Mentorship of early career academics in Tanzania: issues and implications for the next generation of academics

ABSTRACT Mentoring has been pointed out as one of the strategies for early career academics (ECAs) preparation and support for smooth transition into academia. Although ECAs mentoring has been widely studied, and the findings have indeed been informative and illuminating, the issues facing ECAs have yet to be adequately addressed, particularly in Tanzania. Drawing from open-ended questionnaires, focus group discussions, and interviews, we examined the issues surrounding the mentoring of ECAs in selected Tanzanian universities. Academic exploitation, isolation, and lack of support from senior academics were among the key issues facing ECAs. The findings further revealed that regular mentorship training, peer support, and institutional policy framework as strategies in place to subdue the mentoring challenges. We argue for universities to have policy framework on professional development through ECAs mentoring to fill out the gap on pedagogical, research and public engagement deficits for ensuring capable next generation of academics.

effective support system. Mentoring is important to ECAs in that it fosters professional skill development, professional confidence, scholarly productivity, career advancement, reduces stress and enhances job satisfaction (Harker, O'Toole, Keshmiripour, McIntosh, & Sassen, 2019). Certainly, effective mentoring relationships can advance ECAs and other faculty by providing them the means and support to cope with challenges in their career path (Oberhauser & Caretta, 2019, p. 56). Since not all universities offer pedagogical training in their doctoral programmes, mentorship remains the best and less costly option to orient ECAs into HE teaching while continuing to work. More importantly, mentoring is more useful because it is an ongoing process that takes place all time long in their career path even when there is a change or paradigm shifts in HE teaching.
While Universities' core activities are teaching, research, and community engagement; studies indicate that little emphasis is placed on preparing ECAs to fulfil such roles (Budge, 2014;Leibowitz, Bozalek, Van Schalkwyk, & Winberg, 2014;Stes & Van Petegem, 2011). This lack of preparation results in ECAs facing challenges ranging from insufficient knowledge of the language of academia to difficulty teaching within a challenging 'student surge' environment (Ssempebwa, Teferra, & Bakkabulindi, 2016). Moreover, achieving quality teaching in many universities today poses a major challenge due to massification policies coupled with diverse students in terms of age, learning experience, socioeconomic status, and cultural background. Therefore, there is a need to prepare ECAs systematically for working in such diverse and challenging academic environments to facilitate their effective teaching, job commitment, and research output.
The global university ranking game as a key reference for quality and effectiveness of institutions (Bernasconi & Véliz, 2016) has brought universities into league tables partly driven by increasing competition, audit requirements, growing HE sector, and great emphasis on university ranking (Probert, 2013). Such an increase in competition among HE institutions has prompted the management of some universities to opt for socializing their academics and students in both research and teaching to ensure they maintain their positions in university league tables (Jarvis, 2019). Along with such changes, in some countries there is notable casualization of academic staff which involves hourly paid academics (Bennett, Roberts, Ananthram, & Broughton, 2018).
In the university systems of some countries, ECA mentoring has, to a large extent been a norm due to its ability to reduce feelings of isolation and to increase confidence, self-esteem, and professional growth (Hobson, Ashby, Malderez, & Tomlinson, 2009;Marable & Raimondi, 2007). For example, a study conducted in European countries indicated a formalization of mentoring focused on teaching, with 52 of 82 surveyed UK universities (63.4%) having a policy that required new academics to obtain at least a postgraduate certificate in teaching (Gosling, 2010as cited in Teferra, 2016b. Specifically, to teach in one of these universities, academics have to be accredited through the UK Professional Standards Framework (UKPSF) (HEA, H. E. A, 2011). Such accreditation is achieved by meeting benchmarked standards, either by completing an accredited course or by submitting a portfolio of evidence to be mapped against the standards (Jarvis, 2019). In educational systems of other countries such as the US, Hong Kong, Germany, and Sweden, ECA mentoring, especially in teaching, has been made university practice through induction and specific facets of teaching in HE (Bean, Lucas, & Hyers, 2014;Lee & Bush, 2003). The purpose of such preparation is to ensure ECAs' consistent performance and outcome among academics, as well as their satisfaction, retention, and satisfactory promotion rates.
Although not all doctoral graduates join universities as academics, those who do enter academia with little or no pedagogical preparation for teaching and other university responsibilities often find that the job requirements do not match their university training and skills. Consequently, some ECAs display a lack of pedagogical competence in their teaching (Greer, Cathcart, & Neale, 2016), leading some of them to criticise their university curricula for producing 'half-baked' graduates (Behari-Leak, 2017). Although some universities in Australia, Canada, the UK and USA have introduced teaching academics, Bennett et al. (2018) revealed that the career path for teaching academics is uncertain surrounded by an absence of role models in their profession among other things. Lack of role models may suggest the need to have mentors as role models so that ECAs have role models in their professoriate. Teferra (2016) questioned how ECAs can learn to teach if their doctoral training does not include teaching certification. Some evidence indicates that ECA mentoring is a relegated responsibility in some countries because of the informal nature of the mentoring process (Gale, 2011;McLean, 2004;Montgomery et al., 2014;Nakanjako et al., 2011;Teferra, 2016;Udegbe, 2016). Clearly, not all academics are trained as university teachers during their professional training, while others, despite being trained as classroom teachers, lack the ability to lecture -which is a dominant mode of teaching in universities. Such weakness affects not only the academics' ability of lecturing but also students' understanding when are taught by teachers who lack best pedagogical skills to teach university students.
Although the topic of ECA mentoring has been widely studied, and the findings have indeed been informative, two issues have yet to be adequately addressed, specifically in Tanzania: 1) the challenges facing ECAs in their job; and 2) potential strategies for successfully addressing these challenges. To fill this research gap, this study was pioneered to answer the following two research questions: a). What challenges do early career academics face regarding teaching, research, and delivery of public service? b). What strategies are in place to address such challenges? The theoretical reasoning of this paper builds on the previous HE research in early career academics and policy frameworks that would govern ECAs effective professional development.

Conceptual framework for mentoring ECAs
The Cognitive Apprenticeship Model (Collins, Brown, & Holum, 1991) is a theory of social learning that requires collaboration of learners in a community of inquiry with experts and peers. Notwithstanding some criticisms, cognitive apprenticeship model was chosen in this study over other existing mentoring models because it is One-on-One mentoring model and largely converges on the advancement of cognitive skills for multifaceted career preparation which is hardly offered by other models such as mentoring panel and group mentoring. In HE, a cognitive apprenticeship model can be employed to prepare potential generations of academics by stressing on the use of knowledge and skills to varied circumstances. Based on this model, we argue that, in order for ECAs to learn how to teach, conduct research, and engage in public service, a mentoring arrangement is required that involves peers and senior academics in their respective universities and departments in particular.
In the present study, the mentoring model used involves ECAs and their mentors in interdependent work-related practices and social interactions (Brown, Collins, & Duguid, 1989). These work-related practices involve explicit cognitive processes and work-related skills and tasks stemming from reciprocally advantageous collaborative working relationships that lead to mutual reciprocity between mentee and mentor (Kram, 1988). This mentoring model is believed to facilitate collaboration and shared problem-solving by involving ECAs and mentors in a reciprocal exchange of ideas and information (Mullen, 1998). The model employs six mentorship methods to develop expertise -modelling, coaching, scaffolding, articulation, reflection, and exploration. Figure 1 presents the Cognitive Apprenticeship Model.
While the mentor initiates these methods, the model does not position the mentor as the sole expert; rather, during the cognitive apprenticeship process, the mentee (the ECA in this context) ought to be knowledgeable about methods of cross-examining drawn from scholarly tradition and about using explicit heuristics and structured learning tasks to shape knowledge while engaging in realistic practice. It is believed that, by engaging in learning guided by more experienced colleagues, mentees will become cognitive experts. Overall, ECAs' effective engagement in teaching practice requires the regular support of experts particularly on pedagogical aspects, since teaching usually involves practical difficulties directly connected to problem-solving (Kuo, Hwang, Chen, & Chen, 2012).

Tanzania's higher education system and academic profession
Higher education in Tanzania, as in many countries, marks the final level of formal schooling. HE is made up of universities, university colleges, and university centres and involves both public and private providers. Completion of HE programmes takes three to five years, depending on the specific degree programme requirements. In terms of governance, HE institutions in Tanzania are semi-autonomous, with the Tanzania Commission for Universities (TCU) being in-charge of coordinating and overseeing both public and private universities and ensuring that no university becomes a degree mill.
The history of Tanzanian HE can be traced back to the early 1960s, soon after the country had gained its political independence from Britain, in 1961. Thus, the history of HE and what was then Tanganyika's political independence are intertwined. Upon independence, Tanzania (then Tanganyika) had only one HE institution (i.e. Dar es Salaam University College) which was an affiliate college of the university of East Africa together with Nairobi University College of Kenya, and Makerere University College of Uganda, before eventually becoming the University of Dar es Salaam (Ishengoma, 2007;Mkude, Cooksey, & Levey, 2003;Msola, 2005). Since 1960s, the HE sub-sector in Tanzania has slowly grown from 13 students within a single faculty to a system of 43 universities (as of 2019). From the 1960s to the early 1990s, HE in Tanzania was entirely provided and largely financed by the state.
However, following the introduction of neoliberal policies and the resultant increased demand for HE graduates from the mid-1990s on, the landscape of HE provision has expanded to allow private providers (Ishengoma, 2007;Mkude et al., 2003). Since then, private providers have been legally permitted to establish and operate HE institutions and grant academic degrees, as per the Tanzania Universities Act of 2005 (Mgaiwa & Ishengoma, 2017;Ishengoma, 2007;TCU, 2015). As of 2019/20 academic year, Tanzania had 43 universities and universities colleges with a total enrolment of 209,144 students (NBS, 2019), far behind the 510,685 students then enrolled in neighbouring Kenya (Kenya National Bureau of Statistics, 2017, p. 55). Of the 43 universities and university colleges in Tanzania, 16 are public universities and university colleges, while 27 are private (NBS, 2019).

Preparation of academics for teaching in Tanzanian universities
Globally, several countries have introduced a mandatory pedagogical training to university academics (Hanbury, Prosser, & Rickinson, 2008;Ödalen, Brommesson, Erlingsson, Schaffer, & Fogelgren, 2019); due to the introduction of massification policies in HE (Shaw, 2018). However, the deficit of HE teacher training appears to be a global phenomenon. For example, in their seminal work covering 17 countries including Australia, Argentina, Canada, United Kingdom, China, Italy, Hong Kong, Japan, and others, Bennion and Locke (2010) revealed that only a few of academics were reported to have received training in pedagogical aspect or instructional skills of teaching. Additionally, 37% surveyed UK universities had no policy that required new academics to have postgraduate certificate in teaching (Gosling, 2010as cited in Teferra, 2016. Equally, a study by Aškerc and Kočar (2015) in Slovenia revealed that academics did not even have minimal theoretical pedagogical knowledge to teach. These findings suggest that doctoral programs in these countries offer virtually no preparation in pedagogy other than research.
Similarly, doctoral training in Tanzania, as in some countries across the globe, strongly focuses on developing students' research skills at the expense of their pedagogical competency, especially for those aspiring to work in the academia. Consequently, ECAs are less prepared for the pedagogical requirements of their academic roles as university teachers. Similar observations have been made by other scholars; for example, Haynes and Petrosko (2009) noted that graduate programmes typically prepare students well for research but are less successful in preparing them for academic careers, particularly HE teaching. As such, some scholars have argued that the problem confronting African HE is not just the lack of eligible lecturers but also the fact that doctoral students are ill-prepared to teach upon completion of their doctoral studies (Udegbe, 2016). This observation suggests the need for a more specific strategy for preparing ECAs for HE teaching not only in terms of pedagogical aspects but also developing their research and public service engagement competency.
The shortage in the Tanzanian labour market of PhD graduates qualified to work in universities compels most universities to employ junior staff in training posts, for which they must apply to enter university master's and doctoral programmes. However, in the course of their postgraduate training, only those pursuing educational programmes take a formal course that would prepare them for teaching. Since there is no training that result into certification for higher education teaching, universities in Tanzania assume that anyone holding a master's or PhD certificate and having an acceptable grade point average (GPA) can teach in a university. This assumption is misleading because HE teaching is a practice that requires one to be trained to teach effectively; in fact, even academics with a background in education need a formal training to teach in HE as it differs from teaching in junior and high schools, and requires specific teaching skills. Scager, Akkerman, Pilot, and Wubbels (2017), drawing on Enyedy, Goldberg, and Welsh (2006) and Kelchtermans (2009) established that HE teaching involves a multifaceted mix of obligations and inherently involves dilemmas, and that addressing these dilemmas requires a mechanism to facilitate effective teaching, especially for ECAs.
There is a paucity of research on ECA mentoring in Tanzania, as in many African countries, leaving such critical questions as how ECAs learn to teach, conduct research, and engage with the community, and the mentoring challenges they face in their careers largely is unaddressed. Additionally, some studies (e.g. Mgaiwa, 2018) indicate that at the university of Dar es Salaam, the country's flagship university, research output has been slowly declining compared to universities in other African countries, such as South Africa, Ghana, and Nigeria.
ECA mentoring is important because of past government public sector employment freezes. Consequently, academics in public universities in Tanzania today are aging and are dominated by junior academics, hence the need to mentor new faculty members. However, little attention is given to preparing ECAs to take up their roles once they join universities (Nakanjako et al., 2011;Reddy, Searle, Shawa, & Teferra, 2016;Udegbe, 2016). It is against this background that this research became imperative and timely, aimed at answering the following research questions: What are the issues that affect ECA mentoring in Tanzanian public universities? What are the potential mentoring strategies for ECA mentoring in Tanzanian universities? In answering these questions, this study situates the mentoring issues and challenges and their associated devastating consequences in the broader context of African universities and those outside Africa with similar mentoring contexts. The study seeks to devise effective mentoring strategies in Tanzanian HE through institutional policies and practices.

Method
This was a qualitatively driven mixed methods study in which both qualitative and quantitative data were collected to determine the perception of different groups and assess their attitudes and opinions on issues surrounding ECA mentoring in Tanzanian universities. This approach was chosen because it enabled the collection of qualitative and quantitative data from a wide population in a short period using a range of methods (Creswell, 2014, p. 4). Also, a descriptive survey is an appropriate means of describing the incidences, frequency, and distribution of population characteristics. Additionally, the approach is amenable to generating credible information from a large group of people within a short time at a relatively low cost (Ary, Jacobs, Sorensen, & Razavieh, 2010).

Research setting and sample
This research was conducted in three research sites in Tanzania (i.e. the University of Dar es Salaam, Mkwawa University College of Education, and Ardhi University). In accordance with ethical guidelines governing research, these universities were anonymised during the analysis and reporting of the findings. All three universities were purposively selected, for three reasons -i.e. their accreditation status, their representativeness (in terms of degree offerings that reflect the profile of other universities in the country), and their age. In terms of accreditation, all three universities were accredited and chartered at the time of data collection. Regarding their representativeness and age, the University of Dar es Salaam was the oldest of the three universities and had both the largest and most diverse number of degree programmes and the largest number of academics in the country; Ardhi University was selected to represent recently established universities with programmes focusing on land use and earth sciences; while Mkwawa University College of Education was chosen to represent recently established university colleges in the country that focus on education degree offerings as part of the country's response to its teacher shortage.
To ensure this research was carried out in accordance with ethical principles, the researchers obtained ethical approval from the University of Dar es Salaam before communicating with the surveyed universities and research participants. A research consent form was prepared, which all invited participants signed after receiving assurances concerning their anonymity and the confidentiality of the information they would provide. During the analysis and reporting of the findings, pseudonyms (e.g. University 'A', 'B', etc.) were used to represent the universities, and refrained from using personal names when quoting participants; instead, initial ECAs were used to represent the names of participants from each university (e.g. ECA, University A, B, or C). The targeted population for this research comprised all academics, heads of department (HoDs), and directors of human resources and administration (DHRAs). Academics were purposively included because they were thought to have information on institutional issues regarding ECA mentoring, while DHRAs and HoDs were purposively selected because they are concerned with the day-to-day activities of all staff development programmes at their institutions.

Data collection and analysis
Three data collection techniques were employed in this research -focus group discussions (FGDs), interviews, and questionnaires. Qualitative data was collected through FGDs and face-to-face semi-structured interviews. Nine interviews were conducted (three in each university surveyed), and a total of six FGDs were held (two in each university), with each FGD consisting of five to six participants. Initially, a total of 200 questionnaires were distributed to randomly selected ECAs in the three studied universities. However, only 165 questionnaires were finally returned, representing 82.5% response rate. Of the 165 respondents, 114 (69.1%) were males while 51 (30.9%) were females. The respondents' distribution for this study is presented in Table 1.
The data collected through FGD was based on the interactions within the group. Each FGD lasted for about one hour, while the duration for interviews ranged from 35 to 45 minutes. Interviews and FGDs were deemed appropriate because of their flexibility in that both enabled researchers and respondents to obtain extra information. Their views were compared with those of the academics in their respective institutions to validate the data, hence providing triangulation and complementarity. To be included in the FGD, staff had to hold a lower academic rank -i.e. a tutorial assistant, assistant lecturer, or lecturer (PhD graduate) with less than five years of service in academia. Both interviews and surveys were self-designed based on the intensive literature review and personal experience of the researchers. The first step of data collection consisted of semistructured questions through in-depth interviews and FGDs to ECAs in the three surveyed universities. Then, a paper and pencil survey with quantitative items were distributed to ECAs to authenticate information obtained through interviews and FGDs. The semi-structured interview questions were related to respondents': (1) most significant career events regarding socialization into academia particularly related to pedagogical aspects; (2) Key mentoring challenges facing ECAs regarding their teaching, research, and public service.
To ensure the quality information is collected, data were iteratively generated and analysed (Carlsen & Glenton, 2011). Data collected from FGDs assisted in creating initial themes which were further explored by face-to-face individual interviews. Questionnaires were conducted after the analysis of the FGDs and interviews to make the triangulation more effective. Thus, DHRAs and HoDs who participated in interviews were able to offer information on ECAs' institutional mentoring issues. Data collected through interviews and FGDs was all recorded-on audiotape. Quantitative data was collected through semi-structured questionnaires and served to complement the information collected through interviews and FGDs. All necessary procedures, including seeking informed consent, were followed. During FGDs, participants were first asked icebreaker questions regarding their work experience and afterwards, they were asked the core questions regarding ECAs mentoring. Specifically, questions revolved around what challenges they face as ECAs based on their core functions of academics and how such challenges they think can be addressed. By acting as a moderator in each FDG, the researchers were provided with high-quality data in a social setting, as respondents used their and others' experiences to explain the cases they presented.

Data analysis procedures
The data collected was sorted and placed into categories, in accordance with the research objectives. The qualitative data obtained from FGDs and interviews was analysed through thematic analysis, while the quantitative data obtained through questionnaires was analysed using the Statistical Package for Social Sciences (SPSS) Version 20.0, which generated the frequencies and percentages used in the reporting of the findings. Because the study used a mixed methods approach, data collected through questionnaires complemented those obtained through interviews and FGDs for triangulation purposes, both to make the findings more reliable and valid and to simplify and widened the scope of their interpretation.

Findings
The overall aim of this research was to examine the issues surrounding ECA mentoring in Tanzania public universities and identify strategies that could be used to address these issues.

Lack of institutional mentorship arrangements
While creating formal institutional arrangements for mentoring is critical, six out of the nine interviewed HoDs and DHRAs indicated there were no formal mentoring arrangements for ECAs other than informal arrangements that were always not monitored by university management: We do not have any formal mentoring of early career academics. However, this does not mean that there is no mentorship taking place at all. [From what] I know, some senior staff involve junior academics in their research for data collection and analysis (DHRM, University C).
One HoD extended a similar statement: We don't formally assign employed ECAs to mentors, assuming that senior staff know that mentoring is one of their responsibilities because it is stipulated in their appointment letters that will have to do some mentorship for ECAs (HoD, University A).
From the other side of the equation, an experienced ECA from University B proffered the following comment on mentoring arrangements for ECAs at his institution: . . . the practice of mentoring is not visible at our institution especially in research part, but in teaching process there is an element of mentoring because the process is guided by faculty meetings where formal arrangement can be made on teaching, e.g., teaching matrix. Regarding research, young academics are to struggle in their own as it is treated as a personal issue (ECA, University B).
A similar observation was made by an ECA at University A,' Beside the wisdom of the leader, which can be [offered] clandestinely, there is no formal mentoring arrangement. It is [dependent on one's] personal will to arrange mentoring because there is no formal institutional arrangement (ECA, University A).
Data collected through questionnaires largely supported the qualitative data. For example, 126 participants (76.3%) responded 'NO' when asked whether or not there were any mentoring arrangements for ECAs in their respective institutions, while only 39 (23.6%) responded 'YES.' Although 23% seems to be a small number, it might arguably suggest there was some form of mentoring in some surveyed universities. Lack of mentoring among ECAs affected their early class teaching, with some complaining that it was as if they had been thrown into the classroom without having anyone to assess how they taught and provide feedback that would help them to reach a higher level of teaching competence. Therefore, since HE teaching certification is not an employment requirement in Tanzania's HE institutions, there is a need to capitalize on mentoring so that ECAs can become effective in their teaching.
At University C, the DHRA, after admitting that her institution had no formal mentoring arrangements for ECAs, was asked to explain how ECAs with no teaching background learn how to teach. She responded: Although we believe that teaching cannot be learned in a semester or through induction, rather it is a lifetime career experience, I think there is a need to [sic] rethink it by learning from other universities (DHRM, University C).

Lack of institutional mentorship guidelines
While there is pressure for universities to ensure threshold standards for their academics' research and teaching so as to maintain their positions in university league tables (Jarvis, 2019), there are concerns about the lack of guiding policy instruments available to achieve such a goal. One thing that was clear in participants' descriptions of their institutions was their discernment of governance instrument discourses. Both interview and FGD participants described their university as lacking important instruments for guiding the socialization of ECAs into academia. While the participants' descriptions of this weakness differed from one institution to another, most emphasised the lack of a policy framework that would encourage mentoring practice in teaching and research: As you know, the scope of mentoring is wide, and we as an institution have no strict internal guideline requiring assigning each ECA we employ to a senior staff (DHRA, University B).
Although some participants provided fewer examples to support their arguments, they still emphasized that the lack of a policy framework stifled effective ECA mentoring. At University A, for example, one of the HoDs reported that their university had neither a formal mentorship programme nor policy guidelines to support such a process: As an institution, I would say that we have no policy on that, and it has been an individual wish for senior staff to help in mentoring junior staff, although the process might be critical (HoD, University A).
The ways in which interview and FGD participants spoke about policy frameworks indicate even more clearly the issues they raised. While policy guidelines are believed to be important instruments for guiding institutional practices and the effective implementation of institutional plans, the foregoing accounts provide evidence that universities in Tanzania lack not only formal mentoring arrangements for guiding mentoring of their ECAs but also institutional policy guidelines for mentoring in general. The key issues arising from this lack of a formal mentoring policy and practice are the inadequate pace of ECAs' academic growth and the poor quality of their teaching, which has led to students boycotting some lecturers because ECAs do not understand how to teach or fail to understand student needs: "Sometimes I taught so fast such that students could not understand, and as a result, they boycotted my course. Policy guidelines would not only require mentors to evaluate their mentees' and provide feedback thereon but also highlight the need for mentees to understudy their seniors' teaching, research, and public engagement. Thus, policy guidelines would be effective in sustaining the mentoring process. As matters stand, it is difficult for mentors to abide or maintain their mentoring because mentoring is not an institutional requirement and ECAs cannot officially demand it.
Isolation of ECAs from senior staff: Although the cognitive apprenticeship mentoring model facilitates collaboration and shared problem-solving in a reciprocal exchange of ideas and information (Mullen, 1998), in this study, ECAs described having been isolated by senior academics. During the FGDs, ECAs identified isolation as one of their worst on-the-job experiences: Senior academics tend to isolate novice academics especially us TAs, and they are not supportive but usually give us nick names like 'nyoka', metaphorically meaning that we are useless because we are not supposed to teach until we go for further studies (ECA, University C).
While ECAs in the surveyed universities felt isolated due to a lack of mentoring, Remmik, Karm, Haamer, and Lepp (2011) indicate that mentoring helps to create communities of practice for teaching and to reduce academic isolation among novice academics. Although it is clear that studies have shown that mentoring activities help ECAs to join learning communities through which they experience cooperation and support from their mentors (G. S. Åkerlind, 2005;Barrett, Ballantyne, Harrison, & Temmerman, 2009;Walker, Gleaves, & Grey, 2006). In contrast, some of the ECAs in the present study noted that isolation was one of the issues confronting them: 'Just imagine we are isolated from participating in assessment of teaching practice of our students' (ECA, University B). Although isolation of ECAs is used to benchmark assessment of quality, especially in the case of sensitive activities like teaching practice, ECAs still questioned how they would learn such duties because the job assignment they received after completing their PhD including teaching practice, was not a part of their doctoral programme: 'But the question is how will we learn to assess because our master's and PhD degree will not be related to assessment of teaching practice? To me, this is isolation because the activity involves some payments (ECA, University C). Another FGD participant added that, in some circumstances, some senior staff did not feel comfortable even having coffee in the same restaurant with ECAs.
I have learned that because of their academic arrogance, senior academics sometimes see ECAs as helpless staff, so it is very rare to find senior staff involve ECAs in their research project for the purpose of learning and when there are some incentives (ECA, University B).
The foregoing accounts reiterate that apart from the universities lacking a formalised system of mentoring, the culture of collaboration itself is weak among academics; this seems to be rooted in the hierarchies of academic ranks and academic arrogance, the effects of which are more pronounced among the ECAs as they are so dependent on senior academics. Some ECAs described being isolated from senior academics in many of their jobs which they felt stifled their career growth and negatively affected their professional abilities, teaching and research confidence, and community engagement. More than half of the ECAs participating in the FDGs raised the issue of isolation in some activities such as teaching practice supervision, opportunities to under-study their mentors in classrooms, and administration.
Academic exploitation: Academic integrity and excellence are principles that universities and their academics alike are required to uphold. However, university academics have never totally been free of academic misconduct (Gatmen, 2011). Although academic exploitation was not an issue across the surveyed universities, some ECAs raised it as a challenge to the mentorship relationship in their respective universities. While the Cognitive Apprenticeship Model assumes that mentoring can create a reciprocally advantageous collaborative working relationship between the mentee and the mentor (Kram, 1988), this study established that some ECAs in some of the surveyed universities were unfairly used intellectually by their seniors, by having to participate in their mentor's research without compensation, acknowledgement, and authorship: I was asked to collect specimens, and later, I did some lab work and crafted part of the manuscript without any agreement, albeit with some payment for data collection. To my surprise, I just saw a paper published, and I was not one of the authors (ECA, University A).
Another statement, from a participant in University C echoed this: My supervisor, who is also my fellow senior staff, gave me an assignment for a major paper in the class. I then collected data and wrote an assignment . . . To my surprise, some years later, I found some data I used in my assignment was used in his publication without acknowledgment or credit (ECA, University C).
Although the issue of academic exploitation was not common in the surveyed universities, the foregoing quotations indicate that there was no clear agreement regarding ECAs' engagement in their mentors' work, nor concerning the assessment and acknowledgement of their intellectual contributions. The quotations further indicate there was unethical use of ECAs' work and intellectual contributions which is clearly an unfair dealing in research practice.
Overall, DHRMs and HoDs had similar thinking on whether or not there was any formal mentoring for their ECAs. They generally indicated that there was no formal mentoring, but they all argued for the existence of an informal mentoring arrangement for ECAs. On the other hand, ECAs were generally of the view that there was no formal mentoring and that they were not adequately supported by the university after their appointments for a smooth transition into academia.

Potential mentoring strategies
The second objective of this study was to examine potential strategies for effective ECA mentoring. To this end, the researchers collected qualitative data through FGDs, the analysis of which yielded three thematically generated strategies: (i) providing regular training for mentors; (ii) putting mentorship frameworks in place; and, (iii) establishing peer support as part of mentoring. Participants believed that having these strategies in place would make mentoring more effective and improve both the pace of ECAs' academic growth and the quality of the services they offered to students, especially in terms of teaching.
Regular training for mentors: With the idea of giving mentors an opportunity to share their wealth of experience and knowledge with the upcoming generation of academics (Mendez et al., 2019), ECAs raised regular training as a major way of improving mentoring and subsequently the teaching quality of ECAs. They emphasised that regular training is important for moulding effective mentors: I believe that if mentors are given regular training [sic] it will help to mitigate some of the challenges that we discussed today. For example, issues of authorship and unfair academic dealing can easily be dealt with if mentors are trained (ECA, University C).
FGD participants in some surveyed universities insisted that such training should be provided regularly to allow mentors to refresh their experience: The fact is that knowledge changes over time. It is important that mentors are regularly trained to keep up to date with mentoring practices (ECA, University A).
If appropriate training was provided, it would help mentors avoid inappropriate mentoring practices that could harm the mentees with whom they had relationships. Although mentoring has been challenged by research and administrative demands (Sambunjak, Straus, & Marušić, 2006), regular training will increase not only mentees' career success but also their career selection, advancement, and productivity.
Institutionalization of mentoring guidelines: Participants also expressed the need for institutional frameworks to increase the effectiveness of mentoring activities among ECAs by clearly stating both mentors' and mentees' obligations: It is important to have an institutional instrument like a guideline to guide the conduct of mentoring. I believe it will help to have a common standard for the process and a fairer relationship between mentors and mentees (ECA, University B).

Fostering peer support
Given their difficult experiences with mentoring, ECAs in the surveyed universities were of the view that peer support mechanisms should be established to enable ECAs to support one another, such as by sharing experiences that helped them to learn. During an FGD, one ECA stated: . . . [sic] I would suggest that ECAs support each other, I believe, our experience and knowledge about the work is different. Therefore, learning from each other is a useful tool to grow together as colleagues (ECA, University A).
Another ECA suggested that peer support was also needed as a way of overcoming isolation: 'Because sometimes we are isolated, I would love to see we are supporting each other based on our own potentials as ECAs' (ECA, University C).

Discussion
As noted in the findings section, four key issues were found to affect ECA mentoring in Tanzanian universities -i.e. a lack of formal institutional mentoring arrangements, a lack of institutional mentoring policies or frameworks, the isolation of ECAs from senior academics, and the exploitation of some ECAs by senior academics. These findings resonate somewhat with the results from other studies conducted in African universities (see example, Nakanjako et al., 2011;Teferra, 2016;Udegbe, 2016). As such, the deficits in mentoring arrangement, policies and support is a global phenomenon (e.g. Johnson, 2016;Nganga, Bowne, & Stremmel, 2020;Scager et al., 2017). Although lack of mentoring is a global issue, it appears to be more commonly adopted as standard practice within European universities than in the African context. Moreover, in this research, neither policy frameworks nor formal mentoring arrangements appeared to exist in the surveyed universities, supporting the findings from other scholars in African countries. For example, a study conducted by Nakanjako et al. (2011), at Makerere University indicated that many ECAs did not have mentors, as the university had no formal mentoring arrangement. As such, mentors were not adequate for those who needed mentoring, and mentoring occurred in an ad hoc manner.
This lack of mentoring arrangements and mentoring policies is a problem not only in Tanzanian universities but also in universities in other countries in the region. Arguably, this state of affairs might suggest several issues concerning mentees, such as the inadequate pace of ECAs' professional growth and job dissatisfaction (Bean et al., 2014), and the poor quality of teaching. Specifically, these findings may cause ECAs to lack pedagogical skills to teach, insufficient confidence to defend their claims as academics and the ability to do research, both of which would have been learnt through a community of inquiry with peers and experts as advocated by Cognitive Apprenticeship Theory (Collins et al., 1991). Indeed, this deficit might affect academics' pedagogical ability to teach in HE, because universities in Tanzania do not offer higher education teaching certification for those aspiring to be or who are already employed as university academics. As a result, ECAs are likely to face challenges in handling classes because HE teaching is different from that found in secondary schools. This situation will affect not only ECAs but also their students, particularly through the poor quality of ECAs' teaching, which some scholars argue may result in student attrition (Willcoxson, 2010).
While these findings echo what other African scholars (Lunsford, 2012;Montgomery et al., 2014;Thomas, Lunsford, & Rodrigues, 2015) have found regarding negligence and inadequate mentoring, in contrast, findings from South African universities show a supportive environment for ECAs through a variety of programmes, including mentoring and the University Education Induction Programme (UEIP) (Reddy et al., 2016). While engagement in formalized mentoring programmes has been found to be effective for ECAs (Mendez et al., 2019), the present study suggests that the lack of such programmes may negatively affect the productivity of ECAs in terms of research publications and community engagement. Additionally, Åkerlind (2007) argued that a lack of mentoring for ECAs may result in research misconduct, because ECAs are not prepared to accept and grow within the bounds of an ethical academia career path.
Policy frameworks that make the mentoring process clear can attract senior academics and motivate them to see mentoring as a means of fulfilling their careers and influencing ECA's personal development, research productivity, and success in obtaining grants. Mentoring arrangements would also provide a harmonious environment for peer support among ECAs. Other studies, conducted in Mozambique, Australia, and the UK, for example, indicate that when ECAs do not have a supportive environment they may end up demotivated and demoralised, and may even quit the profession (G. Åkerlind, 2007;Cossa, Buque, & Fringe, 2016;Hemmings, 2012;McKay & Monk, 2017). Since mentoring is the most trusted process for developing teaching skills and research knowledge, skills, and confidence among ECAs in HE (Gale, 2011;Kearney & Lincoln, 2016), it is important that university management give the issue due consideration. This particular finding echoes McKay and Monk (2017) finding that inadequate preparation to teach and a lack of confidence are key challenges facing ECAs and may pose some difficulties for them in handling large classes. This situation may be particularly true in Tanzania, where universities are currently operating in the midst of challenging students surges every academic year. ECAs' lack of a professional background in HE teaching may affect their teaching confidence, especially when handling large classes.
Drawing on educational debates about the quality of university teachers, ECAs isolation not only results in dilemmas in their teaching (Scager et al., 2017) but also stifles their confidence in teaching, research, and community engagement (Hemmings, 2012). In the present study, isolation was reported to be a standard phenomenon in the studied universities, perhaps as a result of ECAs' working without adequate support from senior academics. Although one could argue that ECAs could seek support on their own, this could be difficult, since they might lack the confidence needed to seek such support, due to a lack of guidelines and policies to guide them and the academic hierarchies imbedded in academia. It is clear from the literature that working as a beginner individual researcher is too isolating (Hemmings, 2012) and may cause ECAs to feel they lack support and cannot bounce ideas with colleagues from whom one might expect a friendly ear.
Although exploitation was not reported as common to all universities, it can be argued that it is a consequence of power relations between academics and a vicious circle caused by a lack of mentoring. For example, the lack of mentoring policies in universities might allow senior academics to fall into unethical practices, such as wrongly using their power and expertise to exploit ECAs academically. For example, in the present study, some ECAs in the surveyed universities claimed that senior academics might refuse to acknowledge their intellectual contributions to research projects and deny them deserved authorship credit. Academic exploitation is one of the most pressing issues among ECAs and senior academics (Gray, Thompson, Clerehan, Sheard, & Hamilton, 2008;Martin, 1986). Generally, these findings advance previous research on ECA mentoring by shedding light on the issues that confront the mentoring process in higher education. This study also contributes to a better understanding of potential strategies for effective mentorship practices in universities in Tanzania and beyond.

Implications for the next generation of academics
As part of our contribution to the HE research, we argue that effective teaching on the part of the next generation of academics requires effective mentoring programmes that focuses on ECAs' HE pedagogical aspect and research skills. Based on the findings as reported in this article, it is evident that the fate of the next generation of academics in Tanzania is not promising, especially in terms of their teaching quality should this situation persist. The lack of mentoring arrangements and mentoring guidelines may imply that ECAs, both in the early stages of their career and thereafter, will be ineffective in their teaching, research, and community engagement as a result of lacking mentoring not only in pedagogical aspects that would enable them to effectively teach but also they will have deficits in research and public engagement.
Since Tanzania does not provide HE teaching certification for doctoral students, the lack of mentoring arrangements, especially with regards to HE pedagogical aspects, implies that Tanzanian ECAs are unprepared for both the pedagogical requirements for teaching in HE and the learning psychology of adult learners. Therefore, the next generation of academics is likely to be unable to effectively teach and do research at the HE level, which may result in increased student attrition due to the poor quality of teaching. Existence of powerful neoliberal forces in HE systems associated with competition among academics and established imperative of 'publish or perish' as well as limited time to engage in mentoring could be some of the reasons for some academic's reluctance to take up mentorship responsibilities. Based on the aforesaid reasons and the present findings, this paper recommends for inclusion of mentoring activities in academic career progression and/or awards for mentoring excellence to foster a mentoring attitude among academics. As an outcome of this study, this study also recommends for universities to employ a cognitive apprenticeship model as a framework for ECAs mentoring as introduced in section 1.1.
The lack of mentoring arrangements accounts for ECAs' lack of teaching confidence, since they are neither pedagogically trained to teach nor given guidance on how to teach at the HE level; a lack of preparation to do their job leads to a lack of teaching confidence. This may imply that the next generation of ECAs may not only have low self-efficacy but also lack the necessary pedagogical teaching and research skills, which might negatively affect the quality of university education in general. Overall, the findings imply that the next generation of academics will not have adequate knowledge and experience to teach and do research at the HE level or engage with the community because they are not pedagogically mentored for teaching and other academic roles.
Additionally, the findings imply that the exploitation and isolations of ECAs may contribute to declining research output and even the existence of both professional and research misconduct among academics in HE institutions. A decline in research output could negatively affect both the quality of education at universities and university rankings as research output is a critical factor in university visibility and quality. To tackle the issue of academic misconduct, two solutions are recommended from this study: First, asserting guidelines for academic conduct and creating mechanisms through which academic misconduct can easily be reported and proper actions taken. Secondly, rewarding research collaborations which involve ECAs. This will not only reduce academic misconduct but also will foster mentoring of ECAs by senior staff.

Conclusions and recommendations
The mentoring issues found to affect ECAs in this study are evidence of a weak institutional system of preparing ECAs to engage effectively in teaching, research, and public service. The study also contributes to a better understanding of effective strategies for the reform and improvement of mentoring for the academic growth of ECAs. Finally, this research paper offers the following recommendations: The findings suggest that it is critical that universities as a whole devote sufficient effort and resources to ensure that ECA mentoring is not only common practice but a process strongly backed up by institutional policy and guidelines. Ideally, institutionalizing a policy framework will provide clear expectations for both parties to address their joint and separate needs. To this end, university management should have mechanisms to ensure that mentoring is both properly supervised and commendably executed. Doing so will motivate senior academics and help them to extend their mentoring practice even to their undergraduate students. This will extend their academic network and remind them that learning is a social process that should be built within collegial settings, in which mentoring is believed to promote scholarship of learning and teaching. It is also imperative that all beginning academics undergo an HE teaching training and certification process. Although this might be costly, it is important as it will provide ECAs with opportunities to develop a greater pedagogical understanding of teaching in the university context. Such pedagogical training is particularly important for graduates who secure jobs in universities as academics.