LGBTQ+ non-discrimination and religious freedom in the context of government-funded faith-based education, social welfare, health care, and aged care

Anti-discrimination laws around the world have explicitly protected LGBTQ+ people from discrimination with various levels of exceptions for religion. Some conservative religious organisations in Australia are advocating to be allowed to discriminate against LGBTQ+ people in certain organisations they manage. The political debate in Australia has focused on religiously affiliated organisations that provide services in education, social welfare, health care, and aged care. We argue that religious exceptions allowing discrimination should be narrow because they cause considerable harm, reinforce, disadvantage and because LGBTQ+ people are deserving of respect and rights. We draw on a national representative survey to demonstrate that the views of some conservative religious lobby groups do not represent the views of the majority of religious people in Australia or the views of the majority of Christian people.

Anti-discrimination laws around the world have explicitly protected LGBTQ+ people from discrimination with various levels of exceptions for religion (Tebbe, 2017).These exceptions are contentious, with some religious actors seeking to extend their ability to discriminate, and others seeking to minimise limitations on the right of LGBTQ+ people to non-discrimination.To the former, the Australian federal government has tabled (in 2021) a Religious Discrimination Bill and associated legislation (Attorney-General, 2021) aimed at protecting 'religious freedom'.Some conservative religious organisations in Australia are advocating to be allowed to discriminate against LGBTQ+ people in the context of certain organisations they manage, those which provide government-funded education, social welfare, health care, and aged care to the general Australian population. 1  One of the justifications for the Religious Discrimination Bill and associated legislation is that exceptions that allow faith-based organisations to discriminate in the provision of public services are broadly supported by people of faith, and by Australians more generally.For example, reporting on a poll commissioned by the Australian Christian Lobby claimed that 60% of respondents 'supported [religious] schools' rights to not employ staff "in conflict" with their religious beliefs' (Ferguson, 2019, p. 1).The research question that this article addresses is whether Australians, and in particular religious Australians, support discriminatory practices in publicly funded faith-based organisations.
There are strong arguments against exceptions that allow faith-based organisations to discriminate in the provision of public services.We argue later that Australia's international human rights commitments require that the Australian government prohibit discrimination in the delivery of government-funded services by faith-based organisations.Doing so would be consistent with common international practice, as Gogarty et al. (2018) show very few of the international jurisdictions that permit same-sex marriage individuals, religious bodies, and religiously affiliated organisations.The legislation was given impetus by the Commonwealth government's Religious Freedom Review of 2018 which was commissioned to assuage concerns that some religious Australians had in response to the legalisation of same-sex marriage in Australia after the 2017 voluntary Marriage Law Postal Survey.
Though 'religious freedom' was not the most prominent issue of public debate in the push for marriage equality, there was some exploration by legislators of the religious 'right to discriminate' in the context of LGBTQ+ rights.The Marriage Amendment (Definition and Religious Freedoms) Act 2017 (Cth) established the rights of two individuals to marry irrespective of gender.It also established the rights of 'ministers of religion, religious marriage celebrants, chaplains and bodies established for religious purposes to refuse to solemnise or provide facilities, goods and services for marriages on religious grounds' (McKeown, 2018, n.p., emphasis added).Some wanted this freedom-to-refuse to extend to the provision of services in non-religious contexts.For example, Senator James Paterson proposed an alternative Marriage Equality Bill which would have provided extended protections to 'protect the freedoms of other individuals and businessesfor instance, florists, cake makers and photographers' (Collett, 2017).However, Senator Paterson did not introduce this bill.
Religious lobby groups have been major contributors to the public debate about the freedom of some religious people to refuse to employ and provide services to LGBTQ+ people in faith-based organisations that provide government-funded services to the general population.This debate is about the extent of the right of religious bodies to engage in discriminatory practices that have a broad impact on other Australians.

Conservative Christian responses
Even though 52% of the Australian population identified as Christian in the 2016 Census (Bouma et al., 2022), Christian lobby groups depict themselves as 'a minority who are under threat from mainstream norms which give recognition to, and acceptance of, LGBT+ people' (Freedom for Faith, 2018, p. 24).Following the 'Yes' vote in the Marriage Postal Survey, then Federal Treasurer Scott Morrison echoed this language.Morrison described the nearly 5 million 'No' voters as a new 'minority' whose 'broader views and beliefs' were now 'under threat' (Commonwealth of Australia, 2017, p. 12347).This discourse of marginality is a common tactic utilised by conservative Christian lobby groups (McIvor, 2020, p. 85).
Beyond this, in the debates about marriage equality and the religious freedom bills, influential and vocal religious groups claim to speak on behalf of mostor all -Christians.These Christian groups often describe discrimination exceptions as central to their religious freedom and advocate their retention and expansion. 2Existing exceptions give broad discretion to religious bodies to discriminate, such as to dismiss a school teacher who comes out as gay, or to refuse welfare services based on a person's gender identity.One of the most vocal advocates for exceptions is the Anglican diocese of Sydney.The Sydney diocese describes discrimination exceptions as necessary for Christian institutions to 'participate in national life as Christians', suggesting that without them Christian bodies could not 'be' Christian or be involved in public life (Anglican Church Diocese of Sydney, 2018, p. 15).Similarly, the high-profile Australian Christian Lobby and its sister organisation, the Human Rights Law Alliance (HRLA), also purport to represent the 'Christian constituency', claiming to be the 'voice of Christians … in the public square' and to 'bring a Christian perspective to policy makers' in government (Australian Christian Lobby & Human Rights Law Alliance, 2018, p. 1).
The Anglican diocese of Sydney has raised fears that reducing the scope of discrimination exceptions could interfere with faith-based conditions attached to its provision of welfare services, such as foster care.The diocese submitted to the Religious Freedom Review that all its community services must be offered conditionally, consistently with its teaching that 'heterosexual … marriage is both the norm and ideal' (Anglican Church Diocese of Sydney, 2018, p. 14).The HRLA made similar submissions with respect to commercial businesses operated by Christians, arguing that they should have a right to discriminate against LGBTQ+ clients and staff (Australian Christian Lobby & Human Rights Law Alliance, 2018).In order to show what religious freedom in commercial settings would entail, HRLA gave examples of disputes it was involved with.These included supporting the refusal by an employer to use appropriate pronouns for transgender staff, refusals by medical practitioners to refer to or provide certain health services, and a booking refusal by a commercial campsite due to the inquiring group's support of same-sex attraction (Australian Christian Lobby & Human Rights Law Alliance, 2018).
However, it is not clear to what extent the views of Christian lobby groups mirror those of the religious people they claim to represent.Our research is designed to address this question.

Service provision by faith-based organisations
Upwards of one third of the Australian workforce in sectors such as education, social welfare, health care, and aged care, are employed by organisations with religious affiliations (Oslington, 2012;Rowe, 2017).These organisations hire employees who may be LGBTQ+ or have LGBTQ+-affirming attitudes that may conflict with the employer's exclusionary attitudes and practices toward LGBTQ+ people.Such exclusionary attitudes and practices are inconsistent with civic and governmental service goals, such as the aim to provide universal access to high-quality education, social welfare support, health care and aged care.
The history of Catholic education in Australia provides a demonstrative example.Up until the 1970s, Catholic schools were attended almost entirely by Catholic students and received very little government funding.Catholic students experienced considerable socio-economic inequality (Gaffney, 2014).The Menzies government gradually increased funding support to Catholic schools from 1965, and the Whitlam government introduced needs-based funding to Catholic schools in 1972 (Warhurst, 2012).These developments came after strong government lobbying by Catholics and a strike by Catholic school students.More recent growth of Australia's non-state education sector was a product of the 'Howard Government's removal of the cap on new schools in 1997' and the government provision of 'establishment grants' to new non-state schools (Cranston, 2010, p. 187).At the same time as the number of Catholic schools expanded, there was also a declining number of Catholics in the Australian population (Bouma et al., 2022).As a consequence, Catholic schools no longer provide education solely to Catholic students.
Religiously affiliated schools are becoming less a service provided by a religious organisation to its adherents, and increasingly more like a state-funded service provided to the general population but managed by a religious organisation.Within Catholic schools nationally, the proportion of students who are Catholic has declined from 74.4% in 2008to 66% in 2018(National Catholic Education Commission, 2019, p. 14).Some states have even lower percentages, with Tasmanian Catholic schools having only 42.3% of students identifying as Catholic in 2018.Drawing on 2016 Census data, the Independent School Council of Australia (2018, p. 19) reports that non-Catholic independent schools have 46% of their students who are non-Catholic Christians, 28% with no religion, 15% Catholic students, and 11% from other religions.A small number of schools restrict their students to those who identify with the religion of the school.For example, Moriah College (2018, p. 3) 'only accepts enrolment of a child … who is Jewish in accordance with halacha'.However, enrolment restrictions such as this are not typical of most non-government schools.
A similar pattern can be observed in the provision of social welfare.Maddox (2005) observes that a transfer of substantial portions of welfare services to the churches occurred during early 2000s.For example, the federal government outsourced a support program to assist unemployed people to find work, and outsourced programs for the provision of early intervention for children and families, so that religious institutions took over services that had previously been provided by government-run agencies (Maddox, 2005, p. 235).As a result of these government contracts, many faith-based agencies have greatly increased in scale, and they have increasingly employed people from outside their related faith communities.

The harm caused by LGBTQ+ discrimination
In Australia, there is substantial evidence for the harm that discrimination causes LGBTQ+ people, as employees, as students, and as health care and social welfare recipients.The Private Lives 3 (PL3) study shows that the overall mental and physical health of LGBTQ+ people is significantly lower than that of the general Australian population (Hill et al., 2021).According to Rosenstreich (2013), these lower health outcomes are directly correlated to the experience of discrimination and exclusion.Meyer and Frost (2013, p. 252) suggest that social stressors such as 'stigma, prejudice and discrimination' can cause a lower sense of general wellbeing in sexual and gender minorities.Horner (2013), in a survey of 415 LGBTQ+ respondents across Australia, reported that 18% experienced some form of discrimination at work.More recently, a survey of approximately 2500 found that 36% of respondents had witnessed homophobia at work in the last 12 months (GLEE@PwC, 2018).The Diversity Council of Australia's survey of LGBTQ+ people in the workplace found that 40% of respondents felt 'invisible' in their workplace, experiencing the workplace culture as 'don't ask, don't tell' (Brown et al., 2018).Barrett, Lewis and Dwyer (2011) found that in Queensland in 2010, despite inclusive policies, 54% of survey participants experienced discrimination in the workplace.Ferfolja (2013) identifies that lesbian and gay teachers in New South Wales (NSW) state schools experience marginalisation and silencing, despite positive policy and anti-discrimination legislation protecting them.Conversely, inclusive and welcoming practices in the workplace enhance the mental health of LGBTQ+ workers.For example, Brown et al. (2018) found that being out at work improves health and performance outcomes of LGBTQ+ workers.Of the 1600 people surveyed, those who were out at work were twice as likely to feel happy than those who were not, and were 45% less likely to be dissatisfied in the workplace.Being out in the workplace is also beneficial to worker productivity, as less energy is expended in hiding their LGBTQ+ status and more is invested in contributing to the workplace (Conway, 2018).
The harm caused to LGBTQ+ people is particularly clear, for example, in the experience of LGBTQ+ students in Catholic schools in Australia.Norden (2006, p. 25), from Jesuit Social Services, heard stories of depression, isolation and discrimination in Catholic schools: 'A consequence of this discrimination for same sex attracted young people is that they have increased rates of homelessness, risk-taking behaviour, depression, suicide and episodes of self-harm compared to young heterosexuals.' At the heart of these debates is whether it is acceptable to provide publicly funded services to the general population conditionally on faith-based terms in order to ensure the integrity of the religious character of the organisation when such terms undermine the human rights to equality and dignity, and inflict considerable harm on LGBTQ+ people?

Human rights and religious morality
The International Covenant on Economic, Social and Cultural Rights (ICESCR) (1966) sets out that states 'have the obligation to guarantee the enjoyment of [economic, social and cultural] rights to all without discrimination' (CESCR, 2017 GC24; Jameson and Aubry, 2020).Those economic, social and cultural rights encompass the right to work, and the right to services such as health, housing, food, water, and social security (CESCR, 2017 GC24).States are under strong moralif ultimately unenforceablepressure to comply with and report on their voluntarily assumed international human rights obligations.A state does not avoid its non-discrimination obligations by contracting out its functions to private bodies (CESCR, 2017 GC24).
Contracting out public services to faith-based organisations raises the risk that the fundamental right to non-discrimination could, in some circumstances, be at odds with the right of people in faith-based organisations to 'manifest one's religion or beliefs' in Article 18 (1) of the International Covenant on Civil and Political Rights (ICCPR).The conflict is resolved in international human rights law in the following way.Under Article 18(1) of the ICCPR, the human right to manifest religious belief is recognised in relation to 'worship, observance, practice and teaching'; it is not recognised in relation to, say, the conduct of a business or the provision of a service.On that basis, the right to manifest religious belief does not extend to allowing faith-based organisations to give effect to discriminatory beliefs when providing contracted-out public services.But even if the right to manifest religious belief extended to the conduct of a business, a state is permitted to limit the right to the extent that is 'necessary to protect public safety, order, health, or morals or the fundamental rights and freedoms of others' (ICCPR Art. 18(3)).The right to equality and non-discrimination is one such fundamental right (UN HCR 1989 CCPR GC18).This means that the state can permissibly limit the right to manifest a religious belief to the extent that is necessary to protect the right to equality and non-discrimination.Thus, if a religious practice would mean less favourable treatment of people because of their LGBTQ+ status, a state can limit that religious practice to the extent necessary to ensure non-discrimination.
For consistency with its international human rights obligations, therefore, Australian governments could, when contracting out public services to faith-based organisations, require non-discrimination, despite any inconsistent religious belief.This is not, however, the practice in Australia, or what Australian law requires.While all workers and service recipients in Australia have the protection of anti-discrimination laws, there are exceptions to those laws that allow religious bodies to discriminate on the basis of a person's sexual orientation.The Sex Discrimination Act (SDA), for example, allows a 'body established for religious purposes' to engage in sexual orientation discrimination (except in the context of aged care) when doing so 'conforms to the doctrines, tenets or beliefs of that religion or is necessary to avoid injury to the religious susceptibilities of adherents of that religion' (SDA ss 37, 38). 3  There are complexities in the detail of these exceptions across federal, state, and territory jurisdictions (Rees et al. 2018), but the central point is that faith-based organisations are currently allowed to discriminate on the basis of a person's sexual orientation.From a human rights perspective, such exceptions to anti-discrimination laws allowing public service providers to discriminate in their operations puts the government in breach of its international human rights obligations.A government does not, by contracting out its public services, avoid its non-discrimination obligations under the international human rights system.

Method
The Australian Survey of Social Attitudes (AuSSA) (McNeil et al., 2021) collected responses from February to June 2021 in a mail-out questionnaire to individuals randomly selected from the electoral roll, with 1162 completed surveys and a 0.25 response rate (the cooperation rate was 0.7, refusal rate was 0.1).The data were weighted using the weight variable provided with the data that weight by age, sex, and highest education level, using the 2016 Census as the comparison (McNeil et al., 2021).Table 1 provides a comparison of the unweighted and weighted demographic characteristics.Younger respondents and those with only year 12 education are more heavily weighted.The regression analysis (Table 6) was also run with unweighted data and where this produced a difference in the significance of results, this is reported in the notes on the table.
We included questions in the AuSSA that measured support for the right to discriminate, using the following statements with Likert scale response possibilities: 'A conservative religious person working in a cake shop should be allowed to refuse to sell a wedding cake to a same-sex couple' (abbreviated in the tables below as 'LGBTQ+ wedding'); 'Conservative Catholic, Anglican, Jewish, and Muslim schools should be allowed to refuse to employ a teacher because they are LGBT+' (abbreviated as 'LGBTQ+ teachers'); 4 'A conservative religiously affiliated social welfare organisation providing accommodation for homeless people should be allowed to direct LGBT+ homeless people to seek welfare assistance from another organisation' (abbreviated as 'LGBTQ+ homeless').These were all 5-point scales: strongly agree, agree, neither agree nor disagree, disagree, strongly disagree.In some of the analyses below the strongly agree and agree categories, and the strongly disagree and disagree categories have been combined.
Attitudes toward the separation of church and state were obtained with two questions.Other questions asked about the environment: 'Generally speaking, how concerned are you about environmental issues?' (1 Not at all concerned to 5 Very concerned), years of education, personal income, and respondent age (measured in years); about immigration: 'Australia should limit immigration in order to protect our national way
Just over one third agree, just under one third disagree, and one third are unsure.Christians are more likely to agree with advocating Christian values (among Catholics and Anglicans 57% agree, other Christians 73% agree).Unsurprisingly, those with no religion (20%) or who follow a non-Christian religion (17%) are less likely to agree, while those who attend religious services regularly are the most likely to agree (79%).The relatively low rates of attendance at religious services (12.8% attend at least monthly), may partially reflect the influence of Covid restrictions.In the 2018 AuSSA survey, 15.4% of the population attended at least monthly (Evans et al., 2018), a 17% drop in attendance that may be due to Covid.Those who identify with the Coalition (62%) are more likely to agree that the federal government should advocate Christian values, than are those who identify with Labor (29%); of those with no party affiliation, 31% agree, and those who identify with the Greens are least likely to agree (10%).Australians' attitudes toward whether the Christian Lord's prayer should open parliament (Table 2), as is the current practice, mirrors the pattern of attitudes toward whether the Australian federal government should advocate Christian values.Approximately one third (32%) say there should be no prayers in parliament, 20% can't choose and 16% say that parliament should open with a minute of silence.One third of people want religious prayers in parliament, either the Christian Lord's prayer (25.7%) or prayers from a variety of religions (6.3%).Unsurprisingly, both nominal (36%) and attending (52%) religious people are more likely to think the Lord's prayer should be used to open parliament.A small minority of those with no religion want prayers in parliament (12%), and almost half think there should be no prayers (45%) (Table 3).

Attitudes toward LGBTQ+ people among the Australian population
Less than one fifth of the Australian population support religious people being allowed to discriminate against LGBTQ+ people (Table 4).This is a considerably lower level of support than the one third of Australians who support the federal government advocating LGBTQ+ wedding (per cent) LGBTQ+ teacher (per cent) LGBTQ+ homeless (per cent)   A: Agree or strongly agree.Ne: Neither agree nor disagree.D: Disagree or strongly disagree.
Each subscript letter denotes a category whose proportions within the categories of 'Agree', 'Neither agree nor disagree' or 'Disagree' do not differ significantly from each other at the 0.05 level for the Z test of proportions.Cells with low percentages or low numbers are not reliable for comparison purposes.
Christian values or opening parliament with the Lord's prayer.Three questions were asked about discrimination by a religious person and/or organisation: when selling a wedding cake, when employing a teacher in a religiously affiliated school, and when providing services to the homeless.Among the general Australian population, 77% disagree and 16% agree that a conservative religious person should be able to discriminate against same-sex couples while selling a wedding cake, 72% disagree and 19% agree that a conservative religious school should be allowed to refuse to employ an LGBTQ+ teacher, and 74% disagree and 16% agree that a conservative religious social welfare organisation should be allowed to redirect LGBTQ+ homeless people to other service providers.
Women are less likely than are men to support this discrimination, particularly in relation to selling a wedding cake, with 80% disagreeing and only 11% agreeing.
While support for discrimination against LGBTQ+ people is higher among people who are nominally religious, the substantial majority of those people do not think that discrimination against LGBTQ+ people should be allowed in any of selling a wedding cake (78%), employing a teacher (68%) or providing homeless services (67%) (Table 4).Among Catholics and Anglicans, between two thirds and three quarters disagree that discrimination against LGBTQ+ people should be allowed.Among other Christians and those who follow religions such as Islam, Buddhism and Hinduism, between 45% and 61% disagree with such discrimination, although the percentages for non-Christians should be treated with caution because of the small numbers of respondents.
A minority of those who attend religious services at least monthly agree that discriminatory practices should be allowed in any of selling a wedding cake (35%), employing a teacher (41%) or providing homeless services (25%) (Table 4).Further, only one quarter to one third of those who agree that the federal government should advocate Christian values agree with discrimination against LGBT+ people in employment or service provision.Among those who agree with the federal government advocating Christian values, 22% agree that discriminatory practices should be allowed in selling a wedding cake, 32% agree for employing a teacher, and 23% agree for when providing homeless services (Table 4).

Political party identification
Among people who identify with the Coalition, between 22% and 26% agree that conservative religious people should be allowed to engage in discriminatory behaviour toward LGBTQ+ people in any of selling a wedding cake, employing a teacher or providing homeless services, and between 64% and 71% disagree (Table 5).Between 72% and 76% of those with no party affiliation disagree with such discrimination.

A regression model
Table 6 provides a regression model of the predictors of agreement with religious people's right to discriminate against LGBTQ+ people.The dependent variable is a scale measuring agreement with religious people's right to discriminate against LGBTQ+ people, with higher values indicating agreement.Those who agree with discrimination in religious contexts are more likely to be regular religious attendees, believe the federal government should advocate Christian values, be indifferent to LGBTQ+ wedding (per cent) LGBTQ+ teacher (per cent) LGBTQ+ homeless (per cent) A LGBTQ+ wedding: X2(6, N = 1049) = 38.990,p = 0.000.
LGBTQ+ homeless: X2(6, N = 1048) = 36.489,p = 0.000.3.89 * * 0.17 * * p < 0.01.* p < 0.05.R Square: 0.260, F = 8.403, p <0.01, N = 900.Dependent variable: Scale measuring agreement with religious people's rights to discriminate against LGBTQ+ people (Cronbach's Alpha: 0.803).The scale was calculated by summing and dividing by 3, responses to the variables: LGBTQ+ wedding, LGBTQ+ teacher, and LGBTQ+ homeless.The scale variable has a range from 1 (strong agreement with the right of religious people to discriminate against LGBTQ+ people) to 5 (strong disagreement).The regression reported is with weighted data.Using unweighted data results in minor changes to beta values.All the significance levels remain the same, with the exception of 'Green identifying', which has a beta of 0.15 and is not significant with unweighted data.environmental issues, be men, and think that Australia should prioritise freedom of speech.These factors are consistent with those observed as characteristic of conservative Christian lobby groups.Conservative Christian lobby groups emphasise freedom of speech and their engagement 'on environmental matters appears to have been very limited' (Pepper & Leonard, 2016, n.p.).It is also interesting that women are more likely than men to disagree with discrimination against LGBTQ+ people by religious people.This may reflect a broader gendered difference in responses to conservative Christian lobby groups who have strongly patriarchal attitudes toward women (Maddox, 2014).Those who disagree with discrimination in religious contexts are more likely to be non-religious, concerned about the environment, see the government as non-religious, be women, and not list freedom of speech as an Australian priority.This does not mean that people who attend religious services are more likely to agree with LGBTQ+ discrimination.Only just over one third of people who regularly attend religious services agree with discrimination against LGBTQ+ people (Table 4).Rather, it means that those who agree that discrimination against LGBTQ+ people is acceptable are more likely to be people who attend religious services regularly.
It is noteworthy that both antipathy toward minorities and voting Green, while significant, are weaker predictors and only significant at the 0.05 level.Generalised antipathy toward minorities or 'others', as indicated by strongly agreeing that Australia should limit immigration is a weaker predictor than the other variables, Similarly, those identifying as Green voters are more likely to disagree, when compared to Labor voters, that religious people should be able to discriminate against LGBTQ+ people, however the effect is weaker than that of the other variables.Age and years of education are not statistically significant predictors.Income is also not significantly correlated with the discrimination scale, and was not a significant predictor when included in the regression model, but was excluded from the analysis due to the high number of missing cases.
In the regression model, there is no significant influence from identifying as a Coalition voter or having no party affiliation when compared to Labor voting.In the crosstabulation of Table 5, there are often significant differences in attitudes between Labor-and Coaltion-identifying voters.The regression analysis suggests that this difference in attitudes is largely a product of the influence of conservative Christian lobby groups.
Overall, this regression suggests that support for discrimination against LGBTQ+ people in religious contexts is primarily driven by factors common to politically activist conservative Christian lobby groups.It is not related to education, income, age, or affiliation to the two major political parties.Green party identification and antipathy toward immigrants are significant, but weaker influences.
The views of the conservative Christian lobby groups do not reflect the views of the majority of Christian Australians.Among those who agree that the Australian federal government should advocate Christian values, only one third (32%) agree that faith-based schools should be able to discriminate against LGBTQ+ teachers.Only 20% of Catholics, 25% of Anglicans, and 35% of other Christians agree, and 41% of those who regularly attend religious services agree (Table 4).

Conclusion
The AuSSA survey data demonstrates that the majority of Australians do not think that discriminating against LGBTQ+ people should be permitted in the provision of public services by faith-based organisations.Three quarters of Australians do not support religiously affiliated service providers discriminating against LGBTQ+ people, either as employees or as people in receipt of these services.This opposition is much higher than the one third of Australians who are opposed to the federal government advocating Christian values, or commencing parliament with the Lord's prayer.A similar majority of religiously identifying Australians also do not support such discriminatory practices.Among Australians who attend religious services regularly, support slightly increases, but is still in a minority, with the majority indicating they do not support religious service providers discriminating against LGBTQ+ people.Further, among Australians who think the federal government should advocate Christian values, only between 22% and 32% agree with discriminating against LGBTQ+ people in employment or service provision.
Conservative Christian lobby groups often claim their views are representative of all Christians (see for example Australian Christian Lobby and Human Rights Law Alliance, 2018; Freedom for Faith, 2018).Our research demonstrates that this is clearly not the case.The majority of Christians oppose discrimination against LGBTQ+ people in employment and the provision of services.It may be that many Christians do not support discriminatory practices against LGBTQ+ people because they feel that LGBTQ+ people have the right to be treated equally and with respect.Others may simply be resigned, as suggested by McIvor (2020, p. 113) who studied both a legal activist Christian organisation in London and a conservative Christian church.She argues that the public indignation of rights-based Christian activist organisations is often not shared by conservative Christians in the church pews, who have accepted the different expectations of the non-religious mainstream culture 'as the inevitable result of secularisation … [with a] resignation [that] suggests an acceptance of the inevitability of the status quo'.
There is further research to be done into religious practices in the context of government-funded religious schools in Australia that are associated with strict, orthodox or strongly conservative religious traditions.For example, some schools restrict their students to those who identify with the religion of the school.This includes schools run by some smaller Christian traditions such as the Brethren, and some strict Muslim schools (Coleman & White, 2011) and schools run by ultra-orthodox Jews (Goss & Rutland, 2014).Such schools raise a different set of potential issues and harms that may require a different set of considerations (Evans & Gaze, 2010).Rizvi (2014, p. 588), for example, notes that for some strict Muslims in the UK, 'in the absence of [single-sex Islamic] girls' schools, some parents preferred to send their daughters to their home countries or keep them un-schooled'.Vaggione (2018, p. 27) argues that in Argentina, 'the politicization of religious pluralism opens up the possibility that citizens, state officials, and legislators may support legal reforms because of (and not in spite of) their religious beliefs'.That is to say, national culture, religious 'freedom' and being 'Christian' are often conflated with resistance to progressive legislative responses to sexuality by conservative religious actors, such as the Catholic Church in Argentina and Christian lobby groups in Australia.Our data suggests that religious support for progressive sexual politics may be much more widespread than the political and media debate suggests.
A substantial proportion of education, social welfare, health care, and aged care services in Australia is provided to the general population by faith-based organisations.The people who work in these organisations and the people who receive their services are often LGBTQ+ people.In these contexts, Australia has international human rights obligations to limit religious rights to the extent necessary to accommodate LGBTQ+ people's right to non-discrimination.The government's obligation to provide services without discrimination under international human rights law is not changed by the contracting out of service provision to faith-based providers.LGBTQ+ people are deserving of equality and respect, and this should not be eroded by discrimination in publicly funded contexts.Our research suggests that the majority of religious people in Australia do not support discriminatory practices in these contexts.The views of some religious lobby groups and some religious leaders do not represent the views of the majority of religious people in Australia.Similarly, the vast majority of Australians do not support discrimination against LGBTQ+ people in the context of faith-based service provision.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
One asked: 'The Australian Federal Government should advocate Christian values' with a similar 5-point scale.Respondents were also asked: 'The opening of Australian Federal Parliament includes the Christian Lord's Prayer.What do you think of this?' and were offered possible responses: 'The Lord's Prayer should be used to open Parliament'; 'Prayers from a variety of religions should open Parliament'; 'There should be no religious prayers to open Parliament'; 'Parliament should open with a minute of silence';'Can't choose'.The AuSSA survey included a question about religious identity: 'Do you belong to a religion and, if yes, which religion do you belong to?', and attendance at religious services was indicated by the question: 'Apart from such special occasions as weddings, funerals, etc., how often do you attend religious services?' ,b 17 b 63 b,c 30 b,c 17 b,c 53

Table 1 .
Weighted and unweighted data demographics.oflife' (5-point Likert scale from strongly agree to strongly disagree); about political affiliation: 'Do you usually think of yourself as close to any particular political party and, if yes, which party is that?'; and about freedom of speech, respondents were provided with a list that included 'Protect freedom of speech' and asked to indicate which they thought should be 'Australia's highest priority, the most important thing it should do?'

Table 2 .
The Australian federal government should advocate Christian values.
Question: The Australian Federal Government should advocate Christian values.Each subscript letter denotes a category whose proportions within the categories of 'Agree', 'Neither agree nor disagree' or 'Disagree' do not differ significantly from each other at the 0.05 level for the Z test of proportions.Cells with low percentages or low numbers are not reliable for comparison purposes.

Table 3 .
Prayers in parliament.
2(8, N = 1102) = 249.593,p = 0.000.Each subscript letter denotes a category whose proportions among the categories of attitudes toward prayers in Parliament do not differ significantly from each other at the 0.05 level for the Z test of proportions.
: Agree or strongly agree.Ne: Neither agree nor disagree.D: Disagree or strongly disagree.Question: Do you usually think of yourself as close to any particular political party and, if yes, which party is that?.Each subscript letter denotes a category whose proportions within the categories of 'Agree', 'Neither agree nor disagree' or 'Disagree' do not differ significantly from each other at the 0.05 level for the Z test of proportions.Cells with low percentages or low numbers are not reliable for comparison purposes. A

Table 6 .
Regression model of agreement with religious people's right to discriminate.