Pre-service teachers’ experiences of schooling: Implications for preparation for inclusive education

Teacher educators in


Introduction
Teacher educators face the challenge of preparing new teachers for a more socially just education system that is envisaged in policy but that is not yet fully realised in school contexts. While substantial progress has been made towards improving infrastructure and resourcing of South African schools, historical inequalities persist in the basic education system even two decades after the end of apartheid (Maringe & Prew, 2015). Exclusion persists at many points of the education system and takes various forms, including early school leaving (Weybright, Caldwell, Xie, Wegner & Smith, 2017), low literacy and numeracy levels (Pretorius, 2015;Roberts, 2017) as well as the widespread exclusion of learners with disabilities (Majoko & Phasha, 2018). Our concern is the extent to which pre-service teachers have come to regard prevalent, potentially exclusionary structural arrangements as unproblematic or even beneficial. If so, the normalised conceptions of schooling they hold might very well constrain attempts to prepare teachers who can contribute to a more equitable and inclusive schooling system. In this article we report on the results of an empirical study in which we investigated how a cohort of pre-service teachers perceived arrangements of learner support that entailed fully or partially segregating learners who experience barriers to learning from their peers in ordinary classrooms. On the basis of our findings, we argue that a crucial part of pre-service teacher education curricula should consider current schooling structures and their potential for exclusion and marginalisation of those with barriers to learning.

Literature Review
Under apartheid, segregated special schools were established for (particularly White) learners identified as having "special needs." Urban-based teachers were encouraged to identify learners requiring additional learning support, and refer them to a nearby special school, or allocate them to "aid classes", where these existed. In other schools, the option for learners with barriers to learning was for learners to spend some of the school day in ordinary classes, but then received "pull-out support" for part of the day in a separate class with a remedial teacher or other therapist/s (Walton, Nel, Hugo & Muller, 2009). Very few special schools were built in rural areas and as a result, teachers working in schools outside of major cities were not easily able to refer learners with additional support needs to nearby special schools (Nkabinde, 1993). By default, learners with disabilities were generally accommodated in local ordinary schools, without necessary support, or they were excluded altogether (Department of Education [DoE], 2001;Gardiner, 2008).
South Africa has sought to transform education through the introduction of inclusive education into ordinary schooling (Themane & Thobejane, 2019). The model of inclusive education proposed by White Paper 6 has a range of services, where separate special schools remain for learners deemed to have high support needs (DoE, 2001). The assumption of the benefits of segregated special education provision has been challenged internationally. In her synthesis of the international literature over several decades, De Bruin (2020:58) shows the social and academic benefits of inclusive classrooms for learners with and without disabilities. She argues that " [c]ollectively, this body of research refutes the claim that students with disability are better off in segregated special settings." Tchombe (2017:23), the United Nations Educational, Scientific and Cultural Organization (UNESCO) Chair for Special Needs Education concurs, saying that "[m]aintaining some pupils in special schools and support classes still seems discriminatory and cannot be justified on the basis of equity." South Africa, like many other African countries, is a signatory to the United Nations Convention on the Rights of Persons with Disabilities (UNCRPD). General Comment 4 of the UNCRPD makes it clear that segregation within educational settings should be ended, saying that "[t]he right to non-discrimination [for learners with disabilities] includes the right not to be segregated" (United Nations, 2016:13).
Teacher education is seen as an important determinant of the successful implementation of inclusive education in South Africa (DoE, 2001;Majoko & Phasha, 2018). Policy requires that initial teacher education (ITE) programmes should ensure that newly qualified teachers "understand diversity in the South African context in order to teach in a manner that includes all learners" (Department of Higher Education and Training, Republic of South Africa, 2015:64). However, the pre-service teachers who are expected to contribute to the realisation of a more just and inclusive education system are themselves the product of a schooling system that still reflects many historical inequalities and marginalising practices. Compounding this, recent studies show that few pre-service teachers have a positive attitude towards inclusive education (Majoko & Phasha, 2018;Sze, 2009).
Studies have shown significant differences between the conceptual underpinnings, the focus and the structure of curricula of higher education institutions (HEIs) offering teacher education (Council on Higher Education, 2010;Deacon, 2016). While some ITE curricula adopt a largely psychological or medical-based approach to individualised learner support, others situate inclusive education in terms of a social justice agenda within a sociological frame (Rusznyak, 2016). Some institutions tend to emphasise practical knowledge and classroom interventions, while others focus more on the theoretical underpinnings of inclusive education (Walton, 2017). More recently, the sector has collaborated to develop standards for inclusive teaching for beginner teachers in South Africa (British Council, 2019;Walton & Rusznyak, 2019). The first of these standards emphasises that beginner teachers should develop "agency for social justice and inclusion." This requires that they can "identify attitudes and practices that exclude or marginalise learners" and understand "their global and local histories of exclusion." Through such understanding, preservice teachers should come to see the "development of inclusive education as a response to exclusionary practices" (British Council, 2019:7). We were interested in establishing the extent to which prospective teachers, who themselves had been schooled in particular contexts, held predetermined ideas that potentially affected their preparation for teaching in an unequal but transforming education system.

Conceptual Framework
A prevalent challenge for teacher educators has been how to work with the tacit assumptions about teaching that prospective teachers form through their own schooling. Their apprenticeship of observation (Lortie, 1975) is said to form their first socialisation into teaching. The development of students' apprenticeships of observation is personal and individualised, and assumptions about the nature of teaching depend on the specific set of teachers they had interacted with and practices they had observed during their schooling. Much of the research around the apprenticeship of observation deals with abstracted assumptions about the ways in which pre-service teachers tend to underestimate the complex nature of teaching (e.g., Hammerness, Darling-Hammond, Bransford, Berliner, Cochran-Smith, McDonald & Zeichner, 2005;Pugach, 2006). While the apprenticeship of observation familiarises teachers with their future work context, it offers them an incomplete understanding of teaching practices (Pugach, 2006). Observing the visible routines of teaching does not provide prospective teachers access to the professional knowledge and reasoning that "allows the selection and implementation of different strategies that will support learning for different purposes and different students" (Hammerness et al., 2005:368). International studies (e.g., Tanase & Wang, 2010;Westrick & Morris, 2016) and local studies (e.g., Botha, 2020;Rusznyak & Walton, 2014) have investigated how these tacit assumptions can affect student learning in teacher preparation programmes. Unexamined, such beliefs have the potential to act as a filter to the learning offered by ITE programmes and constrain student teachers' development of effective teaching practices (Hammerness et al., 2005).
Researchers (e.g., Mewborn & Tyminski, 2006;Patchen & Crawford, 2011) argue that preservice teachers should be given opportunities to articulate and critically interrogate their experiences of schooling and their beliefs about the nature of teaching. Without such interventions, newly qualified teachers can easily and uncritically revert to "delivering education in the same manner in which they were taught" (Borg, 2004:274) thereby replicating prevalent practices irrespective of their merit. In university-based ITE programmes, a variety of interventions have been designed by teacher educators to this end (Boyd, Gorham, Justice & Anderson, 2013;John, 2013;Rusznyak & Walton, 2014).
An aspect of the apprenticeship of observation that is not often discussed in the literature is its contextual specificity. It cannot be assumed that students have a shared experience of schooling and can easily imagine how teaching and learning is enacted in different parts of the schooling sector. Although ethical principles and specialised knowledge may guide professional judgment in practice, teachers need to consider contextual factors in addition to considerations of subject content and learner attributes. In as much as teachers' practices are informed by a knowledge-base and notions of professional teaching, Carrim (2019) argues that they are equally informed by the multiple and intersecting identities of teachers as social beings, and the opportunities and constraints to what is possible in the contexts in which they work. Education systems are not politically and ideologically neutral structures that operate outside of society's struggles for social justice (Carrim, 2019). It is, therefore, insufficient for pre-service teachers to learn how teaching practices are enacted in decontextualised ways. We concur with Christie (2018:xxvi) that an imperative of pre-service teacher education programmes should be to "deepen student teachers' consciousness of their situation, so as to understand it as a historical reality that requires transformation." This would require that ITE programmes establish conditions in which preservice teachers understand and are committed to upholding the rights of all students to fair treatment, providing equitable access to learning opportunities and upholding anti-discrimination principles in their classrooms (Carrim, 2019).
To prepare teachers for diverse schools within the context of a post-apartheid South Africa, there have been several interventions that seek to introduce pre-service teachers to unfamiliar schooling contexts (Amin & Ramrathan, 2009;Nkambule & Mukeredzi, 2017;Pennefather, 2008;Robinson, 2014;Rusznyak & Masinire, 2018;. In many of these studies, pre-service teachers were given opportunities to undertake observations or teaching practicums in different contexts to the ones in which they were schooled. The focus of experiences like these is often an outward gaze: how the school contexts are different from what they had experienced, and how the goals of teaching and learning are enacted differently with differing resourcing implications. Attention to diverse schooling contexts does not necessarily present opportunities for pre-service teachers to critically examine their own assumptions about teaching, nor their normalised views of schooling (Ellis, 2010).
In thinking about teacher education within a post-apartheid South Africa, Christie (2018:xxv) observes that "the obvious structural inequalities in the state's education system -in learners' experiences, opportunities and outcomeshave become normalised or regarded as simply too difficult to tackle." Simply understanding the "other" and not considering the limitations of one's own perspective is inherently conservative and may ultimately undermine well-intended transformation initiatives. Relatively few interventions provide explicit opportunities for pre-service teachers to interrogate the taken-for-granted assumptions inherent in their own experiences of schooling. Amin and Ramrathan (2009:73) argue explicitly not merely for acquaintance within a different context, but for the need to "reframe memories" about schooling and to "disrupt their suppositions about schools based on their biographical experiences." We deem it useful for teacher educators to interrogate how existing structures of schooling are implicated in the way pre-service teachers view the benefits and/or drawbacks of prevailing school arrangements.

Methodology
In this article we report on a multiple methods study (Spicer, 2018) in which we investigated relationships between the schooling structures experienced by pre-service teachers and their perceptions about models of learner support. Having obtained ethical clearance for the study, a cohort of 550 first-year pre-service teachers enrolled for a 4year Bachelor of Education degree were invited to participate in this study. Of those invited, 159 gave their consent to participate. The study was conducted before students had received any formal instruction on the history of the South African education system, or coursework covering concepts of exclusion, inclusion, marginalising practices or inclusive education. In the first phase, using quantitative methods, we surveyed the experiences and perceptions of participating pre-service teachers through a self-completion questionnaire (Bryman, 2012). The questionnaire presented participants with the following arrangements that schools might make in meeting the diverse needs of learners.
• Arrangement 1: Learners who experience learning difficulties i are removed from ordinary classes and referred to a separate "aid" class, or they are transferred to a special school. • Arrangement 2: Learners who experience learning difficulties remain in ordinary classes but are withdrawn from the class for a part of the school day to work with a remedial teacher or learning support educator, or to receive treatment from a therapist.
Participants were asked whether they had observed or experienced any one or more of these arrangements during their own schooling. This required a "yes" or "no" answer. They were then asked to articulate what they would regard as the benefits and/or drawbacks of each of these arrangements. The questionnaire thus generated both quantitative and qualitative data which were analysed as the basis for the second phase of the study. Here we present our analysis of participant responses to the first two of these arrangements.
In the second phase we sought further understanding and explanation of the identified patterns (Spicer, 2018) which involved in-depth interviews with a small group of participants. Twenty respondents were randomly selected from the questionnaire respondents and invited to participate in individual face-to-face interviews.
Only 12 respondents accepted the invitation. The interviews were semi-structured with open-ended questions (Bryman, 2012).
Responses to the questionnaire were clustered according to those who had reportedly observed or experienced each of the arrangements of learner support during their own schooling, and those who had not. We then coded participants' open-ended responses according to whether they perceived benefits only, drawbacks only, or both benefits and drawbacks in relation to each of the arrangements. The results were statistically analysed using the Χ 2 test (Seale, 2018) to see whether there were significant differences in responses between participants who had observed or experienced these arrangements and those who had not. Fisher's exact test was used for 2 × 2 tables or where the requirements for the Χ 2 test could not be met. The strength of the associations was measured by Cramer's V and the phi coefficient respectively. The 5% significance level was used. The following scale of interpretation was used: • 0.50 and above strong association • 0.30 to 0.49 moderate and significant association • 0.10 to 0.29 weak but significant association • below 0.10 little if any association This part of the data analysis was done by a professional statistician using SAS version 9.4 for Windows.
Our analysis assessed whether the participants attributed benefits and/or drawbacks to each of the arrangements under consideration, and to whom the benefit or drawback was directed. For example, when a benefit was attributed to an arrangement, we analysed whether the perceived benefit was attributed to the learner who experienced learning difficulties, the teacher, or the other learners in the class. We used thematic analysis (Grbich, 2013) to identify the nature of the benefit or drawback.
The study was designed and executed with acknowledgement of several limitations. Notwithstanding the recognised constraints of the questionnaire and individual interview as data collection methods, and the small sample size that limits generalisability, we had to engage with other issues that might influence the findings. These include the fact that we chose not to collect biographical data from participants so that their anonymity would be guaranteed. This means that the data could not be disaggregated by race, gender, or the type of school the participants had attended. Secondly, two of the authors were, at the time, teacher educators at the institution where the data were collected. Various measures were taken to reduce the possibility of coercion to participate, and interviews were conducted by postgraduate students who had no relationship with the participants.

Findings
In this section, we show how participants' observation or experiences of these arrangements correlated with their perceptions of their respective drawbacks and benefits. We draw on the results of a statistical analysis alongside data from open-ended questions in the questionnaire and interview. We show that pre-service teachers who were familiar with the practice of transferring learners with learning difficulties to segregated special schools or classes were absolutely convinced of its benefits. Those who were familiar with the practice of removing learners from ordinary classes for a part of the school day for dedicated support were harshly critical of the practice. By contrast, those who had not observed these arrangements were better able to articulate both potential drawbacks and benefits.

Arrangement 1: Segregation of Learners with Learning Difficulties
Fifty five per cent of participants reported that they had experienced the arrangement of learners with learning difficulties being removed from ordinary classes and placed in separate special schools or separate "remedial" or "aid" classes. Statistically, there is a weak but significant association (p = 0.036; phi = 0.21) between whether or not students had encountered this arrangement and their perception of whether it was beneficial, detrimental or both. As shown in Figure 1, 98% of the participants who had experience of learners being removed from ordinary classes attributed a range of benefits to the arrangement. None who had experienced this arrangement specified any drawbacks, as summarised in Figure 1 below.

Figure 1 Summary of findings for Arrangement 1: Segregation of learners with learning difficulties (n = 91)
The data reveal that participants held strong opinions about learners who belonged in ordinary schools and those who did not. Just over half (51%) of respondents to the survey used the word "normal" to describe learners who cope well with the challenges of schooling. "Normal" learners were variously described as those who "academically achieve average or above average results"; those who "answer questions correctly" and who are "on the same level [as us] academically and emotionally." Participants were convinced that these were the learners for whom the ordinary school system should cater. Furthermore, participants were convinced that teachers were neither willing nor capable of teaching learners with diverse learning needs and should not be expected to teach all learners in ordinary classrooms. In the questionnaire, two responses revealed a view that those learners who needed additional support were "taking up teachers' time" thereby disadvantaging other learners in the class. One participant explained how "teachers cater to the majority and not the minority without holding back the majority." Others justified the benefit of segregated learning spaces as "remedial teachers have more time to spend on the difficulties without worrying about wasting other students' time" and consequently that learners with learning difficulties "tend to be ignored" and "teachers are indifferent to those who could not keep up [the pace]…." One participant explained her view that "it's the job of a remedial teacher to help these children but a normal teacher teaches normal children. They can't consider the academic or emotional problems that they have." Teachers in ordinary classes "lack patience" and are perceived to be "unwilling to work with learners with difficulties." An adoption and commitment to inclusive pedagogies requires that teachers accept difficulties in learning as professional challenges for teachers, rather than deficits that reside in learners (Black-Hawkins & Florian, 2012). However, these responses reveal a concerning belief that learners with barriers to learning should not be the responsibility of class teachers in ordinary schools.
Some participants argued that some learners were unable to achieve academic success within ordinary classrooms. Their discourse suggests the entrenchment of deficit and potentially marginalising attitudes towards learners who experience challenges with learning. Participants assumed that in segregated spaces there would be benefits including slower pacing of lessons, and greater amounts of individual attention. For example, a participant explained the benefits as being "taught at a pace that they are comfortable with" and that learners "don't have to feel rushed and discriminated against when they do not understand something." Participants in our study were also convinced that in being segregated from other learners there would be self-esteem benefits for learners "with special needs." Segregated learners were perceived as benefitting because they would not "feel as insecure" because "others around them are now at the same level." Segregated learning spaces, said one participant, "would help these learners to feel less stupid" and "would feel safe because no one is making fun of them." A participant explained how learners in these spaces could "feel more comfortable to do the things like reading, because if they did it in front of the Benefits only 98%

Percentage of those who had observed/experienced this practice
Not sure 2%

Percentage of those who had NOT observed/experienced this practice
Not sure 7%

Drawbacks only 5%
Both benefits and drawbacks 2% [ordinary] class, the class would laugh at them." If pre-service teachers remain convinced that inclusion constrains the achievement of all learners within ordinary classes, it is unlikely that they will develop a belief that all learners have the capacity to learn and make progress (Black-Hawkins & Florian, 2012). At the start of their ITE, pre-service teacher participants seemed convinced that there was a specialised pedagogy available in separate special classes that was different to the pedagogy required in ordinary classrooms. Their belief in this "secret pedagogical knowledge" provides justification for their belief that learners who experience learning difficulties may be better off in segregated learning spaces (Walton & Rusznyak, 2013:117). Forty-six questionnaire respondents (29%) assumed that in segregated settings (like in aid classes or special schools), teachers "understand what they are doing"; are "specialised"; "trained for those kids" and "teach them differently compared to a normal school." When asked to elaborate on the special pedagogies of these teachers in the interviews, participants perceived that content could be explained in "more accessible ways" than in ordinary classrooms. Differences also included "more one-on-one" and the "teaching style will be different." There was an assumption that in segregated spaces learners would receive "more attention and help" and "won't fall behind." Instead of regarding the inclusion of all learners in ordinary classrooms as a realisation of their educational rights, they regarded the provision of support as an "additional extra" that falls outside the pedagogies and professional responsibilities of ordinary teachers. The myth of this secret pedagogical knowledge of special education teachers has been revealed in studies both in the South African context (e.g., Walton & Rusznyak, 2013) and internationally (e.g., Lewis & Norwich, 2005). Unless this belief is exposed and critically interrogated, it is unlikely that teachers with these entrenched beliefs would accept the learning challenges faced by learners as a professional pedagogical challenge in their daily classroom practices (Black-Hawkins & Florian, 2012).
Although a large portion of participants (86%) who had not experienced the arrangement in their own schooling also regarded it as beneficial, a higher percentage of participants articulated drawbacks associated with the arrangement or were uncertain about its benefits. Of the pre-service teacher participants who had not experienced the segregation of learners with learning difficulties, 14% were unconvinced that this arrangement offered identifiable benefits. They noted the potentially detrimental impact of segregating learners, and the effect on their self-esteem on being visibly identified as needing remedial supporta point confirmed by Taylor (2014).

Arrangement 2: Learners with Special Needs Separated for Part of the School Day
Eighty per cent of respondents to the survey reported that they had observed or experienced learners who had been withdrawn from ordinary classes in order for them to receive support with a therapist or remedial teacher for part of the school day. The statistical analysis shows a very strong and significant association between whether or not participants had encountered this arrangement during their own schooling and their perceptions of its benefits and/or drawbacks (p < 0.0001; phi = 0.91). Figure 2 shows that most students (78%) who had not observed this practice were unsure about whether this practice could be beneficial or had drawbacks. In contrast, 94% of participants who had encountered this arrangement perceived only drawbacks, compared to 3% of those who had no personal experience of it. Of the 126 pre-service teachers who had observed this arrangement, only seven (6%) of the participants perceived benefits, whereas 119 (94%) perceived only drawbacks. Of the 32 participants (25%) who had not observed or experienced this arrangement, 19% thought that there were benefits. Only 3% articulated drawbacks, with the vast majority (78%) noncommittal, unsure of whether they thought that the practice was beneficial or not. By way of contrast, participants who had experienced this were vehemently opposed to the arrangement, professing concerns for learners who were removed from ordinary lessons.
The concerns expressed by participants who had observed this arrangement centred on the highly visible labelling that occurs when particular learners are removed from ordinary classes for support lessons. Their periodic attendance means that learners who are removed are not accepted as fully legitimate members of the class. Several participants recalled how their teachers had labelled learners with learning difficulties. For example, a participant remembered how his teachers would say, "Come here my slow students", [and to others], "You are my geniuses, you go over there." These observations had led learners to "grow up knowing that you have been classified." In these incidents, the capacity of learners to achieve was not only pre-determined by teachers but also made public to all other members of a class community. Using derogatory labels openly in the classroom environment paved the way for "slow students" to be "teased" or "made fun of" by other students. The participants thought that prevalent bullying would damage the self-esteem of those learners who faced difficulties within ordinary classes. One participant described how her teachers had "tried to instil tolerance" but she felt that ordinary classroom environments were "dependent on the tolerance of the learners" and that the bullying behaviour of learners were "out of the teacher's control." Very few participants questioned the bullying culture that they had encountered. Instead of motivating for learning spaces that were safe for all, pre-service teachers supported the removal of targeted learners from ordinary learning spaces.

Discussion
We make three assertions based on these findings. The first is that pre-service teachers' schooling experiences do more than shape conceptions about the nature of teaching, as the apprenticeship of observation literature suggests. Schooling experiences also shape ideas about various (potentially exclusionary) schooling arrangements. Pre-service teachers who have not encountered learning support arrangements are much more likely to consider potential advantages and disadvantages of these arrangements and to take a more tentative, nuanced perspective. By contrast, those who have observed or experienced particular arrangements, held significantly more entrenched views about their merits. As long as prospective teachers continue to accept or reject practices based on their personal experiences rather than on research evidence and critical analysis, they are unlikely to commit themselves towards inclusive teaching practices of all learners in their classes.

Drawbacks only 94%
Both benefits and drawbacks 0%

Drawbacks only 3%
Both benefits and drawbacks 0%

Percentage of those who had not observed this arrangement
Our second assertion echoes Slee's (2011:50) contention that attitude should not be "decoupled from context." The findings from our study confirm some of the negative attitudes and beliefs that Majoko and Phasha (2018) found among South African pre-service teachers, but it also shows that these attitudes and beliefs are engendered by experiences of schooling arrangements in particular contexts. Their experiences contribute to entrenched and normalised beliefs about schooling. Beliefs about who belongs in ordinary classrooms and the extent to which teachers could/should take responsibility for ensuring that learning opportunities are available to all learners in class has already been established in the minds of pre-service teachers. These beliefs may be antithetical to the aims and ideals of a more equitable and inclusive education system.
Thirdly, exposing pre-service teachers to teaching and learning in unfamiliar contexts is important but insufficient to prepare teachers for the complexities within diverse South African classrooms. Our findings suggest that they would continue to observe unfamiliar school contexts through the normative gaze of their own schooling experiences. To contribute to a critical awareness of exclusionary prejudices, such experiences would need to focus both on the unfamiliar context, and what it revealed to them about their own assumptions of what "normal" schooling should look like. There is an urgent need to subject these assumptions to critical interrogation with a view to disrupt them.

Conclusion
Teacher educators have long recognised the importance of interrogating the assumptions that pre-service teachers bring to ITE about classroom teaching, and providing them with the analytic and conceptual tools to broaden their perspective. In this article we broaden this perspective to reveal that preservice teachers' experiences of schooling structures correlate with sets of beliefs that can influence their commitment to inclusive teaching practices. We return to Christie's (2018:xxv) question of what teacher education programmes could do that would enable pre-service teachers to identify prevalent structural inequalities so that they "are able to engage ethically to work with and against them." To advance social justice, inclusive education cannot be taught as a decontextualised and ahistorical policy initiative that has optional classroom applications. Our findings suggest that adoption and commitment to inclusive pedagogies is unlikely unless preservice teachers recognise that they may hold assumptions that normalise structures in schooling that marginalise and exclude learners. Formal and explicit opportunities are needed in ITE programmes for pre-service teachers to interrogate and disrupt beliefs that have been engrained by many years in exclusionary schooling structures. Our findings support calls for formal coursework that critically examine current schooling arrangements and their potential for exclusion and marginalisation . This would create dedicated spaces for pre-service teachers to examine their normalised assumptions about the structures of schooling. It could also prompt them to examine the extent to which the practices that are so familiar to them may contribute to the exclusion of vulnerable learners.