit good reliability and, therefore, assess the constructs they aim to evaluate (i.e. attitudes towards organizational diversity; attitudes towards trans people; and transphobia and genderism). In the following table (Table 6), it is possible to observe the means of the total sum of items for each scale, along with their respective standard deviations, variances and Cronbach's α values, indicating the instruments' reliability.

High scores on the ATDS reflect negative attitudes towards workplace diversity (Montei et al., 1996), high scores on the ATTI Scale reflect negative attitudes towards trans people (Walch et al., 2012) and high scores on the GTS reflect negative feelings and behaviors towards trans people, associated with transphobia and/or genderism (Hill and Willoughby, 2005). In the ATDS, the average response was 62.03 (SD = 11.563), indicating that it is not possible to conclude that there is a high level of negative attitudes towards organizational diversity in the sample. In the ATTI Scale, the average response was 31.67 (SD = 10.060), making it impossible to assert the presence of a high level of negative attitudes towards trans people in the sample. In question 12, which concerns trans individuals in the organizational context and their acceptance, it was evident that the sample would be comfortable working with trans people. Lastly, in the GTS, for the 32 items comprising the scale, none of the average responses were positive (above 3.50). The average was 50.34 (SD = 19.190), indicating a low result (below 112 points), and thus, it cannot be concluded that the sample demonstrates a high level of negative or discriminatory feelings and behaviors towards trans people, associated with transphobia and/or genderism.

Next, as all used instruments aim to measure similar quantitative variables related to diversity constructs, a correlation analysis was conducted between the variables (Table 7) using Pearson's correlation analysis (suitable for correlations between metric variables) to examine the convergent validity of the instruments (Field, 2009). The results suggest that the analyzed variables exhibit positive and strong correlations (ranging from 0 to 1), respectively: between ATDS and ATTI Scale (r = 0.647; p < 0.001), between ATDS and GTS (r = 0.601; p < 0.001), and between ATTI Scale and GTS (r = 0.828; p < 0.001). These findings suggest that all scales effectively measure similar constructs (Cohen, 1988).

Finally, analyzing the results while considering the contact variable, they suggest that, on average, individuals who have not had contact with trans people have a higher level of negative attitudes towards organizational diversity (M = 66.82, SD = 11.931), as well as a higher level of negative attitudes (M = 34.55, SD = 9.762) and a higher level of negative feelings and behaviors (M = 59.51, SD = 22.300) towards trans people, compared to those who have not had such contact (M = 58.33, SD = 10.927; M = 26.24, SD = 7.474; M = 47.21, SD = 15.791).

In the ATDS, the Levene's test for homogeneity of variances (Table 8) revealed a significance level above 0.05 (p = 0.335), indicating that the assumption of variance homogeneity is not violated. However, in the t-test analysis (Table 9), the significance value – below 0.05 (p=<0.001) – showed that there are statistically significant differences between the groups. In the ATTI Scale and GTS, the Levene's tests for homogeneity of variances revealed significance levels below 0.05 (p = 0.002; p = 0.004), indicating a violation of the assumption of variance homogeneity. In the t-test analyses, the significance values were again below 0.05 (p=<0.001; p=<0.001), demonstrating statistically significant differences between the groups. Therefore, the results from all applied instruments can be attributed to the contact with trans people.

Discussion

Trans people can feel that the organizational context is not a safe and welcoming place (Davis, 2009), creating an environment conducive to experiencing various discriminatory attitudes and behaviors. These can largely be attributed to social discomfort related to the notion that gender is not a fixed construct, but rather something fluid (Beauregard et al., 2016; Davis, 2009; Whittle and Turner, 2017). However, given that being trans is not related to one's job performance capabilities, it would be expected that both organizations and HR professionals aiming to recruit, select and retain qualified professionals would show support for all trans people. This support would create the potential for a healthy professional environment and the establishment of positive professional relationships (McFadden and Crowley-Henry, 2016). According to Beauregard et al. (2016), the creation of an inclusive climate is necessary to facilitate the participation of trans individuals in the job market. The challenges related to their recruitment, selection and retention can arise from organizations' ongoing failure to understand their specific needs.

Analyzing the results on the ATDS scale, which measures negative attitudes towards organizational diversity (Montei et al., 1996), the sample obtained an average score of 62.03 (SD = 11.563). While this score is not high, it also doesn't represent a low score that would indicate only positive attitudes towards diversity exist in that context. In the ATTI Scale, which measures negative attitudes towards trans people (Walch et al., 2012), the sample obtained an average score of 31.67 (SD = 10.060). This score is not high enough to conclude that the sample exhibits elevated levels of negative attitudes towards trans people. In the GTS scale, which measures negative feelings and behaviors towards trans people associated with transphobia and/or genderism (Hill and Willoughby, 2005), the sample's average score was 50.34 (SD = 19.190). This score also does not indicate a high level of negative feelings and behaviors towards trans people within the sample.

Through the Intergroup Contact Theory (Allport, 1954; Pettigrew, 1998), we sought to understand if the results could be related to the contact variable. Although the measure of contact does not indicate the frequency or intimacy of such contact, as suggested by Allport's hypothesis (1954), the responses from the sample seem to be aligned with the literature in the two scales related to trans individuals (the ATTI Scale and the GTS), where, on average, individuals who have not had contact with trans people exhibit higher levels of negative attitudes (M = 34.55, SD = 9.762) and higher levels of negative feelings and behaviors (M = 59.51, SD = 22.300) towards them, compared to those who have had such contact (M = 58.33, SD = 10.927; M = 26.24, SD = 7.474; M = 47.21, SD = 15.791). In both scales (the ATTI Scale and the GTS), the tests for homogeneity of variances revealed significances below 0.05 (p = 0.002; p = 0.004), violating the assumption of homogeneity of variance, and in the t-tests analyses, the significance values were once again below 0.05 (p=<0.001; p=<0.001), indicating statistically significant differences between the groups. Therefore, it is possible to assert that the results can be attributed to the variable of contact with trans people.

Regarding the limitations of this study, there are several that do not undermine its validity but rather offer directions for future investigations. Firstly, the sample size (N = 184) and gender distribution, where 72.3% of the sample is made up of women (n = 133). While the results cannot be generalized to the entire population of HR professionals with Recruitment and Selection related roles in Portugal, the use of a sample of 184 individuals, where 72.3% are women, to represent the Portuguese reality is questionable.

Secondly, another limitation is the extensive size of the questionnaire (97 questions), which reduce the response rate. Additionally, the study's topic, where misinformation, lack of knowledge, stereotypes, prejudices or preconceived ideas about trans people can serve as a limitation, creating uncertainty among the participants. Lastly, it is noted as a final limitation that the used instruments were originally developed in non-Portuguese populations, which could influence the gathered responses due to the potential impact of cultural values. It is also important to underline the potential presence of social desirability bias, a phenomenon that can distort results (Larson, 2018). This phenomenon occurs when socially desirable responses are given, either exaggerating or minimizing actual behaviors (Kwak et al., 2021). Despite efforts to mitigate its occurrence through the use of anonymous and confidential data collection methods, its absence cannot be guaranteed (Fernandes and Randall, 1992; Larson, 2018).

To address these limitations, it could be interesting in future research to repeat the application of the instruments in a larger sample (in an attempt to generalize the results), with a better gender distribution (aiming for a better understanding of the differences according to gender), or even with a sample across multiple countries (to compare different realities and try to understand cultural differences related to the topic).

Another suggestion would be to use self-developed scales. It is also important to note that, despite the valid and rich data collected, it was obtained using a single data collection method – the questionnaire survey (quantitative method). Therefore, in the future, it might be relevant to incorporate additional complementary data collection methods, such as interviews (qualitative method), to further substantiate the findings.

Conclusion

Based on the achieved results and their respective analyses, it was possible to conclude that the sample of 184 HR professionals with Recruitment and Selection roles does not seem to exhibit high levels of negative attitudes towards organizational diversity, high levels of negative attitudes towards trans people, or high levels of transphobic and/or genderist sentiments or behaviors. All these variables are correlated. However, extremely low values that could suggest the presence of only positive attitudes and behaviors were also not obtained. It is important to note that these results should be interpreted considering the composition of the sample – young and educated who were motivated to respond to this type of questionnaire.

Breaking down the study's objectives, by examining the relationship between attitudes towards organizational diversity and attitudes towards trans people, as well as correlating these responses with the levels of transphobia and/or genderism in the sample, it was possible to understand that the results suggest strong correlations among all variables, indicating that all the scales used measure similar constructs that are correlated with each other (Cohen, 1988). In this sense, the first hypothesis (H1) was confirmed – Negative attitudes towards organizational diversity have a positive correlation with negative attitudes towards trans people, as well as the levels of transphobia and/or genderism in a sample of HR professionals with Recruitment and Selection roles.

Through the Intergroup Contact Theory (Allport, 1954; Pettigrew, 1998), the aim was to understand whether the results could be related to the variable of contact with trans people. The scores on the two scales concerning trans people (the ATTI Scale and the GTS) aligned with the literature's suggestions, as individuals who had no contact demonstrated higher levels of negative attitudes, feelings and behaviors (including transphobia and/or genderism), compared to those who had such contact. Thus, the second hypothesis (H2) was confirmed – Attitudes towards trans people and the levels of transphobia and/or genderism of HR professionals with Recruitment and Selection roles are negatively related to contact; being possible to conclude that contact with trans people acts as a predictor of negative attitudes and associated behaviors. This pattern has been highlighted in prior studies (e.g. Costa and Davies, 2012; Hill and Willoughby, 2005; or Willoughby et al., 2010).

Finally, it is relevant to question: if, according to the results on this sample of professionals, elevated levels of negative attitudes towards organizational diversity and specifically towards trans people are not demonstrated, what might lead to the trans population not being integrated into the job market in the same way as the cis population, experiencing difficulties in obtaining and/or maintaining employment due to discrimination based on their gender identity and/or expression and having unemployment rates approximately twice as high as cis individuals (Davidson, 2016; Grant et al., 2011).

Regarding the practical applicability of the results, it is important that HR professionals and the organizations they work for effectively integrate trans people, showing openness to receive them, by developing organizational programs and policies specifically focused on achieving this goal, making trans people feel welcome in the job market. The impact of HR policies on the integration of trans individuals into the labor market is evident in the existence of six types of policies: (1) antidiscrimination policies – explicitly prohibiting discrimination (or harassment) based on gender identity and/or expression, enhancing the safety and sense of protection for trans individuals in the workplace; (2) policies focused on transition or gender affirmation processes – outlining clear guidelines, offering support to individuals undergoing or desiring to undergo these processes, and ensuring specific health needs (including coverage by health insurance) related to hormonal therapy or surgeries, when applicable; (3) policies focused on inclusive or gender-neutral facilities such as locker rooms or bathrooms, creating an inclusive and safe environment for trans individuals; (4) policies focused on training all individuals in organizations to increase sensitivity and awareness of the topic; (5) policies that promote an inclusive work environment, which may include support groups, diversity and inclusion programs, open communication channels, inclusive language usage, among others; and (6) inclusive attraction, recruitment and selection practices for trans individuals, including inclusive gender-related communications, advertisements and recruitment processes (e.g. Baggio, 2017; Beauregard et al., 2016; Beauregard et al., 2021; Ozturk and Tatli, 2016).

However, it is also pertinent to question whether discrimination is not so deeply rooted in social and structural aspects that it is not solely present in the moment of entering the job market and in the recruitment and selection processes involving HR professionals (Baptista et al., 2023b). It may be a much more complex and multidimensional issue, relevant to address.

Table 1

Index of fit for the model initially proposed by Montei et al. (1996) - ATDS

Factorial models	χ2	df	CFI	RMSEA	RMSEA 90%CI		SRMR
Bi-factor	1191.39	370	0.84	0.11	0.10	0.12	0.12
Hierarchical	2384.58	406	0.61	0.16	0.16	0.17	0.19
3 correlated factors	1393.12	405	0.80	0.12	0.11	0.12	0.13

Source(s): Authors' own work

Table 2

Index of fit (ATDS)

Factorial model	χ2	df	CFI	RMSEA	RMSEA 90%CI		SRMR
Bi-factor	559.528	321	0.87	0.06	0.06	0.07	0.05

Source(s): Authors' own work

Table 3

Index of fit for the model initially proposed by Walch et al. (2012) – ATTI Scale

Factorial model	χ2	df	CFI	RMSEA	RMSEA 90%CI		SRMR
1 factor	308.80	189	0.98	0.08	0.07	0.10	0.09

Source(s): Authors' own work

Table 4

Index of fit (ATTI Scale)

Factorial models	χ2	df	CFI	RMSEA	RMSEA 90%CI		SRMR
2 correlated factors	401.91	169	0.98	0.09	0.08	0.10	0.06
Bi-factor	273.09	150	0.99	0.07	0.05	0.08	0.05

Source(s): Authors' own work

Table 5

Index of fit for the model initially proposed by Hill and Willoughby (2005) - GTS

Factorial models	χ2	df	CFI	RMSEA	RMSEA 90%CI		SRMR
2 factors	879.71	463	0.97	0.07	0.06	0.08	0.08

Source(s): Authors' own work

Table 6

Descriptive statistics (total)

Scale	N	α	RANGE	MIN	MAX	M	SD	VAR
ATDS	184	0.85	61	34	95	62.03	11.563	133.699
ATTI	184	0.92	41	20	61	31.67	10.060	101.205
GTS	184	0.93	94	31	125	50.34	19.190	368.246

Source(s): Authors' own work

Table 7

Pearson's correlation analysis

Scale		ATDS	ATTI Scale	GTS
ATDS	Pearson's correlation	1	0.647*	0.601*
	Sig (2-tailed)		<0.001	<0.001
	N	184	184	184
ATTI Scale	Pearson's correlation	0.647*	1	0.828*
	Sig (2-tailed)	<0.001		<0.001
	N	184	184	184
GTS	Pearson's correlation	601*	0.828*	1
	Sig (2-tailed)	<0.001	<0.001
	N	184	184	184

Note(s): * The correlation is significant at the 0.01 level (2-tailed)

Source(s): Authors' own work

Table 8

Levene's test for homogeneity of variances (contact)

Scales	N	M	SD	F	Sig
ATDS				0.936	0.335
With contact	76	58.33	10.927
Without contact	51	66.82	11.931
ATTI Scale				10.404	0.002
With contact	76	26.24	7.474
Without contact	51	34.55	9.762
GTS				8.752	0.004
With contact	76	47.21	15.791
Without contact	51	59.51	22.300

Source(s): Authors' own work

Table 9

t-test analysis (contact)

Scales	t	df	Sig. (2-tailed)	Mean difference	SD difference	95% Confidence Interval of the difference
						MIN	MAX
ATDS
Equal variances assumed	−4.139	125	<0.001	−8.50	2.053	−12.557	−4.432
Equal variances not assumed	−4.067	101	<0.001	−8.50	2.089	−12.638	−4.351
ATTI Scale
Equal variances assumed	−5.426	125	<0.001	−8.31	1.532	−11.344	−5.280
Equal variances not assumed	−5.152	88	<0.001	−8.31	1.614	−11.518	−5.106
GTS
Equal variances assumed	−3.640	125	<0.001	−12.30	3.379	−18.987	5.611
Equal variances not assumed	−3.407	83	0.001	−12.30	3.610	−19.479	−5.119

Source(s): Authors' own work

This paper is classified under the “Corporate Social Responsibility” management area.

Data availability statement: The data that support the findings of this study are available from the corresponding author [J.B.], upon reasonable request.