Gender in initial teacher education: entry patterns, intersectionality and a dialectic rationale for diverse masculinities in schooling

ABSTRACT This paper contributes to debates about gender and diversity in teaching. Exploring the underrepresentation of males, socially constructed as a privileged group, is a complex task for researchers working within a social justice framework. Drawing on a national dataset (N = 5,627), we explore entry patterns to primary and post-primary Initial Teacher Education (ITE) programmes, socio-demographic backgrounds and career motivations of male and female ITE applicants and entrants in Ireland. Findings point to possible reasons for the underrepresentation of males in primary teaching, including competitive performance-based selection systems, Irish language competency requirements and a predominantly denominational school system. An intersectional analysis demonstrates the diversity which exists among male ITE applicants. New insights into the complexities of male positionalities in teaching can enhance dialectic debates about reasons for and implications of the underrepresentation of male teachers as well as the transformative potential of diverse and ‘caring’ masculinities for schools and society.


Introduction
Female teachers predominate in primary and post-primary schools in most Western countries. Calls for increasing the proportion of men in the teaching profession have resulted in many failed policy initiatives and attracted criticism from feminist and profeminist scholars (Skelton 2007(Skelton , 2009Vibert et al. 2004). While research exploring equity of access to teaching for other underrepresented groups, including those from minority and disadvantaged backgrounds, has received positive attention and increasingly also some policy support , gender remains an anomaly in teacher diversity research. In this paper, we re-establish the importance of considering male patterns of representation in teaching. As well as considering possible reasons for male underrepresentation, we highlight the importance of an intersectional analysis of the socio-demographic characteristics of male ITE applicants.
Our study forms part of the wider Diversity in Initial Teacher Education (DITE) in Ireland project, funded by the Irish Research Council. DITE explores compositions of primary and post-primary teacher education applicants and entrants across state-funded initial teacher education (ITE) providers in Ireland. To date, the project team has examined the participation, career motivations and experiences of various under-represented groups, including people from minority ethnic Heinz 2015, 2016) and/or minorityreligious (Heinz, Davison, and Keane 2018) and/or lower socio-economic backgrounds (Heinz, Keane, and Foley 2017;Lynch 2018, 2020), people with disabilities (Keane, Heinz, and Eaton 2018) and/or members of the LGBTQ+ community (Heinz, Keane, and Davison 2017). Analyses have confirmed the homogeneity of the Irish teaching body which has been predominantly White, Irish, and female, with significant under-representation from a range of minority and/or socially disadvantaged groups.
The DITE project has been situated in, firstly, the 'equity of access' perspective focusing on the benefits to individuals from under-represented groups who may encounter barriers to progression into the teaching profession. We have, secondly, emphasised the 'equity through access' rationale, i.e. the benefits of a diverse teaching profession for students, schools and wider society (Childs et al. 2011;Keane and Heinz 2015). As such, we are philosophically aligned with the 'teacher diversity agenda' which emphasises the importance of teacher diversity in working towards a more just and equitable societyand education system . Through our research we explore ways of attracting, facilitating entry for, and supporting individuals from under-represented groups on their journey into and through the teaching profession.
While the under-representation of males in ITE has been noted in our previous work, to date it has not received the same attention as other characteristics. Indeed, addressing the under-representation of males, socially constructed as a privileged and powerful group (Connell 1987(Connell , 2003, is a complex task for teacher diversity researchers whose work is framed by social justice concerns. Traditionally, given their societal positionality and construction, the under-representation of males in teaching has not been regarded as an 'equity' issue; instead it has been considered more of a non-constrained choice/ decision in relation to other careers, especially in terms of primary teaching, rather than being due to any systemic barriers. In this context, the discourse has constructed the relative under-representation of men as essentially unproblematic, with Martino and Rezai-Rashti (2012) arguing for a focus on the recruitment of good teachers, irrespective of their gender. While the gender equality lens has commonly and justifiably focused on women, and while many simplistic arguments calling for redressing the gender imbalance in teaching have rightly been refuted, a significant body of research has demonstrated that gender is of particular significance in the opportunities, experiences and expressions of being a teacher (O'Keeffe 2016;Skelton 2009).
Our study needs to be contextualised in the context of schooling and teacher education in the Republic of Ireland. There are 3,420 primary and 722 second-level schools, employing 67,403 teachers (DES (Department of Education and Skills) 2019). Ireland has the second highest number of single-sex schools in the EU (120) (Eurydice 2010). Ninty per cent of primary schools and over half of second-level schools are Catholic, guided by an explicit Catholic ethos and, in the case of primary schools, implementing daily religious confessional instruction (Coolahan, Hussey, and Kilfeather 2012). Irish language education is compulsory with exemptions available only for students with certain learning difficulties or migrant students. Despite the forced shift to English during and following British occupation of Ireland, Irish is the country's first official language and government policy aims to increase its use (Government of Ireland 2010).
Teachers need to be registered with, and ITE programmes are accredited by, the Irish Teaching Council. Undergraduate programmes are four years, and postgraduate programmes are two years, in duration, incorporating academic, practical and research elements. Entry to ITE is competitive; selection for undergraduate programmes is based on applicants' Leaving Certificate (final school examinations) performance, and applicants have to reach specified minimum standards in Irish, English and Maths, with a high standard (recently raised) for Irish (DES (Department of Education and Skills) 2020). At postgraduate level, points for undergraduate degree performance are combined with points for relevant experience and/or further qualifications (Heinz 2008(Heinz , 2013a.
In this paper, we re-establish the importance of considering male patterns of representation in teaching. To do this, we draw on our analysis of the national DITE dataset describing ITE applicant and entrant populations across all state-funded ITE programmes for primary and post-primary teachers in Ireland. In our analysis we focus on the following research questions: (1) What is the gender composition of applicants and entrants to the various ITE programmes (undergraduate and postgraduate, primary and post-primary) in Ireland? What are the success rates of male compared to female ITE applicants in current selection systems? (2) What are similarities and differences with regard to male and female ITE applicants' and entrants' age, ethnicity/ies, nationality/ies, sexuality/ies, socio-economic group/social class backgrounds, dis/ability, religion, education background and career motivations? (3) How does gender intersect with other aspects of identity?

Gender imbalances in the teaching profession: trends and explanations
Female teachers predominate in primary and post-primary schools in most Western countries. This predominance is most pronounced at primary level where women are in the great majority across Europe (87%) and most other upper middle (72%) and highincome countries (82%) (UNESCO Statistics 2020). In Ireland, the proportion of females in primary teaching increased steadily from just over half (57%) at the beginning of the 20 th century, to almost three quarters (73%) in 1979 (Greaney, Burke, and McCann 1987), and further to 85% in 2019 (DES 2020). In contrast to this widespread 'feminisation' 1 of teaching across Western countries, women remain under-represented in primary teaching in the least developed countries (for example 16% in Togo, 27% in Sierra Leone, 41% on average in low income countries (UNESCO 2020)).
At post-primary level, the percentage of females among teaching bodies is significantly lower than at primary level. On average, 69% of post-primary teachers in Europe are female (UNESCO Statistics 2020), as is the case in Ireland (DES (Department of Education and Skills) 2019). In contrast, women account for only 27% of the post-primary teaching body in the least developed countries, an even smaller proportion than that observed at primary level (UNESCO Statistics 2020). Indeed, examining the different proportions of women in teaching across the world, UNESCO suggest that female representation in teaching bodies can be taken as an indicator of economic development in different regions Ibid (UNESCO Statistics 2020). Despite the predominance of women among teacher cohorts in Ireland and other developed countries, school administration and management have, for a long time, been largely led by men, with proportions of women principals significantly below their representation among the teaching force (Drudy and Lynch 1993;Reynolds 2012).
Australian research offers three explanations for the increasing 'feminisation' of the teaching body: Firstly, teaching salaries overall are lower relative to other professions, especially for men. Secondly, cultural factors tend to stereotype teaching as 'women's work', especially at primary level (Sargent 2001;Skelton 2009), and, thirdly, a fear of being wrongly accused of child abuse can deter males entering teaching, particularly at primary level (Davison 2007; MCEETYA (Ministerial Council on Education, Employment, Training and Youth Affairs) 2003). Research in the Irish context also confirmed that students' thirdlevel course choices followed gender stereotypes (Drudy et al. 2005). Drudy et al. (2005) suggest that the gender compositions of teaching populations are influenced by the level of economic development and degree of urbanisation of regions and countries; by the economic policies pursued by state administrators, especially educational administrators; and by gendered beliefs about the nature of women and men.

Too few men? The rationale for gender diversity in teaching
The question of whether the 'feminisation' of the teaching profession should be a matter of concern has been hotly debated for quite some time. Calls for more male teachers and male teacher recruitment initiatives have met intense critique by feminist/pro-feminist researchers (in Finland, the USA, Australia, New Zealand, the UK and Ireland) who have refuted many of the flawed assumptions underpinning the perceived need for more male teachers (e.g. Davison 2007;Cushman 2012;Frank et al. 2003;Mills, Martino, and Lingard 2004;McDonald 2020;Skelton 2007). Three core flawed arguments underpin the various strands: Firstly, the idea that male teachers are needed to 'redress' boys' perceived educational disadvantage (Weaver-Hightower 2003) is contradicted by research indicating that teacher gender does not directly affect students' academic outcomes (Cho 2012;Skelton 2003). Indeed, given that research has shown that females outperform males in higher education and also in teacher education (Drudy 2006;Heinz 2013b), a counterargument could point to women's particular suitability to teaching based on their stronger academic performance. Secondly, the importance of male teachers has often been linked to the need for boys to have access to positive male role models in school. The role model argument has been contested on the basis of its ambiguity, on the one hand, and the resulting essentialisation of masculinity and masculine characteristics which are portrayed as somehow threatened by the 'feminised' education system on the other hand (Martino 2008(Martino , 2009Martino and Rezai-Rashti 2012). Thirdly, the notion that schools need male teachers who can act as father figures for fatherless children lacks empirical backing and appears incoherent when the temporal nature of student-teacher relationships, and various cultural norms, are taken into account (cf. McGrath et al. 2020;Hunter 2019;McDonald 2020).
It is clear that the question of whether, and if so why, a gender balance in the teaching profession should be a policy priority requires significant further debate and research. The issue of gender diversity in teaching is particularly problematic as it concerns the underrepresentation of a traditionally power-holding group. As such, the underrepresentation of males in the teaching profession does not fit easily in the teacher diversity context where underrepresentation is, usually, conceptualised within contexts of changing population compositions, disadvantage, and inequities. Calls for efforts to diversify the teaching profession and to increase participation of individuals from minority ethnic backgrounds, lower socio-economic groups, as well as LGBTQ+ people and those with disabilities, have been firmly rooted within a social justice conceptual framework, within which calls for more male teachers do not appear to fit.
Nevertheless, some research has identified less contested reasons why male teachers are needed in schools, and why ever-increasing levels of women in teaching may be problematic. For example, Drudy et al. (2005, 154) raised three concerns; the issue of 'equality in the labour market', the socialisation of pupils, and the hidden curriculum of schooling with regard to occupational hierarchies characterising the educational system. Based on the findings of their study, Drudy et al. (2005, 155) recommend that 'policy makers, legislators, administrators and educators should adopt and emphasise caring as a core human value, one which is as central for boys and men as it is for girls and women.' The recent work of McGrath et al. (2020) expands the reasons offered by the Irish study. Taking an interdisciplinary and intracultural approach, McGrath et al. analyse the need for male teachers at four levels: At the child level, they explore the role of schools in the development of children's gendered and sexuality-related identities and understandings. Drawing on social cognitive theories, the authors theorise that 'limited observation of male teachers may result in children's erroneous generalisation of all teacher characteristics as female-specific traits, perpetuating the view that women are better suited to the teaching profession' (ibid. 6). They argue that children's gender knowledge is extended upon observing men also demonstrating both feminine and masculine traits. Furthermore, students' experiences of positive and diverse male gender representations, and their observation of male teachers working with female school leaders, may help to reposition care as important for all genders and challenge beliefs about male dominance. At the classroom level, given the research demonstrating that same-sex dyadic (student-teacher) relationships may be closer (cf. Spilt, Koomen, and Jak 2012) and that male teachers may be more encouraging and supportive of male students (Mullola et al. 2012;Bosacki, Woods and Coplan 2015), McGrath et al. raise the possibility that male students may benefit from relationships with male teachers. At the organisational level, they emphasise the benefits of a representative workforce on interactional experiences, policy development and outcomes whilst recognising issues of token visibility, polarisation and assimilation. Finally, at societal level, greater male teacher representation can help to challenge rigid (social) constructions of masculinities (and femininities) and stereotypical gender roles. Beyond workforce-'balance' justifications, McGrath et al. argue that the participation of diverse men in teaching may help to 'promote acceptance of alternative masculinities while destigmatising the participation of men in the lives of young children ' (2020, 9).
The analyses and discussion provided in the next sections explore gender patterns in ITE in Ireland as well as intersections between gender, other socio-demographic backgrounds and career motivations.

Methodology
DITE is the first national study in Ireland collecting data from applicants and entrants to all state-funded undergraduate and postgraduate, primary and post-primary, ITE programmes. Internationally, it is the first study to compare the profiles of applicants and entrants within and across a wide range of programmes. Three data sources were analysed to explore gender diversity in ITE programmes in Ireland: (1) the Central Application Office's ITE data set (N = 4,070), comprising data from all applicants to undergraduate ITE programmes in Ireland; (2) the Postgraduate Application Centre's data set (N = 1,557), comprising data from all applicants to postgraduate post-primary ITE programmes; and (3) the DITE survey data set (N = 2,240), comprising data from an anonymous voluntary survey administered to all undergraduate and postgraduate ITE applicants (prior to selection and programme commencement).
To ensure anonymity of survey participants, we used a hashed algorithm system allowing us to track applicants to entrants during the process of merging survey data with anonymised application outcome files. The DITE questionnaire contains 79 questions, a mixture of closed-and open-ended items. Response rates ranged from 43% for undergraduate primary applicants to 47% of postgraduate post-primary applicants. Response rates for all entrant groups were above 50%. The most important variables analysed for this paper include: • Biological sex (male, female, intersex, other, please specify); • Sex (male, female, intersex, other, please specify) • Gender identity (female/woman', 'male/man', 'transgender', 'a gender not listed here, please specify); With regard to exploring sexuality, it is recognised that the use of a single open-ended question presents certain limitations, for example, the possibility of a conflation of various different, yet highly correlated, domains of sexual orientation. The inclusion of sexuality as a diversity dimension in the DITE survey followed significant negotiation, resulting in the final, open-ended, question which satisfied all stakeholders while simultaneously allowing participants to freely express their sexual orientation and avoid sexual identity labels, which can be perceived to oversimplify the nature and experience of same-sex sexuality (Diamond 2002).
The empirically validated and internationally widely used 2 'FIT-Choice' (Factors Influencing Teaching Choice) framework (Watt and Richardson 2007) was used to measure the career motivations of male and female respondents. The scale measures: i) antecedent 'socialisation influences'; ii) 'task perceptions'; iii) 'self-perceptions'; iv) 'values'; and v) 'fallback career' (see Table 6). All factors are measured by multipleitem indicators with response options from 1 (not at all important) to 7 (extremely important). Cronbach alpha values ranged from 0.62 for 'fallback career' to 0.93 for 'social utility value', with most values around 0.80 indicating good levels of internal consistency of the scale.
The quantitative data were analysed using the Statistical Package for Social Science (SPSS) for Windows (version 26) to perform frequency tests, multiple response tables, chisquare tests of independence, and t-tests to explore significant differences among applicant and entrant cohorts, as well as between male and female respondents. Mean scores and standard deviations were calculated for all FIT-Choice items, factors and higher order factors. Independent sample t-tests and, where equal variance was not verified, Welch's t-test were implemented to investigate potential differences in ratings between male and female respondents.
Qualitative data were analysed thematically, grouping similar descriptions (for example attitudes towards teaching religion, main reasons for choosing a teaching career) into categories (Heinz, Davison, and Keane 2018;Keane and Heinz 2016).
The following abbreviations are used throughout the Findings section for the various ITE programmes: UG P (Undergraduate Primary); UG PP (Undergraduate Post-primary); PG PP (Postgraduate Post-primary). In the Tables presented, the following abbreviations are used: App (Applicants), Ent (Entrants), and Non-Ent (Non-Entrants).

Sex/gender composition of ITE cohorts
The participation rates of males differ significantly for primary and post-primary ITE cohorts (see Table 1). Representation of males among UG P ITE applicant and entrant cohorts (21.4% and 17.6% respectively) is less than half of that in UG PP (56.6% and 45.4%) and PG PP cohorts (37.6% and 39.5%). DITE survey findings indicate a small number of applicants and entrants identifying as intersex.
Analysis of the gender composition in different UG PP ITE programmes revealed significant differences in male and female participation rates with the proportion of males ranging from 100% (N = 18) in a Design Graphics and Construction programme, to 0% in three programmes for Home Economics (N = 59). Practical/technology subjects attracted significantly more males (N = 116) than females (N = 5). However, Physical Education programmes enrolled similar numbers of male (N = 56) and female (N = 54) students, and Science education programmes attracted more females (N = 86) than males (N = 32).

Sex and prior academic performance
Independent sample t-tests indicated significant differences between the prior academic achievement of male and female UG ITE applicants as measured by their Leaving Certificate (final examinations) points:  Males applying for PG programmes indicated significantly lower levels in their Irish proficiency compared to females. No significant gender differences in self-reported proficiency in Irish were noted for applicants and entrants to primary ITE programmes.

Sex and age
Analysis showed that male UG PP applicants and entrants were significantly older than their female counterparts: No significant differences were detected with regard to males' and females' mean age in UG P and PG PP cohorts.

Sex, socio-economic group, and social class
Overall, p values indicated no significant differences between the socio-economic group or social class backgrounds of male compared to female applicants and/or entrants in ITE cohorts. However, z-scores indicated some significant differences in a small number of categories: • UG P Entrants: A significantly higher proportion of male (20%) than female (10.2%) UG P entrants came from farming backgrounds (z = 2.1). • UG PP Entrants: A significantly higher proportion of male (32.4%) compared to female (17.9%) UG PP entrants came from manual skilled backgrounds (z = 2.4). • PG PP Entrants: A significantly higher proportion of male (26.5%) compared to female (15.8%) PG PP entrants came from an Employer and Manager backgrounds (z = 2.2).

Sex and sexual orientation
Sexual orientation was explored with PG PP applicants. Overall, the 587 responses to the open-ended item were categorised as: i) 'heterosexual' (N = 564), and ii) 'nonheterosexual' (N = 23). A small number (N = 8) of ambiguous responses could not be categorised. It is important to note that a significant number of respondents (N = 137) did not respond to this question.
Of the 23 respondents identifying as non-heterosexual, eighteen were male and five were female. Chi-square tests confirmed that there were significantly more male nonheterosexual than female non-heterosexual individuals applying for PG PP programmes (X 2 (1, n = 559) = 14.576, p = .000). Table 3 details the religious affiliations of male and female ITE applicants and entrants to the various ITE programmes. The stated affiliation of the large majority of male and female applicants and entrants was Roman Catholicism. Chi-square tests indicated significantly higher proportions of males compared to females stating that they had 'no religion' for the UG P entrant (z = 2.4) as well as the PG PP applicant (z = 4.0) and entrant (z = 2.3) cohorts.

Sex and religion
For UG P ITE applicants, we explored religious attendance and religiosity in greater detail as religion is particularly relevant in the predominantly denominational Irish primary school sector. Our analysis of respondents' religiosity indicated that a significantly higher proportion of male than female UG P applicants (34.3% compared to 24.5%) consider themselves 'not a religious person/a convinced atheist' 3 , X 2 (2, N = 688) = 6.36, p = .042, Cramer's (V= . 096).
An open-ended item explored UG P ITE applicants' perspectives on teaching religion as part of their teaching role. Following a thematic analysis of the qualitative data, 4 responses were categorised as: 'Agreeing', 'Complying', 'Wanting to teach all religions', 'Mixed feelings', 'Disagreeing', 'Indifferent' and 'Don't know'. Chi-square tests indicated significant differences between the feelings of male compared to female UG P applicants, X 2 (4, N = 508) = 12.54, p= .014, Cramer's (V= 0.157), 5 with more males than females 'complying' rather than 'agreeing' with teaching religion (Table 4).

Sex and ethnicity/ies, nationality/ies, and dis/ability
Chi-square tests found no significant relationship between sex and ethnicity/ies, sex and nationality/ies, or sex and dis/ability/ies for any of the ITE cohorts.  Table 5 illustrates intersections between male sex and other socio-demographic variables, illuminating the complexities of socio-demographic backgrounds and their interpretations of apparent privilege and disadvantage. The analysis illustrates high proportions of White Irish (settled) male applicants coming from lower social class backgrounds (including semi-skilled and unskilled). While the numbers of minority ethnic male (as well as female) applicants have been very small in Ireland Heinz 2016, Heinz and, our sample shows that there are some males with non-White or mixed ethnicities. Furthermore, a smaller number are positioned at the intersection between male-minority-ethnic-lower social class. We also note a significant number of male-nonheterosexual applicants who are evenly split between higher and lower social class groups and males with disabilities in all cohorts.

Career motivations of male and female ITE applicants
The survey included one open-ended question asking respondents to state their main reasons for choosing a teaching career, followed by the validated FIT-Choice Scale (see Table 6 for results of our t-test analyses comparing mean values for male and female primary and post-primary applicants). T-test results showed significantly different ratings for a number of factors for females and males with females rating 'intrinsic' and 'social utility'-type motivations as well as their 'satisfaction with their career choice' significantly higher than their male counterparts. Most of the statistically significant differences had small to medium effect sizes (see η 2 values in Table 6) with the only exception of 'working with children and adolescents' indicating a large effect size.
However, an analysis of ranking patterns of motivational factors indicates that, overall, career motivations are very similar for male and female applicants with both, males and females, rating their enjoyment of teaching, their perceived teaching ability, as well as the social contribution (or 'altruistic') value of the teaching career highly and, importantly, significantly higher than various extrinsic ('personal utility') factors related to job security/ transferability and/or time for family. A strong interest in their subject area was also particularly notable among male and female post-primary respondents. Furthermore, both, males and females, were very unlikely to consider teaching as a fallback career, with mean scores for this factor well below the midpoint. Analysis of the qualitative data (N = 1,188) confirmed this pattern of similarity demonstrating that the vast majority of male (N = 321) and female (N = 867) applicants across the various ITE programmes and socio-demographic groups were predominantly motivated by intrinsic and altruistic-type reasons and/or their desire to work with and help children and/or young people. 'Making a difference', 'inspiring/helping children and young people', 'enjoying being/working with children', 'personal positive learning/prior teaching/ voluntary youth work experiences', and/or a 'belief in teaching ability' featured strongly across the responses of all cohorts. There was a noticeably stronger emphasis on 'enjoying' and 'wanting to help children' and/or 'being good with children' in responses from both male (N = 96) and female (N = 514) primary applicants compared to post-primary applicants, as illustrated in the quotes below: Notably fewer male primary (11%) and post-primary (11%) applicants stated (in their open-ended responses) that they 'always' wanted to teach, or work with children/ young people, compared to female primary (21%) and post-primary (14%) applicants.

Summary of findings
We have provided a detailed analysis of gender entry patterns showing that participation rates of males are significantly lower in primary (18%) compared to post-primary (40-45%) ITE programmes. At post-primary level, we noted some significant discipline-specific imbalances, especially in practical/Engineering subjects and Home Economics, which may, in turn, impact gender representation in different types of schools where these subjects may or may not be offered (for example in single-sex schools). The sociodemographic profiles as well as career motivations of male and female applicants and entrants were very similar.
With regard to differences, we noted significantly lower prior academic performance and lower confidence in competency in the Irish language among male applicants in various ITE cohorts. More males than females indicated that they had 'no religion', did 'not consider themselves a religious person' and, among the primary cohort, 'complied with' or 'disagreed' rather than 'supported' the teaching of religion in primary schools. Finally, more male than female respondents to the post-primary questionnaire identified as non-heterosexual.
The intersectional analysis demonstrated high proportions of White Irish (settled) male applicants coming from lower social class backgrounds, a small number of males with non-white or mixed ethnicities, and an even smaller number positioned at the intersection between male-minority-ethnic-lower social class. A significant number of nonheterosexual males, evenly distributed across social class and disability/non-disability groups, was also noted.

Considering reasons for the under-representation of males in primary ITE
Critical examination of our data points to a number of possible reasons for the underrepresentation of males in the primary teaching body in Ireland while also exposing tensions between competing agendas. While the points-based Leaving Certificate selection system is often assumed to be fair and objective and taken as an indicator of a highquality teaching body in Ireland (Coolahan 2003;Heinz 2013a), the lower performance of male applicants at school results in them being less likely to secure a place in primary ITE. Given that national and international data show lower performance of males compared to females at school level (OECD 2015), it is possible that more males who might be interested in primary teaching refrain from applying believing that they will not get adequate points and/or do not have a sufficiently high level of competency in the Irish language. While the recent increase in the required standard of Irish (DES (Department of Education and Skills) 2020) may support the goals of the Irish language strategy, it may simultaneously result in further decreasing numbers of male applicants (and entrants) to primary ITE in Ireland in the future.
Religiosity has been declining in Ireland over the last decade (Heinz, Davison, and Keane 2018) and our data suggest that males may be less likely to 'fit into' a predominantly Catholic school system where school ethos is interlinked with religious values and practices and where religious instruction is part of a teacher's job. We have previously called for consideration of the systemic barriers faced by minority ethnic and/ or religious, LGBTQ+, as well as disabled individuals who may be less likely to be successful in primary ITE selection systems, and/or not apply in the first place, and/or face significant challenges as primary teachers due to tensions between their personal and professional lives (Heinz, Keane, and Davison 2017;Keane and Heinz 2016;Neary 2013). The analysis presented in this paper shows that the previously identified barriers for underrepresented groups may also be reasons for low participation rates of males in primary teaching. However, given that males are, in terms of their societal positioning, not considered a disadvantaged group, is a call for male teachers justified? What are the benefits of having more male teachers in schools?

Caring masculinities and the importance of an intersectional analysis
Our findings have indicated that male and female ITE applicants and entrants are motivated by very similar (intrinsic, altruistic and child-centred) factors. While the call for more male role models is often vague or problematic, our data supports the rationale for the importance of male teachers on the grounds of the nature of their career motivations -being consciously concerned with care.
Building on Elliott's (2020) feminist framework of 'caring masculinities' we see male teachers as uniquely positioned to engage in gender equality 'because doing care work requires men to resist hegemonic masculinity and to adopt values and characteristics of care that are antithetical to dominance' (254). As was evident in many of the qualitative responses of our male primary ITE respondents, care work can provide 'nourishing and satisfying models of masculinities for men [. . .] enriching their lives in myriad ways' (253). With greater recognition of the benefits of the positive experiences of care, for both masculine and feminine identities, teacher education and wider society will likely be more effective in attracting males to teaching than paradoxical calls for stereotypical male role models. A greater focus on diverse and caring masculinities can, furthermore, temper the contradiction of broadening diversity by being inclusive of a non-equity group if we embrace diverse masculinities rather than assume that hegemonic masculinity is required to address the 'feminisation' in education.
Our intersectional analysis has illustrated the complexities of, and tensions within, discourses of gender, power, privilege and dis/advantage. In particular, we noted a high proportion of White Irish (settled) males from lower social class backgrounds, and the presence of non-heterosexual men, both of whom are generally considered to be 'subordinated' or 'marginalised' masculinities (Elliott 2020). More nuanced representations of the multiple points of intersection between gender, privilege, culture and sexuality are needed to challenge the myth of a single type of male role model.
Critical masculinity scholarship has long advanced the need to look at masculinities as plural and diverse to explore contradictions and tensions between men (Connell 1995;Davison 2007), to examine subordinated masculinities and marginalised men (Bereswill and Neuber 2011), and to break down essentialist notions of gender (Butler 2004;Davison 2007;Fuss 1989Fuss , 1995. Our findings support a rationale for more men in primary education, not as simplistic role models for boys, but rather to demonstrate the diversity amongst men and the possibility of caring masculinities to positively impact children, teaching, schools and society (McGrath et al., 2020). Connell (2003, 29) argues that 'masculinity politics should be thought of as dialectic, not an expression of fixed identities and positions'. The persistent call for more male teachers has, so far, rarely explicitly named the need for diverse masculinities. Elliott's (2020) framework of 'open' and 'closed' masculinities in the 'margin-centre' framework may offer a powerful tool for future research and debates about the possibilities and challenges associated with multiple and flexible positionalities of men and masculinities in teaching. According to Elliott (2020, 4), marginalised masculinities may shift and change moving either towards fixed and normative identities 'at the centre' or towards 'the margin', which may 'lock them out of the power and privilege at the centre' but offer 'greater space and radical possibility for more fluid and open expression'. Existing qualitative research exploring male teachers' experiences in Ireland (O'Keeffe 2016;Neary 2013) indicates that complex cultural and institutional factors, including gender stereotypes, attitudes towards caring and the Catholic ethos of primary schools and ITE programmes, significantly constrain expressions of diverse and 'authentic' masculinities and, similarly, flexible 'movement between centre and margins'.
Intersectional analyses of masculinities in teaching may, ultimately, help to shape a new discourse of men in teaching that recognises multiple positionalities and identities of male teachers, both in Ireland and internationally. The argument that different men are well positioned, but currently constrained, to provide different perspectives which can effect change and contribute to gender equality provides a dialectic-based rationale for increasing male representation in teaching.

Reflections for the future
Diversifying the teaching profession is a complex matter, not least in relation to gender. Our findings draw attention to the particular ways in which gender matters in teaching.
While some argue uncritically for a greater gender balance in the teaching profession, our research points to the existing diversity among male applicants and entrants to initial teacher education and to the importance of an intersectional analysis in shaping theory about the need for male teachers. The significance of a dialectic-based rationale for more men in teaching, and the related importance of diverse, including caring, masculinities in schooling, requires a much more dynamic and flexible conceptualisation of gender, men, male teachers, and masculinities in education to be recognised by all educators, researchers, administrators, and policy-makers in Ireland and internationally. As part of this, essentialist gendered assumptions about male and female teachers, and related discourses about the 'feminisation' of teaching, need to be challenged and cultural and institutional constraints recognised in order to appreciate and mobilise the potential significant benefits of diverse men and masculinities (as well as diverse women and femininities) in teaching.
While having diverse men in teaching is not a panacea for all ills, and will not by itself guarantee positive change in schools, recognising its potential benefits offers useful ways of negotiating the complexities inherent in the gender in teaching debates, and of affirming the value of men in teaching, without denigrating the role of women therein.

Notes
1. While the term 'feminisation' is used to refer to the teaching profession that is numerically dominated by women, the term suggests that the feminine presence is inherently problematic and implies a deficit. We believe this term needs to be examined critically. 2. See Heinz (2015) for an overview of the use of the FIT-Choice Scales internationally. 3. Question and categories were adopted from the European Values survey. Collapsed categories were used for Chi-square tests: A religious person, Not a religious person/a convinced atheist, Don't know. 4. See Heinz, Davison, and Keane (2018) for more details regarding the analysis of this qualitative item. 5. For chi-square analysis only categories 1-5 were used. Chi-square analysis could not be performed for entrant cohort as minimum expected cell count assumptions were not met due to small numbers of male entrants.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by the Irish Research Council [RPG2013-1].

Notes on contributors
Manuela Heinz is a lecturer (Teaching and Learning and Research Methods), Director of Teaching and Learning and joint Programme Director of the Professional Master of Education at the School of Education, National University of Ireland Galway. Her research and publications focus on initial teacher education and teacher professional learning.
Elaine Keane is a lecturer (Sociology of Education and Research Methods) and Director of Doctoral Studies in the School of Education, National University of Ireland Galway. Her research and publications centre on widening participation in higher education and diversity in initial teacher education.
Kevin Davison is a lecturer and joint Programme Director of the Professional Master of Education at the School of Education, National University of Ireland Galway. His research and publications focus on gender in education and science outreach.