The understanding of what constitutes inclusion in education has been a global development over time beginning in 1948 with the United Nations (UN) adoption of the Universal Declaration of Human Rights (UDHR). This declaration highlighted social justice, equality and the right to education for all (EFA). The UN Convention on the Rights of the Child in 1989 affirmed the right to EFA children and emphasised that this right should be non-discriminatory. At the Jomtien conference in 1990, EFA was proposed, and equal rights to education for the disabled were emphasised. In 1994 in Salamanca at the World Conference on Special Needs Education, policy revisions for the development of inclusive education were considered. The Salamanca Statement and Framework for Action on Special Needs Education (UNESCO 1994) shifted EFA’s emphasis, and whilst acknowledging equal rights for the disabled, extended focus to acknowledging each child is different and has unique learning needs, and that educational systems and programmes should accommodate for diversity. It was suggested that learners with barriers should be accommodated in mainstream school settings, and EFA achieved through schools adopting an inclusive orientation (UNESCO 1994). Moreover, the Salamanca Statement affirmed the need to combat exclusionary and discriminatory practices and change societal attitudes towards difference. The EFA goals were reiterated at the World Education Forum in Dakar in 2000, where they were endorsed and adopted and where the needs of the poor and disadvantaged were included as important for consideration in acknowledging diversity. These international developments demonstrate a shift from an initial focus on special educational needs and disability primarily to a broader consideration of recognising difference and diversity and combatting exclusionary pressures and practices as a key priority for inclusion in education.

Alongside international developments clarifying and elaborating on what inclusion in education means, individual countries were simultaneously developing their own educational, legislative and policy responses to developing inclusive education systems within their unique contexts.

Miles and Singal (2010) suggest that inclusive education needs to be considered in relation to specific cultures and contexts to attend to educational inequalities specific to those contexts. This suggests that whilst inclusive education is a global common principle, it cannot be practised in uniformity. Although there is a consensus on the importance of inclusive education, its interpretation and the programme of action will always differ (European Agency for Special Needs and Inclusive Education 2017). We argue that the understanding of inclusion cannot be comprehended by disconnecting from other situations around the world as challenges to the actualisation of inclusivity are universal (Armstrong, Armstrong & Spandagou 2010; Walton 2015) whilst acknowledging different contextual realities. Therefore, schools and governments should look at incidents in communities that perpetuate inequalities in order to understand features that promote exclusion (Slee 2001). With this understanding, it is notable to explore the specific context of Botswana.

Despite Inclusive Education shifts from a specific focus on special education to a broader focus that recognises disability as one aspect of diversity internationally, Botswana still tends to operate from a standpoint of special needs. Hence, Botswana continues to establish special education units for students with disabilities in both rural and urban areas of the country (Dart 2007; Molosiwa & Mpofu 2017). The most recent and first of its kind being the establishment of a Special Needs department at one of the TVET institutions to cater for students with mild intellectual disabilities. With this practice, Botswana has neglected to recognise the broader conceptualisation of inclusive education and its subsequent reconsiderations as a unified education system. This failure also contradicts the conventional Botswana Inclusive Education Policy, which states that:

The existence of the Inclusive Education policy proves that there is a common understanding from policymakers in Botswana, but that this is not necessarily being implemented in practice. The existing implementation demonstrates how Botswana, despite the introduction of an Inclusive Education policy in 2011, has not shifted from the traditional view that the education of children with disabilities is only possible in segregation. We are of the opinion that given where Botswana is operating from, inclusive education in Botswana vis-a-vis the goal of international policy on inclusive education has remained a policy statement and has failed the expectations of the principles of inclusive practices (Muzata et al. 2019). This challenge is not only to Botswana but also a challenge faced in most of the Southern African countries (Pather & Nxumalo 2013) that are failing to address barriers in the system that restrict the education of marginalised individuals including children with disabilities. The key challenge is the lack of understanding that inclusion is a flexible way to allow improved support in the education system, and it is not about merely identifying learners with special educational needs who are deemed to be educated separately.

In terms of developments in the TVET sector, Botswana like other African countries was compelled to respond to the global policies such as Salamanca Statement and Framework for Action on Special Needs Education of 1994, even though the progress has been challenging (UNESCO 2015). The inclusion of marginalised communities and people in Botswana follows the global pattern of action in response to Development Goal No. 4 and the EFA (UN 2015). Sustainable Development Goal 4 speaks of ensuring inclusive education and equitable quality education as well as opportunities for all (UN 2015). This goal aims at reducing barriers to skills development in TVET education (UN 2015). In line with this goal, Botswana is reported to have taken a stride to ensure the right to education of people with disabilities through its longstanding National Policy on TVET of 1997 (Mmolai 2019; Ndzinge-Makhamisa 2019), which originated before the SDGs. The vision of the policy is to provide access to vocational training with disadvantaged groups as a priority, and this stance tallies with Botswana’s action measures on the implementation of inclusive education.

Botswana’s articulation of inclusive education refers to ‘an education system that includes and meets the needs of all, including those with special educational needs, those with life circumstances, health, stages of development or any other circumstances’ (GOB 2011:4). One could understand Botswana’s articulation of inclusive education and inclusion to imply that the society has obligations to take care of each of its members regardless of mental, physical, behavioural and emotional status. If Botswana understands inclusion from a societal obligation standpoint, it is necessary to consider in more detail the policies that support this stand.

Botswana’s first policy on education entitled Education for Kagisano of 1977, the subsequent Government Paper No 2 of 1994 and the National Revised Policy on Education (RNPE) have been instrumental in leading Botswana’s education system in an inclusive direction. The policy recognised that Botswana’s breakthrough to economic diversification is through VET. The RNPE advises that whilst the industry commits to specialised vocational education, the government should be in charge of primary vocational education. This initiative was viewed as the means to achieve economic diversification and become instrumental in addressing issues of theory and practice mismatch (Pheko & Molefhe 2017). The need for vocational education and training (VET) in Botswana was impelled by the country’s economic history (Mupimpila & Narayana 2009; Siphambe 2009), where vocational education was not viewed as a priority and educating for white-collar jobs (workforce in offices). Educating for white-collar jobs meant a focus on theoretical rather than practical courses in many cases. In terms of inclusion, three policies promote inclusion and inclusive education in TVET. Firstly, the VET Policy of 1997, which calls for ‘increase in access to VET by making education inclusive and equitable whilst addressing issues of quality and cost-efficiency’. The second policy is the RNPE, which has a special provision for an education, which caters for children and young people including those with disabilities (Republic of Botswana 1994). The third is the Inclusive Education Policy of 2011, which has 10 goals of which goal no 4 is significant to this article. It states that ‘action will be taken to improve vocational training for young people for whom the current system of vocational training is unsuitable’ (GOB 2011:1). The adoption of these three policies did not, however, mean that these vulnerable groups would access education in practice. The enrolment of individuals with disabilities in TVET is in fact still minimal (EFA Country Profile 2015).

Botswana’s educational development is interpreted with the National Development Plan (NDP), which usually run for 6 years. National development plans in Botswana are reflective frameworks in which government checks signs of progress in the different sectors of the economy. Therefore, it is not surprising that the objectives of the initiative of vocational education continue to change. The study informing this article took place within the NDP 11 plan. The NDP 11 plan commits to the following:

Whilst the initial vision of TVET in Botswana was for productivism, the reflection from the current National Development Plan (NDP) indicates that the development of human capital is significant to the economy. Hence, the current status of vocational education is not solely government domain, as the growth of private vocational institutions is on the increase (Richardson 2013; Samboma 2017). Not only is the current NDP objective to promote human development but embraces inclusion through inclusive education directives with the emphasis that everybody who has the right to education should be given the opportunity. The NDP 11 provides for the expansion of brigades and technical colleges (Richardson 2013). Brigades are community-led vocational schools, whereas technical colleges are government-led institutions of technical and vocational training. Another government initiative that promoted an increase in formal access for people with disabilities was to earmark certain brigades and technical colleges to offer different programmes for different disabilities. This initiative enabled the admission of the first cohort of students with mild intellectual disabilities in 2012 to the technical college selected for this study. Although this can be viewed as a positive step, the reality of the implementation of inclusive education on the ground seems to be rhetoric. There is still inequality and exclusion of marginalised students in Botswana’s higher education (Makwinja 2020). This supports the consideration of how TVET institutions practise inclusion.

Admission of students with disabilities in the technical college selected for this study follows a process of interviews with candidates and their parents to discuss the educational psychologists’ report, which is a prerequisite for admission. Suitable prospective students are then selected based on current ability levels and the perceived fit with the programme. There is, therefore, accommodation and modification in terms of the admission processes. The college initiative follows the country’s Affirmative Action Framework (Ministry of Education and Skills Development [MoEDS] 2013), which has the goal to ensure that no one is left behind by admitting students regardless of test scores. Current practice in the TVET college is to separate students with disabilities from the so-called ‘normal’ students so that classes for students with disabilities are offered separately. The separate provision is justified as allowing students with disabilities to learn at their own pace in order to have epistemological access. A pragmatic justification for separate classes is that students with disabilities doing hospitality operations have a separate kitchen where equipment has been modified. For example, stove handles have been lowered for any students in wheelchairs. Some classes, however, are still on the second floor of a double storey building making them inaccessible to some, especially when the lifts are not working. Whilst the college attempts to offer students with disabilities the opportunity of learning on their own, it simultaneously portrays a negative view of the students’ disabilities. On a different note, the college’s justification of this separation is that programmes for students with disabilities are below the level of what is offered in the same hospitality operations at diploma level for students without disabilities. This segregation has resulted in labels and othering of students in the department. At the same time, the label is seen by the management and policymakers in the ministry as a means to identify the educational needs of students with disabilities. What the college is experiencing resonates with the ‘dilemma of difference’. The dilemma of difference as coined by Terzi (2005) and Norwich (2010) is a choice between treating all students as the same so that no one feels different and provides similar support and resources or whether to label and treat students differently so that they can benefit from individualised support. Besides the concern for students learning, the college is also faced with the operational mandate. That is, whether the department should be a service department in which it serves the educational needs of all students in the college or as a special unit focusing on the students in their section only.

When students have completed a unit, the student’s disabilities are catered for through modification and enrichment of assessment. Assessment is modified for those students experiencing difficulty and enriched assessment is provided to those for whom the materials are found to be too easy. This speaks of an inclusive assessment practice in which there is a ‘need to review and modify test items to ensure that they have maximum accessibility, without changing the properties of the test items’ (Elliott, Frey & Davies 2015:2). More importantly, the assessment criteria determine whether students with disabilities have achieved the unit independently, with assistance or not achieved at all. This allows these students to be re-assessed until they can achieve. This is also part of an inclusive practice in which students are assessed for checking skills acquisition rather than grading so that students are not left behind.

Initially, students with disabilities at the TVET College were also socially separated from others as their hostel accommodation was separate from the rest of the student community. However, the current situation is that students with disabilities are paired with students without disabilities to promote social integration. Positive results of the initiative include integrated sports where teams were formed and ultimately improved their sense of belonging. Pairing students with disabilities with students without disabilities may, however, be reinforcing difference, labelling and othering. According to Norwich (2010), there is a dilemma of identification. This dilemma means that the students may not be able to enhance their social skills because they are treated by the so-called normal differently but if they are identified according to the disability, it may help the college to secure what will benefit them. For example, the hostel doors were modified to allow wheelchair users. In practice therefore, it becomes clear that at the selected TVET College, formal access to the institution has been facilitated for students with disabilities, but the question of whether full inclusion beyond formal access is being achieved remains.

One TVET college in Botswana was selected as the site for this study using convenience sampling as it is one of the two institutions that offer programmes for students with disabilities. The participants were purposefully selected for the study to get relevant information for the study. Purposive sampling is a subjective non-probability sampling in which participants are chosen through the judgement of the researcher (Patton 2014). The experiences of students could not be attained from people other than the students themselves; hence, purposive sampling was relevant. Participants were 14 students with mild intellectual disabilities who were enrolled in the Hospitality program. These students were chosen on the criterion that they are students with mild intellectual disabilities and as the technical college admits this specific disability. A characteristic list was made to further sample participants based on the ability to understand both Setswana and English, ability to communicate well, the chronological age of 18–23 years and programmes of study (in this case, Hospitality Operations). Ethical clearance was obtained for the study from the affiliated higher education institution, and all procedural ethical processes were followed, including obtaining informed consent from all participants, ensuring anonymity and maintaining confidentiality. The following data collection tool was used to construct participants’ meaning of their experiences of inclusion at the selected site.

In-depth individual interviews were conducted with both the currently enrolled and graduated students’ participants. The interview was guided by a semi-structured interview schedule where questions focused on the participants’ experiences of inclusion and exclusion in the TVET college. The individual interviews were critical in having a ‘conversation with a purpose’ with participants. A semi-structured approach was adopted to ensure attention to critical key questions related to students’ experience of inclusion but also to allow for probing and elaboration of these responses. Considering the realities that student meanings of experiences are different and unique to individuals, we found it essential to use individual interviews so that we could probe further and clarify participants’ meanings so as not to misinterpret their real experiences (Rubin & Rubin 2012). Fourteen individual interviews with currently enrolled students were conducted. The interviews were employed to contextualise their meaning through their voices, and therefore they served as a space for understanding those voices (Creswell 2014). These voices were audio-recorded verbatim and later transcribed.

The study’s data analysis was informed by its epistemological position, which is to interpret meanings. We employed an iterative process of data analysis. An iterative process means that we had to look at the data as a whole through its parts and vice versa to ultimately come to a context in which the whole and the parts were embedded (Smith, Flowers & Larkin 2013). With this process, we had to keep an open view of what the implication of the data might mean. For instance, data items were analysed individually looking at what students said, what it meant for each data set and ultimately what the interpretations were. This progression aided us to arrive at what Holloway and Todres cited in Braun and Clarke (2006) call ‘thematising meanings’. In the words of Braun and Clarke (2006:6), ‘thematic analysis is a method of identifying, analysing and reporting patterns (themes) within data’. Prior to thematising, the following process was ensured.

Recordings from the individual interviews were translated verbatim from the vernacular to English, and the English versions were then transcribed. Particular attention was paid to the individuals’ contextual experiences. Engaging with data assisted us in extracting and identifying significant meanings as told by the participants. Transcripts were manually coded following three steps. Step one was to open the code drawing from the literature review to identify essential elements that could possibly make a pattern by comparing and categorising (Saldaña 2013). The second step was the axial coding, looking at the causal relationships (Saldaña 2013). In this step, a sequence of thematic codes was generated by employing the Capabilities Approach (affiliation, agency, freedom, capabilities and functioning). This follows Neuman’s (2006) idea that at this stage a clear link between the concrete findings and the theoretical themes should be noticeable. Thematic analysis was not limited to original themes but was flexible to patterns that came out and were helpful to answer the research question beyond the initial themes. The third step was selective coding, which helped to develop explanations and move beyond the descriptive analysis of the data. We looked for themes through the thematic latent analysis level. The latent analysis level demands that a researcher goes beyond the surface meaning but reaches themes through interpretations of the data by identifying features and ideologies that go beyond the meaning (Braun & Clarke 2006). For instance, in this study, we analysed interview extracts by not only looking at what the literature refers to as elements of inclusion but we also used information related to the research question and its relation to the Capability Approach.

Ethics clearance was obtained for the study from the University of the Witwatersrand, reference number: 2017ECE017D.

The experiences of students in relation to inclusion are summarised in Figure 1, which shows indicators of how inclusionary practices were identified from the data. Four key areas were identified, namely (1) social inclusion; (2) epistemological access; (3) formal access and (4) respect, recognition and dignity.

The analysis showed that students experience some level of good inclusionary practices by being socially included. Students reported having been able to interact with others socially and form friendships. The interpersonal skill modules helped them to live well with others and have reciprocal relations with some lecturers and the college non-academic staff.

Being enrolled in TVET gave the students epistemological access. Here, students were given the opportunity of gaining new knowledge through accessing learning resources and lecturers knowing their profiles that helped with being scaffolded and remediated where necessary. The students also benefited from the integrated teaching that lecturers employed. Integrated teaching between the vocational lecturers and the special educators enabled the students to be given attention and units objectives tailored to their profiles. In this way, students were able to access content that eventually gave them some capabilities achievement. New knowledge acquired shaped students’ aspirations for their future. Box 1 demonstrates an example of these aspirations as expressed by student participants.

Some students were also grateful for the TVET experience because they felt a sense of belonging. They were able to live well and interact with others. This speaks to the capability of affiliation (Nussbaum 2000). The achievement of this capability meant that these students were able to form social networks, which were instrumental in their forming learning groups, which helped with other capability achievements such as numeracy, literacy, ICT and problem-solving skills to name a few. Also being able to relate well resulted in students being able to accept their disabilities, eventually gaining confidence and self-esteem as well as being able to take risks without fear.

Social inclusion not only benefited students in forming networks for learning, it made a number of students feel welcomed and cared for by other non-disabled students and other peers with disabilities. Ultimately, students could be recognised, respected and feel dignified. The feeling of being accepted was instrumental in students being able to appreciate TVET as it translated to formal access. Formal access in the sense that nearly all students felt that the impediments such as not being accepted and treated well would have hindered their learning experience. Therefore, barriers that impeded learning were removed. Removal of barriers also meant that a good number of students were allowed to enrol in TVET, especially when special dispensations were made so that they may qualify to attend TVET. One of the dispensations was that students were not subjected to entrance examinations. Box 2 demonstrates examples of these experiences as expressed by participants.

Whilst some students experienced the above-said inclusionary practices, many also experienced forms of exclusionary practice. This is not surprising as literature reviews indicate that where there is inclusion, there is also exclusion (Ainscow, Booth & Dyson 2006). Figure 2 demonstrates exclusionary practices and its descriptors as identified from data analysis.

At times some students felt epistemologically excluded when certain lecturers failed to help them engage with content. For example, some students were afraid to tell their lecturers when they had a problem with content because certain lecturers were impatient with students. Some lecturers used abusive language, which resulted in students not wanting to come to class or feeling intimidated. In such instances, affected students revealed that they resorted to not asking questions and not participating in class. Box 3 captures student expressions of these experiences.

Some students reported that one particular lecturer used to suspend teaching them as a punishment when she found fault with them. They experienced this as an unfair punishment measure. Students believed that this suspended teaching was used for them because they were ‘special needs’ students. Students cited incidents where they were punished for not knowing that there was a class by coming to class and then not being taught. At times, students also felt silently excluded because they were in class and college and yet not epistemologically benefitting. Silent exclusion (Lewin 2009) refers to enrolling students in schools, yet little learning is taking place in the classroom. This results in constrained engagement with the curriculum as students are physically present but not supported to access knowledge (Walton & Bekker 2013).

Another exclusionary practice experienced by students was marginalisation. Data showed that students were denied free movement in the college as all participants reported that they were restricted to the hostels only and particularly barred from making friends with students without disabilities. Whilst this restriction occurred in the college, it was, however, a practice of a particular member of the department. She threatened students without disability that they would be taken to task if they befriended students with disabilities because it was ‘not allowed’. In contrast, students with disabilities were threatened that they would be expelled from college if they were seen with non-disabled students. The fear that was instilled in students resulted in them keeping to their hostels. From a Capability Approach perspective, this kind of practice meant that students were denied the capability to take control of their environments (Nussbaum 2000) as they had no freedom of association.

They were also not allowed to go outside of college independently. Students were only allowed to leave campus with the permission of their parents who then needed to come and collect them from college. This practice was not welcomed by students who felt that they were caged, and their independence was taken away from them. This is also contradictory to what they are being taught in curriculum regarding independent living. At the same time, students felt that they existed between two personas as, when they are outside college, the society treated them as abled persons whilst in college where they are supposed to grow, they were treated as ‘babies’ or ‘little mice’. Box 4 provides examples of students’ expression of these experiences.

Besides being deliberately restrained by the department rules, some students also reported the use of language of marginalisation. They were called by names by some lecturers and other students without disabilities. These name callings affected the students’ self-esteem as they reported that they saw themselves as those names and ended up believing that they are indeed what those names mean. For example, being called ‘special need’ meant that they are exceptionally not normal hence ‘special’. According to McDaid (2008), often people with disabilities are perceived in terms of the challenges they have without considering their overall contribution in the society and many times they are understood as dependent on others as well as incompetent. Labelling or not labelling in the field of special education is debatable. However, the Special Educational needs and Disability Act of 2001 (SENDA) and the Special education law of 2004, which was renamed the Individual with Disabilities Education Act (IDEA), allow labelling if they are meant to categorise an individual so as to address their educational needs for their individualised learning programs (Boyle 2014; Boyle & Sharma 2015). On the contrary, given the stigma generally allied with labelling individuals, it has been argued that labelling usually ignores the need of the individual and focuses on pathological tendencies (Goodley 2001). This suggests that labelling or categorisation is supported in legislation, where the intention is the provisioning of support, but that the practice of doing so may result in the experience of stigma for individuals so labelled.

Whilst labelling in inclusive education is not something new (Boyle 2014; Boyle & Sharma 2015), we argue that the policy landscape and school system socially constructed the labels, which the collegiate community adopted and used to refer to these students. In other words, the use of the word ‘special needs’ became an institutional practice, which was exclusionary in nature. Becker’s (1963) labelling theory suggests that labels serve as special markers, which are used to identify and classify particular students in institutions or society. Although labelling to other scholars (Booth & Ainscow 1998) is to recognise differences and confer social salience to those differences (Arishi et al. 2017), in the context of this study, some students experienced labelling as a discriminatory practice despite it being used by the department for the purpose of special educational provision. Studies have shown that marginalisation goes hand in hand with stigmatisation. A study by McDaid (2008) showed that professionals are also amongst the populations that stigmatise within their profession. This was the case with this study, where students were stigmatised by their lecturer who dehumanised them verbally, physically and emotionally. Evidence of this is provided in student statements in Box 5.

Some students were labelled and deemed incompetent. One of their lecturers communicated low expectations of them instead of having a positive attitude, which could have countered collegiate community negativity. The stereotyping that some students experienced resulted in them being categorised and pigeonholed. This labelling increased the distance between provisions that students were afforded. For instance, some students without disabilities treated students with disabilities as not equal to them. The behaviour yielded sexual abuses with the belief that students with disabilities would not be able to reason and identify their perpetrators. Not only was a student with disabilities sexually abused, but some students with disabilities were also taken advantage of in terms of their possessions. Furthermore, outside people masquerading as churches and insurance people wanted to rob them of their allowances. Box 6 shows some of the excerpts that demonstrate these experiences.

Negative experiences such as these hindered students’ well-being achievement by limiting their opportunities for expansion of capabilities. It becomes clear that both inclusionary and exclusionary practices were experienced by students with disabilities in TVET. Of the two practices, exclusionary practices were noted more than inclusionary practices in terms of references made by students during interviews. This suggests that TVET is struggling with the implementation of inclusive education suggesting that inclusive policy is not fully realised in practice.