An Investigation into Factors that Contribute to Learner Migration in South African Schools

The purpose of this study was to investigate the various factors which contribute to learner migration in South African schools and to suggest ways in which learner migration can be addressed. A qualitative research approach was used for the study and data was collected by means of individual – and group interviews. The sample of the study consists of one Expert Management Information System (EMIS) member and one principal from a secondary school in each of the nine provinces as well as four parents and four learners from four of the nine provinces. The final sample were 50 participants (n=50). Individual semi-structured interviews were conducted with the 9 EMISmembers and 9 principals, while focus group interviews were conducted with the 16 parents and 16 learners. The findings of the study correspond with the reasons for learner migration in terms of Bronfenbrenner‟s bio-ecological theory model. This model is firmly grounded in reality, thereby allowing the researchers to determine how stakeholders are influenced in terms of learner migration. The findings of the study indicate that learners are migrating to other schools due to, inter alia, the new legislative framework, school management and leadership practices of principals, school effectiveness issues, infra structure of schools as well as socio-economic factors. The study underscores the need for structuring the South African learner migration system that is guided by a coherent system of policies that enables principals to manage learner migration effectively.


Introduction
It is clear from the research literature that there is no single definition for learner migration. According to Hornby et al. (2009), learner migration can be viewed as the movement of large numbers of school learners from one school to another. Internationally, the main reason for this process is in most cases to get access to `better" schools to enable learners to gain knowledge and skills through effective education. In the South African context, authors such as Neluvhola (2007), Lancaster (2011) and Machard (2014) maintain that learner migration refers to the voluntary movement of learners who migrate from predominantly black township schools to urban schools, that is, former Model C and independent schools which primarily served white learners during the Apartheid period prior to 1994. Pampallis (2003), expands on this definition by stating that learner migration refers to the movement of learners from black townships on the periphery of cities to former White, Coloured and Indian schools situated in relatively more affluent areas. This takes place as a result of the belief of specifically black parents and their children that these schools offer a better quality of education. Sekete et al. (2001), agree that the concept of learner migration refers to learners migrating from their home areas to schools outside their township or neighborhood. According to these authors, learners who move to other schools do this without much information of what learner migration entails. As a result, poorer schools in townships are often abandoned as migration leads, inter alia, to poor resources, poor discipline and poor academic performances.
Learner migration is a problematic phenomenon that emanates from dissatisfaction experienced by parents and learners and require an investigation into the root cause of migration. This seems to entail many contributory factors. These factors need to be interrogated to enable the various stakeholders in the education system to address the complex phenomenon of learner migration.
In this regard, Neluvhola (2007) adds that the migration of learners from township schools, which are predominantly black, to urban former Model C schools, that were predominantly white in the previous political dispensation, constitutes a major form of learner migration. Kruger (2002) emphasizes that migration is also rooted in poor matric results, the lack of discipline, low teacher morale, as well as other educational problems in some schools in South Africa which have led to a general drive or exodus of learners from schools across the country. This was confirmed by a study conducted by Neluvhola (2007) that revealed that better matric results, extra-curricular activities, discipline, respect and infrastructure in former Model C and independent schools constituted `pulling factors" from township schools to former Model C and independent schools in selected schools in Limpopo province. Botha (2018), acknowledges that the ineffectiveness of a school culminates in poor outcomes of examination results. Hence, it leads to learner migration to schools that are perceived to develop learners" learning skills and knowledge more effectively. Maile (2004), as well as Bischoff and Koebe (2005) argue that learners migrate to the schools outside their neighborhoods in order to acquire knowledge and skills that will allow them to earn a living. It has also been known that ineffective leadership and poor management lead to total collapse of educative teaching which translate into learner migration (Botha, 2018;Neluvhola, 2007).
Another contributory factor to learner migration is urbanization. Neluvhola (2007) argues that increased urbanization of black families experienced in Limpopo province after the ANC government came into power and all apartheid laws were scrapped, including the Group Areas Act (Republic of South Africa RSA, 1944b), resulted in many black families moving to white suburbs. As a result, learners migrated to schools that had to open their doors to learners of all races. Sekete et al. (2001) and Lombard (2009) believe that learner migration is experienced in all the nine provinces of South Africa and relates to all ethnic sectors of South African population. This view is based on findings undertaken in Gauteng province.
 Legislation and quality education (Neluvhola, 2007;2017).  Transformation in education (Lombard, 2007).  Township youth cultivating their "aspirational dispositions" (Lancaster, 2011).  Commuting to quality schooling (Machard, 2014).  Low achievement due to poor management (Mbokazi, 2015). Against the background of the above information, there is a need to conduct in-depth research to minimize or curb learner migration as far as possible since it could have a disruptive influence on principals of schools, parents and learners. This study is significant because it highlights reasons for as well as challenges and benefits of learner migration in the South African education context.

Theoretical Framework for the Study
A theoretical framework of a study is a theory that is used as a structural basis which underpins research (Ben-David, 2011). Dreyer (2018), defines a theoretical framework as a basis for making connections between new research and existing theory. In this study, the researchers explore Bronfenbrenner"s bio-ecological model from 1979 as theoretical framework that can be used to understand the phenomenon. This theory is based on his ecological model which provides an academic base for this study and explains how the individual child is influenced by and interrelates with the systems of his/her environments. The child is therefore studied in the context of development, which lies within the context of the family which, in turn, lies within the context of the community (Bronfenbrenner, 1979;Mischo, 2014).
Bronfenbrenner"s bio-ecological theory is relevant to learner migration since it shows the importance of acknowledging that the various intertwined environments influence learners to migrate from one school to the other. In this study, various influences of the ecological-model and the relevance of different aspects at each environment surrounding the child are related to learner migration to eventually arrive at recommendations in terms of the phenomenon of learner migration in South Africa.
This framework enables us to gain an understanding of how various levels of environments, namely the microsystem, the exosystem and macrosystem influence learner migration. Although this model was intended to focus on factors that affect the development of children according to certain spheres which surround the child, it is applicable to contextualize learner migration.
Bronfenbrenner"s bio-ecological model is not only used to explain the reasons for learner migration at various levels (or spheres) that surround the developing child during the child"s development, but also to assist in planning intervention strategies that can be used to minimize learner migration.  (Bronfenbrenner, 2005) The innermost layer represented in Bronfenbrenner"s bioecological model, called the microsystem, is the closest environment which surrounds the developing child and influences him/her directly. For the purpose of this study it would refer to factors that directly relate to affecting learner migration. It comprises, inter-alia, the home and the school where the child spends most of his/her time. The so-called mesosystem portrays interaction between components of the microsystem. The second level of the model is the exosystem environment, which surrounds the microsystem. It influences the child in a more indirect manner. The system consists of, inter alia, the School Governing Body, the Circuit offices and District offices of the Department and the Provincial Department of Basic Education. The outermost level of the model is known as the macrosystem which comprises of cultural general values, customs and laws. In this study acts and laws in the macrosystem that influence learner migration are also particularly relevant.
Bronfenbrenner"s bio-ecological theory is beneficial in providing insight into various factors that play a role in the growth and development of a learner. The latter would include phenomena such as learner migration, which constitutes a major change in the life of a learner. Viewing learner migration from the perspective of the bioecological model acknowledges that different environments and their interrelationships would influence learner migration.
The bio-ecological model shows that an individual"s interrelated surroundings at the micro-, exo-and macrosystem level offers a meaningful framework for interrogating learner migration in terms of interlinked social strata. The components of especially the microsystem, which have significant and direct influence on the learner as a result of its powerful mesosystemic action between its components, including the family at home, the school, peers, the neighborhood and the church, could play a significant role in effecting learner migration. The microsystem is, however, also linked with the exo-and macrosystems. It could therefore be said that all the spherical systems of Bronfenbrenner"s bio-ecological model would bear relevance to the phenomenon of learner migration, enabling the researchers to obtain a more holistic view on this issue. The chronosystem which denotes prominent internal and exchanges in the life of the learner also has the possibility of playing a key role in terms of learner migration.

Research Design
In this study the researchers employ the qualitative approach in order to gain an in-depth understanding of the root causes of the problematic issues of learner migration. Qualitative research involves the study, use and collection of a variety of empirical methods such as observation, personal experiences, case studies, life history and interviews (Denzin and Lincoln, 1994). Flyvbjerg (2006), describe the case study as a method that "can close in" on real life situations. A multiple case study was used since views from relevant participants associated with one school in each province (which constitute a case study) are obtained. Furthermore, a case study tests theory against lived reality and enable the voices of the participants to be heard (Mendaglio, 2003).
The researchers wishes to convey the analyzed messages from people who have either experienced learner migration directly or who have dealt with affected parties in each province who have experienced learner migration before.
Data was collected from participants using semi-structured individual interviews with principals from public secondary schools and from EMIS-officers in each of the nine provinces. In addition, focus group interviews were conducted with parents and learners. Due to some logistical constraints a focus group interview for parents and learners were conducted in only four provinces. The use of these methods allowed for triangulation which contributed to reliability and validity of the study. Some of the questions raised during the interviews were (Cohen and Morris, 2011):  What are your reasons for migrating from your previous school?  What do you think are the main reasons for learners migrating from your school?  Do you know which schools in your province experience the highest learner migration rate?  Can you explain why you decided to let your child leave the township or neighborhood secondary school to enroll him/her at the present school?

Population and Sampling
The methods of sampling which the researchers employed was purposive sampling. This method is most often the most appropriate method for selecting participants from the population if in-depth knowledge of participants is required (Dreyer, 2018). In the current study a sample was drawn from secondary schools that have experienced migration of learners: the principals, the learners who migrated to such schools and the parents of these learners. The researchers also approached the Department of Education (DoE) to gain assistance from EMIS-officers who enabled the researchers to identify information-rich schools in each province. In some provinces a list was sent to the researchers while in others, names of the schools were suggested. The researchers contacted the EMIS-officers from which list were obtained and requested them to suggest specific schools. The researchers then purposefully selected a school from each province based on logistical reasons. The total sample of 50 (n=50) were made up as follows:  One secondary school principal per province to which large numbers of learners have migrated from each of the nine provinces of South Africa (n=9).  One EMIS-officer from each of the nine provinces (n=9).  Four parents and four learners from schools in four provinces. These participants were selected with the help of the secretaries who provided the names of the parents and their respective children who, according to them, would be most suitable in terms of providing information (n=4x8=32). Codes were allocated to each participant to ensure anonymity of the participants. Table 1 below gives a summary of the sample and the respective codes.

Data Collection Methods
The methods deployed in this study were a literature study, semi-structured interviews, focus group discussions and observations. Each method is discussed below. The researchers have chosen one secondary school for the pilot study in one selected province. The purpose of the pilot study was to conduct a feasibility study that would help the researchers to execute the process of research. It enabled them to refine the research problem, test the viability of research and familiarize themselves with the field of the study. After permission was granted, the individual interviews were conducted in one secondary school in each province in which learner migration takes place. Permission was gained to interview the principal and the EMIS-officer in each of the nine provinces as well as parents and learners in four selected provinces.

Literature Study
According to Mouton (2003) a literature study is aimed at reviewing a body of knowledge concerning a phenomenon. The literature study revealed an appropriate theoretical framework for this study and informed the questions asked by interviewers. It also provided a valuable frame of reference with which information obtained during interviews could be related.

Interviews
The most common method of data collection in qualitative research is interviewing. The research interview helps researchers understand the closed life-world of the interviewee (Strauss and Myburg, 2000). The main purpose of interviewing is to find out things that cannot be observed with the naked eye (Maharaj, 2005). The researchers believe that the interview was a suitable tool to collect information for this current study. The principals, parents and the learner"s voices could be heard explaining their reasons for migrating and articulating their views of their experiences of learner migration. In the same vein the researchers are aware that interviews are time consuming and that sometimes the interviewee may not be willing to open up. Semi-structured interviews were used. Three different kinds of questions are relevant in terms of semi-structured interviews, namely:  Main questions: The researcher prepares a handful of main questions which guide the conversation.  Probes: When responses lack sufficient detail or depth of clarity, the interviewer asks a probing questions to complete or clarify to an answer, or to obtain explanatory examples and evidence.  Follow-up questions: These questions pursue the implications of answers to the main questions. The researchers selected this method for this study because predetermined questions were deemed necessary to ascertain that relevant information is obtained and that probing questions and follow-up questions are asked to engage participants in giving reasons for migration from their former schools to their present school. Two types of semi-structured interviews were employed for this study, namely semi-structured focus group interviews and indepth semi-structured individual interviews. Semi-structured focus group interviews were conducted with parents and learners in selected provinces, while individual semi-structured interviews were conducted with selected principals of schools because they were be able to clarify issues of learner migrations in detail as they deal with admissions of these learners on yearly basis. Telephonic interviews were conducted in some of the provinces due to logistical reasons.

Observations
Observation was used to complement the face to face interviews in in Gauteng, Limpopo, Mpumalanga and North West Province. Observation used during this study would not qualify as observation in the true sense during which the observer either "focused on situations in which the observer has no control over the behaviour in question, and plays an unobserved, passive and non-intrusive role in the research situation" (Lee, 2000) or participant observation where the observer becomes one of the group (De Vos et al., 2005).
In this study, the researchers were able personally to visit and observe participants during interviews in four provinces, namely; Gauteng, Limpopo, Mpumalanga and Northwest. In the process of collecting data using observation, the researchers fully adhered to the recommendation made by De Vos et al. (2005) as well as Leedy and Ormord (2005) in terms of traditional observations in research, namely that observations should be documented carefully. Puttergill (2013) describes qualitative analysis as the meaning attributed to a social phenomenon, while Johnson and Christensen (2004) point out that in qualitative research, the researcher use data analysis to develop a deeper understanding of their research topic. In this study the raw data which consisted of transcripts of interview and field notes were arranged to reveal manageable themes, patterns, trends and relationships (Mouton, 2003) which were transcribed to provide a complete record of discussions. The researchers analyzed the content of the discussion using Tech"s method as outlined by Babbie and Mouton (2004), Cohen and Morris (2011) as well as (Cresswell, 2014), namely:

Data Presentation and Analysis
 The researchers reads through all transcripts and jots down ideas as they come to mind.  The best interviews are selected and the meaning of each piece of information noted in the margin. This step was not applied since all the interviews were required so that each province could find representation.  A list of all topics are made and similar topics are clustered together.  Topics are abbreviated by means of codes and these codes written next to each segment of data in the transcribed interview.  Categories were formed by grouping topics together and determining the relationships between categories. The qualitative research approach encourages inductive reasoning when it comes to data contact analysis procedure (McMillan and Schumacher, 2014). In this study, inductive reason was applied to isolate themes that have direct bearing on the migration of learners as regard to EMIS-officers, principals, parents and learners. Inductive reasoning create an environment that individuals can be interviewed and studied so that generalization can be formed. A total sample of 50 participants were interviewed who freely expressed themselves in connection with learner migration. The content analysis allows the findings to surface from data. Data interpretation affords the researcher an opportunity to step back and reflect on the topic being researched. These formed the findings in terms of learner migration in South Africa.

Ethical Considerations
Since research ethics is one of the important aspects in the research process because it provides guidelines that guard against harmful effects of research (Dreyer, 2018). The researchers undertook measures to obtain permission from the relevant Departments of Education and from participants. This ensured that the participants" confidentiality and anonymity were protected during their participation in the study, while in all cases written and verbal consent was obtained. The researchers respected the lives of the participants by honoring their privacy, protecting their identity, treating them with respect and by not violating their human rights are protected at all costs as enshrined in the constitution of South Africa (Republic of South Africa RSA, 1994a).
For the purpose of this study, the following ethical issues related to human rights were relevant and were explained to interviewees from whom consent was requested (Babbie and Mouton, 2004;De Vos et al., 2005;Du Plooy, 2000):  Informed consent: The participants were in advance given adequate information on the goal of the investigation and the procedures which would be followed during the study. Informed consent ensures accurate and complete information. In this study, consent was obtained from the principals, parents, learners and EMIS-officials after the provincial Departments authorized the empirical research.  No harm to participants: Since participants should never be harmed or stressed during any research, irrespective of whether they agree to partake in the research or not, the researchers guarded against questions that could have psychological implications, such as, personal circumstances that are related to the affordability of paying high fees in independent schools.  Confidentiality and anonymity: The researcher assured the participants that their right to anonymity would be respected. No names were used. Different letters of the alphabet were used to identify the views held by EMIS-officers, principals, parents and learners.

Interpretation of the Research Findings
A key purpose of this study was to establish the reasons which led to learner migration in South Africa and to relate these to the findings to Bronfenbrenner"s bio-ecological model. This allows for a holistic view and an in-depth understanding of the etiology of learner migration in order to provide recommendations to address potential problems that should be prevented from the phenomenon of migrations. Attention is also paid to the outcomes resulting from learner migration and the challenges associated with it. Discussions of findings from interviews of EMIS-officers, principals, parents and learners is provided and supported by statements from interviews. Pseudonyms were used for all respondents.
The data is presented by discussing overall themes which emerged from an analysis of interviews with EMISofficers, principals of schools, parents and learners. The overall themes are followed by the reasons for learner migration and challenges relating to learner migration. As mentioned, the data collected mainly centered around reasons for learner migration as it is closely related to the main research question, namely: How can the problem of learner migration be addressed and managed?
The analysis of data has been integrated into Bronfenbrenner"s bio-ecological theory so that a holistic perspective on the reasons for learner migration can be presented in terms of the micro-, meso-, exo-and macrosystem to address the issue of migration in a comprehensive manner. This holistic base can serve as a frame of reference for further studies, either qualitatively or quantitatively.
Practices and experiences related to learner migration were shared by interviewees from the nine provinces of South Africa. Two major themes emerged from the interviews and are contextualized in more detail in the next section. The two major themes are the complexity of learner migration and the highly emotive nature of the issue of learner migration.

The Complexity of the Phenomenon of Learner Migration
The complexity of the phenomenon of learner migration becomes apparent if one considers the extensive scope that it entails. Not only does learner migration relate to various reasons that initiate it, it also includes various challenges experienced by principals and EMIS-officers, parents and learners.
The complexity of the phenomenon of learner migration covers different spheres. It ranges from principals" managerial tasks, parental interaction with schools, positive and negative experiences of EMIS-officers and learners, as well as the problematic administrative role of principals when planning for the forthcoming year. In terms of the latter, learner migration becomes extremely complex since schools receive large numbers of application forms from parents whose children migrate from feeder schools ranging from schools situated in close proximity to schools far away beyond the borders of feeder areas. Principals are challenged to come up with strategies to handle the mammoth task of learner migration

The Highly Emotive Nature of Learner of the Phenomenon of Learner Migration
Interviews evoked strong emotions of fear, anger, anxiety, insecurity and joy. Principals seemed particularly appreciative of the opportunity that the interviews gave them to demonstrate positive emotions concerning the effective management systems of their schools, which made their schools known as "better" performing schools which drew learner migrators from various provinces of South Africa . The principals" pleasure and satisfaction was expressed by facial expressions and by tone of voice when they talked about their schools" reputation for good performance. Learners also displayed positive emotions of satisfaction when referring to their choice of school and subjects selected from the curricular menu in terms of envisaged careers.
Mixed emotions were displayed during interviews. Some parents, for example, showed satisfaction pertaining to issues such as regular meetings held at schools to which their children had migrated, which afforded them the opportunity to meet teachers and share ideas concerning their children"s formal education. Others expressed their frustration due to their inability to assist their children with homework since they have migrated to their new schools.

Discussion and Analysis of the Findings
The migration of learners from one school to the other is the result of so-called "pushing" -and "pulling" factors. Migration theorists such as Dovlo (2003) and El-Khawas (2004) advance the "push-and-pull-factor" theory to explain the migrations. Pushing factors refer to hostile conditions that compel learners to leave their schools and seek admission in other schools, while pulling factors are the attractive conditions in receiving schools that draw learners. The following reasons emerged from the study as main reasons why learner migration, involving both pushing and pulling factors, are taking place on such a large scale in South African schools:

Changes in Legislative Frameworks and Policies
The South African Education system reform policies should be viewed in the context of international emerging trends with regard to education policies. Section 5 of the South African Schools Act (SASA) of 1996 (Republic of South Africa RSA, 1996c) states clearly that all public schools are required by law to admit learners and serve their educational requirements without unfairly discrimination. Although there exists such a policy, the schools use a variety of criteria to avoid admission in their schools. Problems associated with admission are, inter alia, that Christian private schools only allows learners who are from Christian families, while some public schools in turn only admit learners who are in the same suburb or area as the school. This problem is even worse when taking into account the reality that most parents from rural areas violate rules by looking for a guardian for their children who stays in the vicinity. In the same breath one can view these criteria as a silent restriction by the schools.
One of the principals (KN-Pr1) interviewed stated: The policy for admissions is the problem. Parents take their children and send them to stay with families where they think are better schools, so the number of our learners decrease all the time, while the EMIS-officer of another province (NW-E1) complained: Although policies exits to prevent this, some schools do not want to admit learners from their feeding area, using excuses such as religion and language to not admit learners.

Effective School Management and Leadership
The importance of effective school management in this process cannot be emphasized enough. It is assumed that the School Governing Body (SGB) and School Management Team (SMT) are recognized structures which should assist the principal in running the school. It is through these joint efforts that principals guide the teachers and learners in establishing a culture of teaching and learning (Engelbrecht, 2006). Poor school management is manifested in lack of discipline and poor learner performance. In the same breath good school management is the result of implementation of good disciplinary practice and effective leadership of the principal. This can be regard as a pulling factor as learners migrate to schools with effective school management systems with good discipline and improved results. One of the parents (L-P2) summarized this when stated: I took my child from this one school because the principal cannot run the school; there are no teaching, the teachers sit around and the principal fails to manage them.
Effective leadership from principals can be related to their approachability and availability that benefit both staff and learners. Effective leadership from principals is manifested in effective schools. An effective leader also provide opportunities for staff development that enables teachers to perform beyond their call of duty. This can be regard as a pulling factor as learners migrate to schools with effective school leaders, again resulting in good discipline and improved results. Another parent (G-P3) said in this regard: The majority of Grade 12 learners failed last year, so there is no culture of learning in the school due to poor leadership, this is why I took my child from this school. A learner form the same province (G-L1) added: I want to be in a school with good results, my future depends on this; this is why I decided to leave this school. The principals cannot lead and we feel and see it every day.

The Implementation of Good Disciplinary Practices
Good disciplinary practices contribute to learner migration. In this study it was indicated that a school with a well-developed disciplinary policy, endorsed and implemented by the principal and staff who discipline learners, is regarded as a good and effective school. This factor is considered as the moral compass of the school and can therefore be regarded as a pulling factor as it prompts learner migration. The results of the findings show that in all the provinces, learner migration is indeed influenced by the lack of proper discipline and a deterioration of moral values. This aspect is one of the major pushing factors with regard to learner migration.
One of the principals (WC-Pr1) of an effective school in a province with the best matriculation results in the country stated clearly: Discipline is the beginning and the end at my school. Without discipline, you cannot achieve good results, and that is what we aim for. The EMIS-officer (L-E1) from another province with one of the poorest matriculation results in the country confirmed this when he agreed with the principal: The biggest problem in schools in my province is a lack of discipline; not only from learners but also from teachers, it all start with the principal. If we do not improve this, our results will not improve; we lose learners to other provinces due to this. One learner (L-L2) from the same province said unconditionally: I have left my old school because there was no rules and regulations; everyone do what they want to and the principal does not even care.

Improved Formal Education
This aspect goes hand in hand with effective school leadership and management and they are indeed two sides of the same coin. The study revealed that former Model C schools as well as independent and private schools offers a good curricula that allows learners many career opportunities. Furthermore, the use of English as medium of instructions is considered as another contributingor pulling factor for learner migration as learners will be conversant with the Queen"s language. Over and above this pulling factor, good Grade 12 results constitute the bedrock on which the road to tertiary institution is constructed.
From the empirical investigation it became evident that good matric results play a prominent role in learner migration and act as an important pulling factor with regard to learner migration. One principal (L-Pr1) said: If you have good teachers, like we do, learners will not leave the school as they know good teaching and good education are they keys to their future, while one of the parents (LP-3) from the same province reiterate: Our children prefer to be teach in English; I will therefore send my child to a school where he is taught in English. The EMIS-officer (L-E1) from the same province responded on this issue as follows: We lose learners to other provinces; they prefer to attend a school where they know there are excellent teaching and learning, and who can blame them for that? One of the learners (L-P3) from the same province concluded: I will go to as school where I know I will get good education, even if I have to travel very far to get there and if it cost my parents more; they know just as good as I do how important a good school is for me.

Structural Educational Changes
From this study it is evident that learner migration is a serious challenge to some schools. Learner migration and the consequent decrease in learner numbers result in the merging of schools which is a common practice in some provinces. One of the principals interviewed (NW-P2) stated unambiguously: Our learner numbers had decreased over the years a lot; this resulted that we had to merge with the school next door. Now we lose even more because of this issue. One of the learners (M-L3) stated: When our school merged with another school, everything got worse. I was not prepared for this and decided to leave the school. Luckily my parents support me all the way. It became clear from the study that the factor of structural educational changes acts as a pushing factor with regard to learner migration as the merging of schools result in learners leaving the school to attend another school.

School Infrastructure
The infrastructure of most schools in rural areas and in townships range from poor to non-existent. Schools, specifically in rural communities, are dilapidated and lack basic services such as sanitation. It is therefore no surprise that a shortage of infrastructure serves as a pushing factor by encouraging learners to migrate to schools that have better infrastructure and sanitation.
One of the principals interviewed (G-Pr1) emphasized: learners come to my school just because we have toilets, labs and a nice looking garden. They are not prepared anymore to stay at a school without a proper infrastructure such as well-maintained buildings, while the EMIS-officer (G-E1) of the same province concluded: We are lucky in our province, the infra-structure of our schools are still generally OK and we need to keep it as such. Learners migrate from other nearby provinces to our schools because they know we have better schools, and we appreciate that. One of the learners (NW-L2) shared this sentiment: My former school did not even has decent toilets and bathrooms. There are no trees, so we sit in the sun the whole day as some classrooms do not have enough space for all of us. In some classes we are more than 60 kids, we can't even hear what the teacher try to say. Another learner (L-L3) complained about them lack of labs: I like sciences and biology, but we don't have any labs. There are no equipment and the teacher is not able to demonstrate anything to us. I wish I can leave this school, but I can't.

Socio-Economic Factors
It would be very difficult to discuss learner migration in South Africa without contextualizing it in the previous political dispensation. Due to racial segregation mainly blacks were confined to rural areas. This scenario has, however, changed rapidly in recent year. This can be seen in changes to urbanization, status, prestige and improved public transport. Urbanization is experienced where parents and their children are moving to urban areas. Learners from empowered families are able to utilize public transport and get admission in urban schools while these from poor families remain in rural schools. The distance from home to work is also decreased. Prestige and status also plays an important role in learner migration. These are all pushing factors with regard to learner migration.
One of the parents (M-P4) addressed the concern of the majority of parents in his province as follows: I am not prepared that my child attends a `black-only' school; this reminds me of the Apartheid days. I am lucky, because I can afford to put my child in a hostel in a former Model C school. I know not all parents can afford this, but this is non-negotiable to us. The son of this parent (M-L3) agrees with his father: I am in a hostel, but a lot of my friend moved to the city; this means they can also now attend this school, and it is so different from my old school. We are taught in English and I feel part of the new generation of learners.

A Safe School Environment
This pulling factor goes hand in hand with socio-economic factors. The interviews established the view that a safe school environment that is conducive for teaching and learning and good accommodation for learners translate into a pulling factor for learner migration. Principals are taking precautionary measures to ensure the safety of learners at all times when crime, including rape on school premises, has escalated to unacceptable high levels.
One of the parents (M-P2) said: Safety is no the number one priority for me; even more important than good education. But these two go together anyway. If I know my daughter is safe, I sleep well at nights. She did not feel safe at her former schools. I am so glad we moved her here. One EMIS-officer (EC-E1) said in this regard: One of our biggest challenges in this province is to make our schools safe again. Problems range from drugs, weapons at school, alcohol and gangs operating in schools. We are losing learners daily as they prefer to go to schools in areas with less crime. A learner (NW-L1) concluded: For the first time in my life I feel safe at school now I can focus on what I am here for: to get the education that will enable me to attend university.

Poor Learner Performance
The interviews with parents and learners from all the participating provinces revealed that learners" poor performance constitutes a driving force of learner migration. Learners migrate from schools without a culture of teaching and learning.
One of the principals (FS-Pr1) stated in this regard: It happens more and more that parents take their children out of my school because their results are not good enough, even when it is sometimes their own fault. In this process, we lose a lot of learners that negatively influence the school. It is, however, not always the fault of the school when a learner do not progress well. One learner (L-L1) who migrated to another school after he failed one grade said: The problem was not with me. The teacher did not do anything in class. Since I moved to the new school, my marks increased a lot. I wish all my friends could join me in my new school. A parent agreed: I wish I had change school for my child a few years ago. From the time that he entered the new school, everything changed. She is now happy and her marks show it; it had even doubled in some subjects. I am so glad for her sake.

Recommendations from the Study
From the findings above, it seems that are very relevant and legitimate pushing and pulling factors in the South African that play a role in learner migration. With these in mind, the following recommendations can be made that derived from this study:  The SMTs of schools where learner migration is experienced in large numbers should be reskilled to enable them to address this problem.  It is challenging to operate an institution on a tight budget. The Department of Basic Education (DoBE) should be hands on in the improvement of teaching and learning practices as well as infrastructure and resources of schools in this situation.  The Apartheid resources backlog of poor infrastructure in many schools in various provinces should be improved.  Discipline is a serious challenge in schools where migration is experienced. It is recommended that a policy be developed in line with the departmental policy that will promote discipline in all schools. The policy should be communicated to different stakeholders, while the SMT and SGB should implement, manage and monitor the policy.