Policies to change attitudes to people with disabilities

Negative attitudes are a major barrier to the equality of people with disabilities. Governments and other organizations have implemented numerous programmes to change attitudes towards people with disabilities. We analyse published evidence about the effectiveness of such programmes using a framework of the interrelationship among three levels of policy intervention to change attitudes: personal level – directed at changing the attitudes of individuals; organizational level – concerning attitudinal barriers in domains such as employment, education and health; and government level – legally mandating behaviour change. The analysis finds that the following policy types can be effective if used together: policies that involve direct contact with people with disability; information and awareness campaigns; education and training about disability; and antidiscrimination enforcement. Policy characteristics that contribute to effectiveness include a positive programme experience for participants; multifaceted and prolonged interventions; and adequate programme resources. Policy effectiveness to change attitudes relies on corresponding reinforcement at all three policy levels. ARTICLE HISTORY Received 25 November 2014 Accepted 15 June 2016


Introduction
Negative attitudes towards people with disabilities 1 are a major barrier to the equality of people with disabilities. In response, governments and organizations have implemented programmes to improve attitudes. Yet, most people with disabilities continue to experience attitudinal barriers to attaining their human rights. Analysing why policies struggle to address pervasive negative attitudes, and which aspects of the initiatives might be effective, is important to resolve ongoing discrimination.
The article analyses published evidence about the effectiveness of policies, programmes and initiatives for changing attitudes towards people with disabilities. It structures the multitude of interventions using a classification framework that was developed with reference to the literature and findings. The framework refers to the interrelationship among three levels of policy intervention to change attitudes: personal leveldirected at changing the attitudes of individuals; organizational levelconcerning attitudinal barriers in domains such as employment, education and health; and government levellegally mandating behaviour change. The three levels of intervention are defined and illustrated with policy examples. The article draws out implications about how the effectiveness of interventions may be improved.

Background to policies to change attitudes
In this background section, we introduce the approach taken in the article to understand attitudes and the mechanisms for change; the classification framework for analysing policies; and the method applied to identify and analyse policies. We use a social perspective of disability, which understands that negative attitudes are a barrier to social equality of people with disabilities in their communities (Shakespeare 2006). This is also recognized in the Preamble (e) of the United Nations Convention on the Rights of Persons with Disabilities (CRPD 2008): ' … disability results from the interaction between persons with impairments and attitudinal and environmental barriers that hinders their full and effective participation in society on an equal basis with others'.
According to Allport (1935) and Petty and Cacioppo (1981), attitudes comprise two components: thoughts and feelings. A person's attitudes and their behaviour are correlated, but not always identical: a person can think and feel in one way, but act in another, even opposite way. Attitudes towards people with disability can be positive or negative, but even when they are positive, there can be a disjunction between perceptions. For example, a person without disability may conceptualize a positive attitude as being 'nice' or 'helpful', whereas a person with disability might find it patronizing (Yazbeck, McVilly, and Parmenter 2004, 97). The strength with which attitudes are held can be affected by factors such as direct experience (Fishbein and Ajzen 1975), modelling (Bandura 1977), values (Rokeach 1973), situational context (Calder and Ross 1973) and exposure to other people (Zajonc 1968). Attitudes are held by individuals but are also formed, reinforced and experienced at a community level.
Research about attitudes to people with disabilities demonstrates that many people hold negative attitudes and that these attitudes are barriers to the equality of people with disabilities (Deal 2007). Younger people and people with more education tend to have more positive attitudes towards people with disabilities (Yazbeck, McVilly, and Parmenter 2004). Negative attitudes may be due to misconceptions, lack of awareness or knowledge, and can be manifested in outright vilification or direct or indirect discrimination (Deal 2007). Research also shows that attitudes towards specific groups of people with disabilities differ, resulting in more disadvantage for women than men, particularly in the workforce; and showing that people without disabilities are less comfortable with people with psychiatric disabilities than with people with physical disabilities (Meekosha 2004;Simkhada et al. 2013;Wallace 2004). Familiarity with people with disabilitiesthat is, knowing them personally as acquaintances, friends and colleaguesseems the most promising way to increase respect, especially if exposure is consistent and recent (Burge, Ouellette-Kuntz, and Hutchinson 2008).
Policy to change attitudes to people with disabilities recognizes that they have the same rights to achieve their goals and aspirations as other members of society, and in ways commensurate with achieving the fullest possible quality of life (Clement and Bigby 2008). The CRPD sets out these rights and obliges governments to implement policies to realize and protect them, including attitude change.
An aim of policies about disability attitudes ultimately is to improve the realization of the rights, inclusion and participation of people with disabilities. Social inclusion can be seen to have two components: relational and distributional inclusion (Fraser 1999). Relational inclusion involves people's sense that they are valued as much as other members of society. This requires respect and non-discrimination on a personal level. Distributional inclusion involves equality of access to opportunities at an organizational level: in social and economic life domains such as education, employment, housing and leisure activities.
Targeting policies only at the individual and organizational levels has the disadvantage that only some people with disabilities experience the benefits. Systemic and large-scale change requires overarching policies at the government level, for example, inclusive education and disability rights legislation. The three levels of policies to change attitudespersonal, organizational and governmentalare equally important, and they interact and reinforce each other (Sawrikar and Katz 2008). Together, they provide a framework for categorizing the variety of policies aimed at changing attitudes. Cook et al. (2014) apply a similar framework to understanding the effectiveness of interventions to address stigma and to improve public health ( Figure 1).
Personal-level policies are directed at changing the attitudes of individuals towards people with disabilities. They involve the following elements, often in combination: information, education, training, positive portrayal of people with disabilities, and supported opportunities for contact between people with and without disabilities. Personal-level policies are based on theories of persuasion from the cognitive and social psychology literature (Brostrand 2006;Hunt and Hunt 2004;Kleeman and Wilson 2007, 15;Petty and Cacioppo 1981). According to these theories, persuasion could come from intensive information campaigns which challenge negative thinking and beliefs, ignorance, misunderstanding, myths, misperceptions, stereotyping and fear. Clarity of the message is important: information needs to be easily understood and specific about what to do.
Organizational-level policies attempt to improve attitudes towards people with disabilities in various life domains, such as education, employment and health. Policy types include training, complaints mechanisms and targeted information programmes, which seek to mitigate the power disadvantages experienced by people with disabilities, by changing the behaviour and attitudes of people without disabilities. At the same time, these policies try to empower people with disabilities to claim their rights to equal access and participation. This is consistent with the model proposed by Scotch and Schriner (1997), and with a social investment approach that intends to realize the potential of people with disabilities (Deane 2009). Effective, long-term improvement of attitudes requires significant levels of combined educational and experiential interventions (Campbell, Gilmore, and Cuskelly 2003;Harvey 1985;Tait and Purdie 2000). While organizational-level initiatives are similar to policies at the personal level in being directed at individuals, they are distinguished through their focus on life domains and types of organizations, such as schools and workplaces.
Government-level policies are initiated by governments and attempt to influence attitudes by mandating behaviour change. This level consists of the policy statements and laws that define actions reflective of positive attitudes, together with the means for implementation and monitoring. Examples are anti-discrimination legislation and inclusive education. This approach is consistent with the theory of cognitive dissonance (Festinger 1957), which assumes that government-level policies, by forcing people to change their behaviour, can eventually change most people's underlying attitudes too, when they reconcile the dissonance between their attitudes and their behaviour. Not all people make that reconciliation, but at the least, they may be forced to change their behaviour in the meantime. In addition, laws that mandate the social and economic integration of people with disabilities provide public confirmation that discrimination 'is no longer acceptable' (Barnes and Oliver 1995, 114). Personal-and organizational-level policies can be established by governments, non-government agencies or advocacy groups. Government-level policies originate from governments, although they may need other agencies and community groups to implement the policies and tailor them to local requirements. Each policy intervention level reinforces the others, and policies may be applied at different levels at the same time.
The distinction that is made here among three levels serves to classify the range of possible policy interventions and to suggest that disability attitudes are, as Cook et al. (2014, 101) put it with relation to health part of a reciprocal web of relationships among individuals, their social networks, and larger social structures. This perspective encourages researchers to consider how effective interventions that target stigma at any level, when well-timed and congruent with conditions at other levels, might have long-term, cascading effects across a system.

Review method
The research was a review of published literature and policy documents about initiatives to change attitudes towards people with disability. It was part of a larger research project commissioned by the Australian Government and received ethics approval from UNSW Australia (Thompson et al. 2011). The review included searches in Google Scholar and specialized databases such as IngentaConnect, Project Muse and Cambridge Journals Online. The search involved snowballing whereby citations within documents were followed to ensure the inclusion of the most relevant academic commentary on attitudes. Sources were included in the review if they offered a critical analysis of attitudes towards disability and of programmes designed to change these attitudes. This criterion eliminated many sources in the review. Other sources were excluded if they described a policy intended to be implemented in the future.
Key words for the search were disability, attitudes, perceptions, awareness, childhood, employment, education, leisure, sport, student, employer, employee, health, transport, criminal justice, criminal, recidivism and incarceration. 'Disability awareness' was a term commonly used in the literature to identify programmes directed towards attitude change.
The review was exploratory rather than systematic and confined to English language materials. A systematic review was not adopted because, as is common in social policy analysis, many of the policy documents are only available in grey literature, which requires an investigative approach. While terminology around different types of literature reviews is varied and inconsistent (Grant and Booth 2009), this project can be characterized as a scoping review (Arksey and O'Malley 2005) due to the breadth of its topic, because it was exploratory rather than systematic and because it did not include quality assessment of the identified literature. The limitation of the research is that it encompasses only a policy analysis of published material. The research could be extended by including follow-up primary data collection with people involved in the policies including people with disabilities, community members and policy-makers; and undertaking a systematic review of evaluated policy interventions for wider attitudinal change, such as antidiscrimination.
In the findings sections below, examples of policies at all three levels of the framework are described and linked to the relevant CRPD Articles. All available evidence about policy effectiveness in changing attitudes towards people with disabilities is included. As was common, where no evidence of effectiveness is mentioned, none was available.

Personal-level policies
Personal-level policies attempt to change the attitudes of individuals, by increasing knowledge and understanding about the experience of disability. Such policies can be widespread or tailored to particular locations or groups of people. This section describes examples of personal-level approaches to changing attitudes towards people with disabilities, grouped into three policy types: awareness raising (through information and training), social contact programmes, and positive portrayal of people with disabilities in the media and the arts.

Awareness raising
Raising awareness about disability is attempted through information campaigns and awareness training (Art. 8 CRPD). Information campaigns provide information about disability to the general public or particular groups. They are implemented by government or non-government organizations, sometimes as part of a larger strategic programme of advocacy and services. Disability awareness training is conducted mainly through courses run by tertiary education institutions or non-government organizations.
An example of an awareness campaign at the international level is the International Day of People with disabilities on 3 December each year. It was established by the United Nations in 1992 (as the International Day of Persons with Disabilities) to raise awareness of disability issues and promote the integration of people with disabilities. In Australia, for example, governments support activities to mark this day (www.idpwd.com.au). The federal government gives awards to people with disabilities and organizations who have improved the lives of people with disabilities; and across Australian states and territories, dozens of awareness-raising public events are organized, including dance parties, street stalls and art competitions. In contrast to earlier years, people with disabilities are involved in decision-making about the priorities of the events. Some of them hold positions as government officials and others are advisers from Disabled Person's Organisations.
In New Zealand, the Ministry of Health has funded a public education programme since 2000 aimed at reducing the stigma and discrimination associated with mental illness. Like Minds, Like Mine includes a wide range of activities, such as TV campaigns, community workshops, internet resources and disability provider training (www.likeminds.org.nz). Research in 2005 and 2007 showed that the campaign was successful in improving attitudes (Litmus Ltd 2008). People were more likely than they had been in 1997 to accept someone with a mental illness as an employee, a workmate, a babysitter or a neighbour.
ONCE (Organización Nacional de Ciegos Españoles), the Spanish National Organisation of the Blind and a non-profit service provider, operates a lottery scheme that provides income for its activities (www.once.es/new/otras-webs/english). As well as raising money, the lotteries keep the public aware of the organization and the rights of people with vision impairment.
Disability advocacy organizations in many countries use public information and education to improve social inclusion. Campaigns may operate at national or local levels. One example from the USA is Disability Awareness Starts Here! (DASH), which campaigns for accessibility of public places in Jefferson County (www.dashproject.org). DASH aims 'to educate the community about why and how good access makes sense for everyone'. It organizes community forums, produces local accessibility maps and organizes public disability awareness activities ('Assume-a-disability' events).
Disability awareness training is the subject of courses provided in many countries by tertiary education institutions and non-government or advocacy organizations. For example, the Canadian Centre on Disability Studies in Winnipeg, Manitoba, offers education about disability to academics and the public, along with conducting research in disability studies (http://disabilitystudies.ca). In Chicago, USA, the Western DuPage Special Recreation Association has conducted disability awareness programmes for the general population, teaching how to interact with someone who has disabilities, reduce the fear of the unknown and recognize a person's abilities (WDSRA 2010).

Social contact
The second type of personal-level policy is to arrange social contact between people with and without disabilities. Research evidence suggests that direct contact with people with disabilities improves people's attitudes (e.g. Burge, Ouellette-Kuntz, and Hutchinson 2008;Kleeman and Wilson 2007;Krahé and Altwasser 2006;Murfitt 2006;Scior 2011;Yazbeck, McVilly, and Parmenter 2004). For example, Jones et al. (2008) found that people who had relatives or friends with intellectual disabilities were more supportive than people without this experience, and Yazbeck, McVilly, and Parmenter (2004) observed that people were less discriminatory towards people with intellectual disabilities if they had interacted with them in the last six months. One study in the UK (Cameron and Rutland 2006) found that reading stories to children aged 5-10 years of friendships between children with and without disabilities significantly improved the children's attitudes by the end of a six-week period. This indicates that interventions to reduce prejudice could be effective with primary school children in contexts where there is no opportunity for direct contact.
However, some studies indicate that social contact by itself does not guarantee success. To improve attitudes, the quality of the interaction between people with and without disabilities needs to be interactive, respectful and pleasant (Hutzler and Levi 2008;McManus, Feyes, and Saucier 2011), it needs to occur in a supportive environment (Murfitt 2006) and it needs to be accompanied by information, because otherwise unfamiliarity may increase anxiety and thus reinforce negative beliefs (Lee and Rodda 1994).

Positive portrayal
The third personal-level policy approach to changing attitudes is positive portrayal of people with disabilities in the media and the arts (Art. 8.2.c, CRPD). Governments and advocacy groups agree that mass media have a powerful role in shaping people's attitudes (DAA 1993; UK Prime Minister's Strategy Unit 2005), as does portrayal and participation of people with disabilities in the arts (Victorian OFD 2010). The approach is based on the social cognitive theory of mass communication (Bandura 1986(Bandura , 2002, which provides a conceptual framework for understanding how media influences people's thoughts and behaviours. Mass communication affects individuals both indirectly, for example, via attitudes and emotions, and directly (Valkenburg, Peter, and Walther 2016).
In Australia, the national Australian Broadcasting Corporation (ABC) stated in its Editorial Policies document (ABC 2009) that the content of any broadcast 'should not use language or images which disparage or discriminate against any person or group' on any of a number of grounds including disability, and instructs its employees to comply with anti-discrimination legislation. The British Broadcasting Corporation (BBC) has a website, Ouch!, that intends to reflect the lives and experiences of people with disabilities. The website (www.bbc.co.uk/blogs/ouch) contains links, news articles, blogs, a message board and a downloadable radio show.
The Canadian Government's booklet A Way with Words (Canadian Government 2006) is directed at the media and responds to requests by people with disabilities that what is said about them be respectful, and that images portraying them do not reinforce outdated stereotypes. The booklet recommends appropriate terminology and images.
In Germany in 1997, the disability rights movement ran a massive media campaign to educate the public about disability rights. The campaign consisted of advertisements on every twelfth billboard across the country. For example, a billboard at a busy intersection said, 'How is a blind person supposed to cross this street?' These were supplemented with TV ads, and slogans on posters, buttons, beer coasters, stickers and postcards disseminated in every neighbourhood. Street theatre, conferences, community meetings, talk shows and public demonstrations were held throughout the country. Opinion polls indicated that the public response was overwhelmingly positive, with 87% of respondents finding the campaign necessary. However, no data exist on whether this approach changed attitudes and behaviours.
The international non-profit organization VSA (Very Special Arts) was founded in the USA in 1974. VSA has prepared an information guide for people involved in the arts who want knowledge about disability and advice on social etiquette and positive interactions (VSA arts 2006). VSA also develops resources for arts programming in schools and communities; supports artists with disabilities; and promotes increased access to the arts for people with disabilities. There are VSA programmes in more than 60 countries around the world.
In summary, a large number and variety of personal-level policies exist around the world, but evidence about their effectiveness in changing attitudes is scarce. Available research shows that a multipronged, long-term public education campaigns that directly involved people with disabilities, contributed to change. There is also strong evidence that prolonged direct contact with people with disabilities improved people's attitudes. To be most effective, direct contact needed to occur in a supportive environment, and involve positive interactions and additional information.

Organizational-level policies
Organizational-level policies attempt to change the attitudes of people in various social and economic life domains, such as work, education, health, leisure or transport. The policy goal is to increase the participation of people with disabilities and to improve their experiences when they do participate. This section describes policy examples in the life domains of education, employment and health. The review found insufficient literature to discuss other life domains such as transport, housing, leisure and corrective services.

Education
Research shows that inclusive school education can improve the attitudes of children without disabilities. For example, studies in Canada (Bunch and Valeo 2004) and Chile (Sirlopú et al. 2008) found that children in schools with inclusion programmes had more positive attitudes towards students with disabilities than children in schools with segregated education. However, inclusion programmes appear to be most effective if they are resourced adequatelywith physical access, assistive technologies, small classes and non-teaching staff to support children with disabilitiesand if mainstream children and teachers are well prepared (Buell et al. 1999;Maras and Brown 1996). In fact, studies from the UK, the USA and Australia (Crabtree and Rutland 2001;Santich and Kavanagh 1997;Stinson, Whitmore, and Kluwin 1996) found that self-esteem and emotional security of children with disabilities are worse if they participate in inclusion programmes that lack adequate support.
To prepare mainstream teachers for inclusion programmes, special training and resources are advocated (Art. 24.4 CRPD). Research has linked insufficient teacher training for inclusion to a lack of confidence, which led to reluctance to have children with disabilities in their classrooms (Deane 2009;Hsien, Brown, and Bortoli 2009). Teacher training could incorporate information about the benefits of inclusion (D'Alonzo, Giordano, and Vanleeuwen 1997) and instruction on topics such as assessment of academic progress, adapting the curriculum and using assistive technologies (Buell et al. 1999). Research strongly supports inclusion training for all undergraduate teaching students and ongoing professional development for existing teachers (e.g. Deane 2009;Hsien, Brown, and Bortoli 2009). Evidence suggests that teacher training can indeed be effective in improving their attitudes (Campbell, Gilmore, and Cuskelly 2003).
Likewise, educating students about disability can help to improve their attitudes towards peers with disabilities (Art. 8.2.d CRPD). According to studies from the UK, Canada and Nicaragua (Burge, Ouellette-Kuntz, and Hutchinson 2008;Maras and Brown 1996;Scior et al. 2015), disability awareness and information programmes in schools made students significantly more inclined to play with their classmates with disabilities, helped them to 'personalize' children with disabilities and see them as individuals who they could interact with, and created lasting positive attitudes towards the social inclusion of people with disabilities.
Awareness programmes for school students exist in many countries, may be implemented by governments or community organizations and take different forms. For example, a local government in south-eastern Melbourne, Australia, ran a Disability Awareness in Schools project that involved local people with disabilities talking to grade 5 and 6 students about their lives (Glen Eira City Council 2007). In the USA, the website of the National Dissemination Center for Children with Disabilities (www.nichcy.org) provides teaching resources via links to other networks and providers, such as a link to the Disability Awareness Activity Packet (Adcock and Remus 2006) that intends to give students some experience of what it is like to have different kinds of disabilities. For example: 'This activity is designed to show how people with autism are bothered by things most people don't notice'. Student attitudes may be changed through prolonged contact, information and the introduction of disability ambassadors (ACT DAC 2007).

Employment
People with disabilities have much lower employment rates than people without disabilities, and negative attitudes of employers and workmates have been identified as an important reason for this (e.g. Snyder et al. 2010). Studies from Australia, the UK, Canada and other countries show that many employers are reluctant to employ people with disabilities (e.g. CCDS 2001; DEEWR 2008; UK Prime Minister's Strategy Unit 2005), especially those with mental health issues, learning disabilities or blindness (Zissi et al. 2007). Reasons for their reluctance include negative employer attitudes, and these are often due to misconceptions about the capabilities of people with disabilities, the cost of workplace modifications, stereotyping and fear of workplace disruption.
To address attitudinal barriers in the workplace (Art. 27 CRPD), numerous training and support programmes for employers have been created. In the USA, for example, the Tilting at Windmills programme was first developed in 1980 and has been widely used in most states (Brostrand 2006). It is designed to increase awareness of attitudes through activities and discussions. The Australian Government provides information, training and links to government financial support via its JobAccess website (http://jobaccess.gov.au), not dissimilar to the Australian Network on Disability www.and. org.au, which includes business, government and not-for-profit organizations.
Approaches to changing employers' attitudes include: leadership from the top; government support to employers in the form of information, resources and recognition; credible and reliable sources of information and awareness training to share best practice; and networks for recruitment and support. Workplaces where managers have personal experience of disabilities are the most accommodating towards recruiting people with disabilities (Erebus International 2006;Gustafsson, Peralta, and Danermark 2014). Initiatives to change co-workers' attitudes include information, contact and training (Haney and Rabin 1984;Krahé and Altwasser 2006;Wallace 2004;Waterhouse et al. 2010).

Health
Access to health and community services is vital to fulfil people's right to health and well-being (Art 25. CRPD). Organizational-level policies often attempt to change the attitudes of health and community services staff by providing training and awareness raising, but evaluations of these programmes show inconsistent results. Some research has found that attitudes do not change or that they worsen after coursework and direct contact with patients with disabilities or mental illness (Arkar and Eker 1997;Chadd and Pangilinan 2011). Other research has found that health professionals' attitudes towards people with psychiatric disability, particularly the attitudes of male professionals, can be worse than the attitudes of the general public (Jorm et al. 1999;Tervo et al. 2002).
It seems that in this sector, training and contact are insufficient to change attitudes. More successful approaches might be to also change the profile of the health workforce by employing people with disabilities, and including consumer representatives in the health sector (Conchar and Repper 2014).
In summary, there is solid evidence that organizational-level policies in the education sector can be effective in changing attitudes. In combination, inclusive school education, inclusion training for teachers and student awareness programmes have been shown to improve attitudes, especially when they are well resourced and operate over prolonged time periods. More generally, in other workplaces, employer attitudes were found to be most positive where employers had prior personal experience with people with disabilities either in the workplace or elsewhere. Available evidence from the health sector indicates that training and contact are insufficient to change attitudes.

Government-level policies
The final level of policy to change attitudes is at the government level. These are government interventions that mandate behaviour change and thus indirectly influence attitudes. The level consists of the policy statements and laws that define actions reflecting positive attitudes, and the means to implement and monitor the policies. Government-level policies include disability rights legislation and mechanisms to support the legislation, such as standards, strategies and human rights agencies.
Their effectiveness relies on interrelationships with the personal-and organizational-level policy interventions, such as related awareness and training programmes and cooperation across government and community levels. Experience from campaigns to change attitudes in various policy areas including health and road safety indicates that a multifaceted and collaborative approach contributes to policy effectiveness (Productivity Commission 2004). Similarly, anti-discrimination legislation appears to be most effective if accompanied by awareness-raising campaigns and education programmes at various levels (Masselot et al. 2006).
For example, Australia ratified the CRPD, obliging federal and state governments to amend legislation contrary to the Convention. This includes policies that affect attitudes, cited throughout this article. Australia also legislated the National Disability Strategy to overcome barriers in life domains such as education, employment, health, housing and income support. The strategy includes improving attitudes towards disability as an aspect of people's inclusion, participation and equality. It also intends to provide leadership for a community-wide shift in attitudes, and it articulates longterm goals across key policy areas which impact on people with disabilities (COAG 2010). In addition, Australian federal and state governments have human rights agencies, Disability Services Acts, Discrimination Acts, Disability Standards and Disability Action Plans to both drive and enforce changed attitudes. These measures articulate the expectation that communities and individuals will use the laws and policies to reinforce personal and local action for attitude change, consistent with theories of attitude change summarized in the background to this article. A review of the Australian Disability Discrimination Act (Productivity Commission 2004) concluded that disability recognition and awareness had been promoted through the combined effect of the various parts of the Act, including information provision, public inquiries, complaint mechanisms, disability standards and action plans, and policy guidelines. An important reason for the effectiveness was the public consultation process.
Inclusive education is a specific example of government-level policy that is intended to change attitudes. Its aim is to ensure the right of children with disabilities to attend education with other children (Article 24 ;Foreman 2005;Hsien, Brown, and Bortoli 2009;Wedell 2005). Inclusive education is also articulated in the UN Convention on the Rights of the Child and the Salamanca Statement and Framework for Action on Special Needs Education (UNESCO 1994).
Mechanisms supporting inclusive education policies include implementation and monitoring guidelines and allocating resources. Implementation initiatives include school-based training; support and resources for peers and teachers to improve attitudes; and individualization of support so that it is specific to children's needs and moves with them through the education system (Deane 2009). Countries with the policies and resources to support inclusive education have the highest participation of children with disabilities in education (UNESCO 2014).
In summary, the example of inclusive education demonstrates how government-level policies can contribute to changing attitudes when they are supported by personal-and organizational-level policies and resources.

Discussion and conclusion
The link between negative attitudes to people with disabilities and barriers to their rights is well established. Some people with disabilities are more likely to experience negative attitudes, such as women and people with intellectual or psychiatric disabilities. Governments have publicly recognized this relationship between attitudes and rights and have committed to changing attitudes with their ratification of the UN CRPD.
While psychology and sociology have determined that the strength of attitudes can be influenced by external factors, as outlined in the background section above, it is not clear what policy interventions are effective to change attitudes. In this article, we applied a three-level policy intervention framework to categorize literature about the effectiveness of relevant policies. The review found many examples of personal-, organizational-and government-level policies across countries. The policies vary in the way they are operationalized (national or local); their time frame (one-off campaigns, annual, over several years or ongoing); their target (whole of population, people in positions of power, peers or strangers); and who initiates them (government, non-government or citizens).
The review found that few policies have been rigorously evaluated. Paucity of evaluation evidence in this field has been confirmed by a worldwide review of initiatives aimed at improving attitudes towards people with intellectual disability (Scior et al. 2015). The limited available evidence points to some policy types that are associated with effectiveness: . Policies that include direct contact with people with disability ('social contact' section). . Information and awareness campaigns (the Like Minds, Like Mine campaign and programmes for school children). . Education and training about disability (teacher and employer training). . Legislation to enforce anti-discrimination measures (Australian Disability Discrimination Act, inclusive education).
The research evidence also indicates that certain policy characteristics contribute to effectiveness: . Direct contact needs to be a positive experience for participants, and be combined with information. . Information, awareness, education and training programmes need to be multifaceted and prolonged. . Any policy needs to be resourced adequately.
These findings are based on scant evidence and need to be verified and expanded through further research. This would include formal, independent evaluations of policies from the outset, to measure change over time and gather robust evidence about the link between policies and attitude change.
The classification framework used in this articlewhich was developed from the literaturesuggests that approaches to change attitudes are more likely to be effective if all three levels are acted upon simultaneously. This is demonstrated in the evidence available for this article regarding children's rights to education. Countries where attitudes to children with disabilities participating in inclusive education have changed the most have introduced multiple interventions at all three levels, including: mandated inclusive education policies (government level); teacher and student training, support and contact (organizational level); and community media (personal level), which expose children, families and teachers to positive portrayals of people with disabilities (UNESCO 2014). The theoretical literature and the framework in this article indicate that multi-level interventions are more likely to be effective because they can address the diversity of disability experience, reinforce positive attitudes and replace negative attitudes through repeated information, emotional engagement and mandated change.
Further research could include evaluations of multi-level policy interventions. These evaluations could also measure change in attitudes and the impact of policies and attitude change on rights and outcomes. A disability rights approach (CRPD 2008) suggests that evaluations include the perspectives of people with and without disabilities, people holding organizational power in the relevant domains and policy officials deciding about which policies and interventions to prioritize. In addition, more sustained attitude measurement would contribute to understanding strategies to change attitudes and the effect on the rights of people with disabilities. Note 1. 'People with disabilities' is the term adopted in this article to be consistent with people first language and the UN CRPD.