Supervision experiences of postgraduate students at an ODL institution in South Africa

Abstract This article expounds on findings of a qualitative case study was conducted in a South African Open Distance Learning university. The aim of the study was to explore the experiences of postgraduate students Open and Distance Learning context. Data was collected through semi-structured interviews and narratives written by the interviewees. Typical case sampling was used to select six doctoral graduates. Data analysis was conducted thematically, and predetermined themes from the literature and those emerging from the data were analysed and interpreted collectively. The findings revealed that many students felt supported by their supervisors, while some strongly felt unsupported and neglected. Moreover, communication is mainly through emails, telephone, WhatsApp, and occasionally through personal meetings. Whilst new technologies enhance support for such students, care needs to be taken that other tried and tested methods of support are not assumed to be anachronistic. Therefore, supervisors need to develop rigorous strategies, for example, Communities of Practice (COPs) to involve postgraduate students in the research community of which they wish to become a part. Moreover, the students are older and in remote rural areas where there is a problem of connectivity and they struggle to familiarise themselves with e-learning.


PUBLIC INTEREST STATEMENT
This research delves into the learning experiences of postgraduate students during the supervision of their research projects. It addresses the need to inquire into the practicalities of postgraduate supervision in the open distance learning (ODL) pedagogical setting. This study examines the supervision experiences of doctoral candidates who graduated from a reputable longstanding ODL institution. The study was necessary because the spatial and temporal implications of ODL pose salient professional and personal challenges regarding supervisor-supervisee interactions that influence the content, progress and delivery of research activities. The study becomes even more revelatory given that higher education institutions are eager to hasten student completion within the minimum period stipulated in institutional policies to enhance their performance. This research demonstrates that it is imperative for students to experience supervision as being developmentally beneficial for them as emerging academics by facilitating the acquisition of scholarly independence, dexterity and self-confidence through constructive criticism.

Introduction
Postgraduate research and supervisor practices are both relatively solitary activities. Erichsen et al. (2014) postulate that postgraduate study is a challenge to many students. The challenge becomes even harder when students in postgraduate programmes complete their degrees through distance provisioning. Moreover, distance education comes with its own challenges due to the physical distance between the supervisors and the students (Moore, 1997;. Distance education then emerges as an educational trend that gives everyone the chance to learn asynchronously and uninhibited by time and space. Among some of its features, distance learning provides a flexible and multimedia-rich educational environment for students (Isik et al., 2010). Additionally, universities, whether they appreciate it or declare it, are becoming more "open" as manifested by a number of contact universities having distance education programmes concurrently with their facet-to-face programme offerings.
The contexts and means through which postgraduate supervision is practised in contemporary universities reflects not only the different and diverse needs of part-time students-often off campus -but also the emerging computer and communications technologies (Evans, 1995;Henry et al., 2001). As mentioned earlier, all universities, whether they appreciate it or declare it, are becoming more "open" universities. Given this context, postgraduate research supervision is then worth considering.
Literature reveals that until relatively recently, there has been little research on the challenges and complexities of distance learning postgraduate research degree programmes as experienced by the students themselves (Nasiri & Mafakheri, 2015). Also, as postgraduate programmes continue to integrate e-learning, hybrid and online components into the curriculum, these complexities and challenges have also become increasingly important to the scholarship on doctoral studies and supervision (Ali et al., 2016). Within the broader movement to examine and reimagine the processes of developing future scholars, facilitating doctoral degrees, and rethinking the role of supervisors in this process, our purpose was to ascertain empirical data on postgraduate students' experiences with their masters and doctoral supervision within the context of Open and Distance Learning.
This article is based on the findings of the study that was conducted in an Open and Distance Learning context in South Africa. The purpose of the study was to explore the experiences of postgraduate students in such learning context. Moreover, there is a shortage of research focusing on pedagogical process and learning experiences of post graduate students' supervision. There is a need to delve deeper into the practical aspects of distance postgraduate supervision.
The exploratory qualitative study on which this article is based seeks to contribute to better understanding of the supervision experiences of postgraduate students in doctoral research supervision (Chamberlain, 2016). The central argument in this article is whether distance supervision presents different experiences in the overall learning experience, or is simply a technology-driven choice. According to Nasiri and Mafakheri (2015), Evans (1995) has registered the opportunities that a distance programme, in the light of new technologies, could provide for good quality postgraduate research supervision.

Students' supervisory needs
Postgraduate students have supervisory needs, which is the form of academic support given to students by supervisors. Those needs include mentoring, nurturing and guiding. Wiegman (2013) argues that by its nature, supervision is in fact, mentoring. The latter author) bases her understanding of the ontological nature of supervision on the fact that a mentor is more than a guide throughout the postgraduate students' research endeavour. She (Wiegman, 2013: online) emphasises the issue thus: They [supervisors] provide both professional and personal advice in transitioning into, and out of, graduate school. They give constructive feedback on writing, teaching, and other elements of career design. They can also serve to help students balance professional goals with their personal lives or give emotional encouragement during challenging times.
The definition of supervision cited above brings to light the all-embracing nature of supervision, which equates supervision with mentoring. More than just providing guidance on research, supervision also covers the provision of assistance with the management of personal life and career life, which directly affect postgraduate students' research studies (Ali et al., 2016;Chamberlain, 2016). Zelditch (1990) further distinguishes the categories of mentoring/supervision as: advisers-people with career experience willing to share their knowledge; supporters-people who give emotional and moral encouragement; tutors-people who give specific feedback on one's performance; masters-in the sense of employers to whom one is apprenticed; sponsors-sources of information about, and aid in obtaining opportunities; models of identity-of the kind of academic to be emulated. Thus, supervisor support is one that is expected to be encapsulating all of these aspects.
Postgraduate mentoring relationships are close, individualised relationships that develop over time between a graduate student and one or more faculty members (Wiegman, 2013). The attributes of mentors include knowledge and skills as a researcher, ethical research behaviours, and personal characteristics and traits that enhance effective personal and intellectual or scholarly development (Borders et al., 2012). Accordingly, research mentors are expected to share their respective areas of research expertise with their mentees. Supervisors, however, do not necessarily have to be knowledgeable about all aspects of research (Fourie-Malherbe et al., 2016;Vyncke, 2012). Notwithstanding, it is crucial for them to be aware of their limitations as researchers and inform the mentees of such limitations, encouraging them to find other resources as needed (Borders et al., 2012).
It is significant to point out that supervisor feedback is very important, as it plays a vital role in supporting postgraduate students' development during the academic research journey (Saleem & Mehmood, 2018). Wang and Li (2011, p. 101) aptly encapsulates this feedback factor, stating that "Students benefit from engaging in intellectual exchanges with their supervisors in order to receive guidance on their research progress and thesis writing". Stracke and Kumar (2014) also corroborate that it is by means of the feedback process that the supervisor guides the student through the research journey towards becoming an independent researcher and a competent critical scholar.
While feedback positively contributes to the academic development of students, it may in some situations be counterproductive and damaging to students' self-esteem, consequently affecting their learning outcomes negatively. This was evident in previous studies by Caffarella and Barnett (2000) and Hyland and Hyland (2001), which revealed that receiving critical feedback can be emotionally difficult for some students. Therefore, in order to keep the supervisory relationships healthy, feedback must not only focus on the "what," but also on the "how" of the way feedback is given and received (Holmes et al., 2010). The supervisor should therefore use the motoring model in which the supervisor and student engage in open dialogic exchanges of insights and ideas, which leads to acceptance in a critical and reflective way (Wang & Li, 2011).

Peer support
Peer support is provided primarily by means of students' involvement in peer groups, which are groups of postgraduate students who meet regularly without their supervisors to discuss the content and process of their doctoral studies (Stracke & Kumar, 2014). According to (Stracke, 2010, p. 2), "the conventional perception of a peer group is often one of a group in which postgraduate students meet without their supervisor". Interestingly, Stracke (2010) suggests broadening the definition of the term "peer group" to mean a group of peers that includes supervisee(s) and supervisor(s). In this inclusive case, Stracke emphasises the importance of a desired partner-like relationship between supervisee and supervisor that is of the peer-to-peer notion, as underpinned in the idea of a community of practice (Lave & Wenger, 1991).
Traditional views of the writing process as a solitary and painstaking task could inhibit postgraduate students from pursuing useful conversations about their writing (Vyncke, 2012). Moreover, for many postgraduate students, the challenge of writing their theses is undertaken without much guidance. Most of the supervisors provide guidance mainly on content and advice on research process (Fourie-Malherbe et al., 2016). They seldom create a conducive atmosphere in which students could engage with each other on their research experiences. Recent research has suggested that spaces for opening discussion on writing are needed and are important in supporting postgraduate writers to develop their academic identity (Fergie et al., 2011).
In their paper reflecting on their own experiences as doctoral students, Devenish et al. (2009) further noted that recognition of the important role that might be played by fellow students in the supervision process was missing from annual progress reports of their university. It is important for postgraduate students to collaborate. The socialisation aspect embedded in such collaboration contributes to enhanced completion rates (Littlefield et al., 2015). The latter view is also confirmed by Gardner (2010) in a study conducted on the impact of socialisation and completion rates in doctoral programmes. The study found that successful programmes offered the greatest opportunity for students to form such social relationships as they felt that they were in a family.
More advantages of peer support have long been echoed by researchers more than a decade ago (Conrad, 2006;Ferguson, 2009;Gardner, 2010). These researchers mentioned the gains of peer support, such as writing support, development of information literacy, the likelihood of completing a degree, a place where students are accountable to the group for their progress, and a place where students provide and receive support for publication.
However, collaboration may be negative for other doctoral students. For example, Pemberton and Akkary (2010) assert that in a peer group setting, some students could be overwhelmed by the fear that their individual student voices would be diminished by the group voice. Furthermore, the potential for positive group dynamics could suffer as a result of group members group not contributing or putting in the same effort as others in the group. Competition may even arise from academic authorship crediting. Nonetheless, collaboration and group work could inadvertently foster negative implications for the individual student. In such instances, the benefits then outweigh the drawbacks (Littlefield et al., 2015).
In addition to normal peer group, Doctoral Community Network (DCN) has been acknowledged by Radda as one of the ways in which postgraduate students at a distance learning institution could have peer support. It is this DCN aspect that provides non-traditional doctoral students the opportunities to collaborate with their peers virtually (Guri-Rosenblit, 1999).
In a study conducted by Andrew to investigate the challenges of post-traditional, distance PhD supervision, the postgraduate students expressed that their main challenge was premised on lack of access to a community of scholars. Thus, it is useful to also consider a notion of Communities of Practice (CPO) as advocated by Paul et al. (2000). COP is defined as " . . . a group of professionals informally bound to one another through exposure to a common class of problems, common pursuit of solutions, and thereby themselves embodying a store of knowledge" (Manville & Foote, 1996, p. 67). In the context of distance supervision, Guardian supervisors, cohorts of PhD students and latterly the team of distance supervisors can be identified as communities of practice which develop and share ideas and build good practice over time. It also important to note that for the ODL institutions that are in contexts where ICT implementation and connectivity is not an issue, it would be easier to translate some of the co-situated practices to the virtual world(relatively easily, for example, finding a common purpose or at least a shared interest (Paul et al., 2000). This position, therefore, cohere with the argument propounded by Wood and Louw (2018) in their proposition for the preparation of postgraduates for participation in the research fraternity (community of practice) rather than as apprentices being readied for artisanship.

Fiscal support
Fiscal support is important in ODL institutions due to the physical distance between the supervisor and the supervisee. Supervisors of international students at distance learning institutions are possibly the most isolated. For this reason, postgraduate students could face specific challenges related to negotiating issues and practices concerned with demands of studying and undertaking research at such institutions, balancing research, work and domestic responsibilities (McCallin & Nayar, 2012, p. 63). Thus, supervising students at a distance learning institution needs extra support as the role of supervisors is potentially problematic due to cultural misunderstanding and limited personal contact (Brooks, 2010;McCallin & Nayar, 2012).
Technology-based approaches are advocated by researchers as a solution to the inherent challenges caused by the physical distance between the supervisor and supervisee. Research) confirms that increase in the use of technology offers more opportunities for students to take part in education and supervision remotely (Carlisle et al., 2013).
However, researchers have registered the concerns regarding the downside of new technologybased approaches in distance learning supervision (Carlisle et al., 2013). Regarding this concern, Brooks (2010, p. 261) states that "many institutions are still struggling to provide appropriate and effective training, development, and reward opportunities for faculty". Moreover, technology-based approaches add on the work load of supervisors as it demands a certain way of administering the research process. For example, the supervisors have to know how to deal with Web-based enrollment documents, grading procedures, and other online classroom tools.
As a result of these technological innovations, supervisors find themselves needing timely assistance when faced with technology-related problems. Brooks (2010) argues that supervisors lack skills and knowledge of using technology-based educational practices because universities are "still struggling to provide appropriate and effective training, development, and reward opportunities" for faculties (Brooks, 2010, p. 262). Also, some of the disadvantages related to distance e-learning supervision as listed in the literature (Chapman, 2006;Dubi et al., 2012;Sindlinger, 2011) have been as cited in Carlisle et al. (2013, pp. 22-23) Thus, technology can be expensive and distance supervision may not be an economically feasible option for some supervisors and supervisees. Additionally, distance supervision requires specialized knowledge of various asynchronous and synchronous methods used to deliver supervision. Also, failures in technology can occur (Watson, 2003) which can disrupt the process or flow of sessions. Finally, the lack of in-person contact between the supervisor and supervisee can be a potential drawback (Vaccaro & Lambie, 2007). As described by Stenbnicki and Glover, some challenges specific to asynchronous supervision methods (e.g., e-mail) include misunderstandings between communicators as a result of the absence of nonverbal cues as well as the risk of communicators leaving out important or contextual information due to the time required to type a communication.
It is therefore crucial for universities to consider some of the disadvantages and challenges of distance e-learning supervision whenever they decide to use it.
There are essentially two types of delivery methods for distance learning (DL) supervision (Guri-Rosenblit, 1999;. Synchronous software programmes, such as the telephone or video conferencing technology which involve live, instant, and real-time interactions; while asynchronous software programmes such as e-mail, threaded discussions, and file sharing programmes involve a delay in time between communications (Clingerman & Bernard, 2004;Chapman, 2006;Sindlinger, 2011). Both synchronous and asynchronous methods of supervision offer both face-to-face and online supervisors a technological assortment of delivery methods to conduct and supplement other supervision approaches. Nonetheless, the contexts through which postgraduate supervision in ODL is being practised calls for the use of computer and communications technologies (Evans, 1995).

Moore's theory of transactional distance
In his Theory of Transactional Distance (TTD), Michael G. Moore posits that in distance learning scenarios, separation between the teacher and students can "lead to communication gaps, a psychological space of potential misunderstandings between the behaviours [sic] of instructors and those of the learners" (Moore & Kearsley, 1996, p. 200). Giossos et al. (2009) further refine this notion in their review of the contemporary relevance of Moore's theory. They stated that: " . . . the particularities of space and time pertaining to teacher and learner which characterise distance learning, create particular behavioural models for the teacher and the learner, psychological and communication distance between them, and insufficient understanding of each other" (Giossos et al., 2009, p. 2).
According to Moore (1997), the nature of the transaction developed between teachers and students in distance learning needs to take into account three factors: dialogue, structure, and learner autonomy.

Dialogue
Dialogue refers to more than simply two-way communication, but takes into account all forms of interaction, "within the context of clearly defined educational targets, cooperation and understanding on the part of the teacher, and, ultimately, it culminates in solving the learners' problems" (Giossos et al., 2009, p. 2). Moore (1997) indicates that the important consideration in this respect relates not to the frequency of dialogue, but to its quality and the extent to which it is effective in enabling the resolution of learning problems the distance learner may be experiencing.

Structure
The second factor Moore (1997) refers to is the nature of the course structure, which is described as the level of the course's rigidity or flexibility. This factor includes aspects such as the extent to which course goals and objectives are prescribed, the pedagogical model used in teaching the course (for example, teacher-versus student-centred learning), the nature of course assessment, and the ability of the course to accommodate individual student needs (Zhang, 2003).

Learner autonomy
The third factor, learner autonomy, is reliant upon the previous two, in that it refers to the sense of both independence and interdependence perceived by learners as they engage in the course. Learner autonomy is intimately tied in with a learner's sense of self-direction or selfdetermination, and this can be significantly affected by the dialogue, the level of rigidity or flexibility inherent in the course design and delivery, and "the extent to which the learner exerts control over learning procedures" (Giossos et al., 2009, p. 2).

Usefulness of Moore's theory in an ODL context
Moore's theory asserts that an inverse relationship exists between the three above-cited factors, in that increases in one could lead to corresponding decreases in others (McIsaac & Charlotte, 1996).
For example, a course with an inflexible structure could lead to a decrease in the quality of dialogue and sense of learner autonomy, thereby increasing the students' perception of transactional distance. Although Moore (1997) does not specify the threshold, he also notes that when course structure decreases below a particular threshold the sense of transactional distance learning could actually increase, due principally to the potential for learner confusion or dissatisfaction.
A number of studies have been conducted to determine the empirical status of Moore's theory. For example, Bischoff (1993), Bischoff et al. (1996), Chen (2001aChen ( , 2001b, and Saba and Shearer (1994). Although not unanimously accepted (Gorsky & Caspi, 2005), these studies generally confirm the usefulness of Moore's theory as a framework against which to analyse distance education practice. As Garrison (2000, p. 3) intimates, theories such as transactional distance "are invaluable in guiding the complex practice of a rational process such as teaching and learning at a distance"; while Jung (2001, p. 257) avers that the theory "provides a useful conceptual framework for defining and understanding distance education in general".
In this article

Research approach
This was a qualitative case study conducted in an Open and Distance Learning University (ODL) in South Africa. A deliberate effort was made to interview postgraduate students who completed their doctoral studies in an ODL university. Such an approach was adopted as the aim was to explore the dynamics of postgraduate supervision in an ODL environment (Rossman & Rallis, 2012).

Sampling
In this qualitative study, purposive sampling was used, based on our exposure to, familiarity and engagement with postgraduate supervision in an ODL context (Rossman & Rallis, 2012). The criterion referenced sampling technique was opted for, and in terms of which the selected six participants were postgraduate students who already completed their PhD studies at the university of investigation (Suri, 2011). In addition, these participants were employees of the self-same university of investigation.

Data collection
Individual in-depth interviews were conducted to collect data from each of the selected six participants. The interview schedule was developed and used to complement the data collection process (Creswell, 2012). The interviews also encompassed written narratives by participants. As a mode of enquiry, the narrative method examines "experiences" as expressed in lived and told stories of individuals (Liamputtong, 2013). McAlpine (2016) avers further that narrative enquiry creates a construction of the self, and the researcher acts as an active agent by demonstrating his or her goals and intentions. Therefore, the researcher as an active agent makes connections between events, shows the influence of the passage of time, and recounts the personal meaning of the experience.

Data analysis
The interviews were recorded on audio tape and later transcribed verbatim. Additionally, thematic data analysis was conducted in conjunction with the usage of predetermined themes derived from the reviewed literature (McMillan & Schumacher, 2010). Similar codes were aggregated as the basis from which to form major ideas or themes from the data. Emerging themes from the data were also analysed and interpreted. Similar responses were grouped together into categorical themes. Such identification of themes provided further insights on understanding the individual views and experiences of the interviewees (Creswell, 2012).

Ethical considerations
Anonymity and confidentiality were observed when reporting on the utterances and narratives of the participants, whose names were not mentioned. Instead, pseudonyms were opted for to protect their identity (Creswell, 2012). Ethical measures, which included informed consent (McMillan & Schumacher, 2010), guarding against manipulating participants were applied during the data collection and reporting processes.

Findings and discussion
The themes discussed in this section are derived from the thematically analysed data obtained from the interviews with the six selected participants. These themes are: mentoring role of the supervisor, feedback provision, communication and attributes of a supervisor. It is worth noting that these themes relate to the manner in which the supervisor-supervisee dialogue at a distance learning institution could be constructed and developed in order to diminish transactional distance (Guri-Rosenblit, 1999; Moore, 1997).

Mentoring role of the supervisor
In this study, we had to determine whether the students understood the role of a supervisor. This was particularly important because both the supervisee and the supervisor should know and understand their roles and responsibilities. In response to the question: What do you think the primary roles of your advisor/chair should be? The responses of the interviewees showed that they had a good idea of the role of the supervisor. Three of the participants (indicated below in their pseudonyms) indicated that a supervisor is as mentor: It is evident from their responses that the interviewees viewed mentoring as part of the responsibilities that should be performed by the supervisor. Their understanding was not consistent with the view of Wiegman (2013), which equates mentoring with supervision. Nonetheless, their responses showed that they have a modicum of understanding the role of the supervisor as a mentor.
However, some of the participants' responses captured the various aspects of what a mentor is supposed to do. These aspects varied from giving guidance and direction, providing feedback, giving support, motivating and advising, as discussed in the sections below.

Providing guidance and direction
The students refer to the role of the supervisors as that of guiding and directing students in their research work. Giving directions could simplify the research process to students, as many of them struggle to make sense of the research process before they even deal with the issue they are investigating (Chamberlain, 2016). The following narrative statements clearly express the interviewees' need for guidance and direction: My supervisor must give guidance throughout. (Tshidi) Guide me to give the study shape. (Nkele) To guide me and give direction. (Lufuno) His primary role is to give . . . guidance . . . direction and advise. (Phindulo) To act as a . . . and guide in the process of writing postgraduate reports. (Itani) What these interviewees are actually stating profoundly is that guidance and direction should be provided from the beginning of the study until the end. Indirectly, they are referring to continuous guidance.

Supervisors should provide support
Support is an important component of supervision. post graduate supervision. The interviewees outrightly mention support as a role of the supervisor as in the responses of the four interviewees: His primary role is to give support . . . (Phindulo) It should be to give support . . . (Tshepo) His primary role is to . . . support. (Phindulo) What is interesting in the responses above is that one of the interviewees went on to show that the supervisor could show support by suggesting relevant material to the student, an aspect that is emphasised by Borders et al. (2012).

Advise and motivate
Advising and motivating are some of the critical aspects of mentoring/supervision (Holmes et al., 2010;Zelditch, 1990). The participants expressed this aspect thus: The supervisor must encourage me to keep in touch. He must follow-up on progress.

Peer support
In order to determine whether the students had any peer support, we asked the question: Was there any informal support from your peers studying in the same institution? Two divergent views emerged. On the one hand, some students felt that there was no formal peer support organised by supervisors except for the postgraduate workshops that are organised by the Department of Graduate Studies. In addition, other students expressed that there was informal peer support which helped them share research ideas, as there was no mentoring support from their supervisors. They expressed these views as follows: Too much of it because we were all supervised by this not available supervisor so we depended mostly on us as a group/team, both Masters and Doctorate. (Funo) Not really, more than good wishes and we supported each other when it was time for brown bags session. (Itani) Yes, there was. We would sometimes have meetings wherein we would talk about our challenges. (Phindulo) Yes, we would discuss our studies and learn from each other. (Tshidi) Yes, only through a postgraduate workshop. As we present our work at the seminar, other students asked questions that assisted in developing the study. (Nkele) On the other hand, there was a view suggesting the absence of any form of peer support because fellow students seemed reluctant to share their research thoughts. The latter sentiment confirms an argument by Pemberton and Akkary (2010), who mention that to some students, collaboration may be negative. For example, Mohau (not real name) expressed it in this manner: "No, we worked in isolation, which I do not promote and I feel like my fellow peers do not want to discuss about the research they are conducting".
As discussed in the literature section, collaboration in the form of peer support helps support postgraduate writers to develop their academic identity, even though there is a possibility of damaging students' self-esteem, negatively affecting their learning outcomes (Fergie et al., 2011;Vyncke, 2012). Also, for learner autonomy to be developed, students have to learn to be independent and self-driven (Giossos et al., 2009), which was not the case amongst the six postgraduate students sampled in the current study.

Feedback provision
From the reviewed literature practitioners and scholars such as Wang and Li (2011), and Wood and Louw (2018), feedback to students contributes immensely to their academic development. The responses of the interviewees in the current study are in agreement with the literature. All interviewees indicated that the supervisor is to provide feedback on their work. Interestingly, instead of just mentioning the feedback role of a supervisor, the interviewees qualified the kind of feedback they envisaged. To them, feedback should be constructive and given promptly.
Staff communication and timely responsiveness are very important to distance-learning students. The importance of timeliness was emphasised by Owens et al. (2009)

Communication
Communication is an integral part of student supervision, especially in the distance learning context characterised by insufficient transactional distance, which may lead to insufficient understanding of the supervisor and the supervisee (Giossos et al., 2009;Moore & Kearsley, 1996). In this case, the supervisor should try to create meaningful and quality dialogue that enables the resolution of learning problems experienced by the distance learner (Ali et al., 2016;Moore, 1997). It is in this communication that structure of the research phenomenon and the research process will be addressed. According to Moore's theory (Moore, 1997), structure includes aspects such as clarifying the goals and objectives of the research and the pedagogical model that will be followed. As suggested Bin the literature, distance learning provides flexible and multimedia rich educational environments for the students (Isik et al., 2010). While some students were satisfied with the manner in which the supervisors communicated with them, others expressed utter disappointment at the lack of communication and contact by the supervisor. The participants in the present study concurred that the predominant mode of communication was the emails, and the telephone, sometimes WhatsApp; and occasionally through face-to-face meetings, which are synchronous distance learning (DL) supervision and delivery methods (Sindlinger, 2011).

Emails
Communication between these students and their supervisors was mostly done via emails, as confirmed in the following narrative statements of the interviewees: Mostly via . . . and emails. Mostly I was the one who initiated communication. (Tshepo) We used more of emails. (Nkele) We communicated through emails . . . (Phindulo) We would write each other emails. (Itani) Lufuno indicated that she had to send several emails before her supervisor responded. This shows that some supervisors did not respond to emails, to a point that they had to be reminded by students, as mentioned by Lufuno below: I'm laughing right now because I would say we were communicating via email, but she'll always say she didn't find the document I sent. For her to find my document, I had to send it more than three times. This happened to all her students. This reaction by Lufuno seems to suggest that some supervisors do not strive to lessen the transactional distance, an act which may hamper the research progress of the students (Chamberlain, 2016). It is important to note that students may feel isolated and neglected by the supervisor, who is actually a mentor.

Telephone and cell phone calls
Telephones and cell phones are some of the effective synchronous modes of distance supervision. These participants indicated that this mode of supervision was used by both the students and the supervisors. Some even used WhatsApp calls as they were inexpensive.

Face-to-face meetings
Six students seemed to be happy with the regularity of face-to-face meetings with their supervisors, as expressed below: He would give me feedback through face-to-face discussions. (Tshidi) I had meetings with my supervisor two to three times in a month. (Mohau) As often as there was a need. (Tshepo) We met only when there was a need. Very seldom, but we interacted full-time online. (Nkele) Once every week, and whenever there was feedback on my submitted draft. (Tshidi) We used to meet at least once in a month. (Phindulo) There was no stipulated period, but we would meet at least twice in a term. (Itani) Mostly via meetings . . . Mostly, I was the one who initiated communication. (Tshepo) What it critical is that many of the students struggle with most of online supervision tools, hence the supervisors make sure they even use meetings in order to guide students, this does not come without problems too because students often find themselves having to pay for transport and accommodation because they live far from the institution. Looking at the literature reviewed in section 2 of this article, it was interesting to note that the students in this study, given their context were keener on having face to face interactions with the supervisors even if it meant spending more on, trans port and accommodation. This implies that the it is important for institution to also introduce funding for students to lessen their financial burden. Hence, supervisors could be given time and finances to meet the students at their locations to have discussions with them.
In contradistinction to the apparent regularity of meetings and frequent (student-initiated) communication, Funo stipulates a different view, indicating that there were no regular meetings with her supervisor: "We met by chance when she's not busy at a restaurant most of the time". This lack of communication was surprising, given the fact that she was enrolled in an ODL university, which has fully functioning emails, e-technologies and a variety of online communication tools which should make it easy for communication to take place. Nonetheless, the majority of the students were satisfied with the frequency of the meetings, as well as the quality of the interaction.

Attributes of a supervisor
The interviewed postgraduate students in this study also referred to the attributes of the supervisor. As a mentor, the supervisor needs to display some specific academic, personal and personal leadership qualities (Borders et al., 2012;Henry et al., 201). The attributes enumerated by the participants were research expertise, professionalism and the ability to plan.
Regarding research expertise, Itani declared that: "The supervisor must also suggest sources relevant to the field and the study that I am doing". In addition, Lufuno expected the supervisor to hold guide her expertly through the research process. To this effect, she averred that, "The supervisor must work with me on my study as she/he is more experienced than me". Nkele had this to say regarding professionalism: "Adherence to professional and research standards". Planning was also mentioned by Mohao, who mentioned that, "The primary role of the supervisor is to help student define and develop realistic study plan through schedule planning".
When responding to the question: Did your supervisor have expertise in your thesis topic? Why do you say so? The responses of the participants generally related to the supervisor's expertise. Interestingly, the participants attributed various reasons correlating to the supervisors and the requisite expertise. Some supervisors had published articles and books in their fields, some gave relevant examples and some were familiar with the research process, as encapsulated in the following statements: The supervisor has the experience in this field of the topic under investigation and has also published extensively in the field. (Mohau) Yes, she did. For example, she could realise that the first instrument that I developed for data collection would not yield what we expeced (Nkele) Indeed, he was published substantially in the field. (Tshidi) He had a lot of expertise in my thesis topic. He has written articles and books (Phindulo) Yes, because of the examples that he would give were seminal sources in the field. (Itani) No, but had expertise in the research process. (Tshepo) However, Lufuno is of a different opinion. She thinks her supervisor did not have expertise as she used Lufuno's information in her own presentations, suggesting that she covertly used her (Lufuno) as her own personal researcher. To this extent, she complained thus: "I suspect because some of the theories she learnt from me, and even bragged to me that she went for an interview somewhere and was telling them about the theories in my thesis".

Conclusions and recommendations
This article investigated the experiences of PhD postgraduate students' supervision at a distance learning university in South Africa. The main question answered by the study was: What are the experiences of the post graduate students that are supervised at a distance? The prominent issues that emanated from the findings were: the need for peer mentoring, supervisor mentoring, the supervisor's degree of expertise based on the students' research topics, and the use of technology as a means of communication between students and their supervisors. The principal issue encompassing these findings is the kind and quality of the interaction, since students viewed supervision as a factor of mentorship and guidance. These finding s point to a need for ODL institutions to differentiate their interaction with students considering that these institutions have different contexts that may need different way of supervising. This article shows that, given the rural context and the age of many students, the students may need more personal interaction with the supervisors.
On the one hand, some students felt neglected by their supervisors, and felt isolated from other students enrolled for postgraduate studies. It emerged that there was no formal peer support organised by supervisors. Students indicated that peer support is mostly informal. Based on the latter, it could then be concluded that the interviewed students were particularly in need for an opportunity to share their learning experiences with other students.
On the other hand, some students held the view that their supervisors satisfactorily attended to their concerns. However, in contrast to the findings of Alston et al., 2005), participants in this study were not seeking contacts with other students and peer groups for social reasons. This finding seems to confirm that peer review is less likely to happen because fellow students may be reluctant to share their research thoughts as argued by authors cited in the literature review section of this article.
The means of communication were emails, the telephone, WhatsApp calls, and face-to-face meetings. The students seemed satisfied with the regularity of face-to-face meetings with their supervisors.
The participants made a number of recommendations, which related to timeous feedback, exposing the students to doctoral defence programmes, supervisors taking a reasonable number of students under their care, more regular face-to-face-meetings, and support.
In view of the findings of the study, the following recommendations are made: Firstly, communities of practice can be seen as a strategy for addressing some of the problems identified as specific to postgraduate research, such as isolation and the need for longer-term maintenance of momentum. Guardian supervisors, cohorts of PhD students and latterly the team of distance supervisors can be identified as communities of practice which develop and share ideas and build good practice over time. As per literature reviewed in this article, this recommendation, therefore, cohere with the argument propounded by researchers in their proposition for the preparation of postgraduates for participation in the research fraternity (community of practice) rather than as apprentices being readied for artisanship.
Secondly, supervisors of international students at a distance can benefit, we believe, from being involved in a community of practice which can enable them to develop shared values, and good practice, staying in touch with institutional demands and provision so that they do not become unnecessarily isolated, but supported in their work. To this end we suggest that an online supervisory support and development programme be initiated.
Lastly, based on the notion that some ODL institutions are still enrolling huge number of post graduate students who are ICT illiterate and less prepared for postgraduate, many of the students struggle to get the desired academic achievement. Thus, it is recommended that future research be directed at exploring the ways in which this problem can be solved.