Supervision in graduate study as a research focus

Analysis of nearly seventy years of the formal existence of graduate degree programs in Brazil has revealed that the process of supervising dissertations and theses has undergone a strong inflection. A first indication is a succinct observation found in the report Parecer n. 977/1965 of the Conselho Federal de Educação [Federal Education Council] (CFE), according to which, upon entering a graduate program, students will be “supervised by a director of studies” (^{Almeida Júnior et al., 2005}, own translation). The change is also seen in a recommendation found in the V Plano Nacional de Pós-Graduação [5th National Graduate Studies Plan] (PNPG) of 2005 to 2010, which expresses the need for the “creation of an efficient supervision system” (Capes, 2010, own translation). The period between these two guidelines had a rich history in which factors that concern the relationship between supervisors and supervisees in the process of research, drafting, defense and socialization of the final work of the students became noticeable.

The literature to which we had access, the research conducted, and the statements we gathered make clear how the supervision of academic studies has undergone metamorphoses, in a movement that ranges from the intuitive-individual to the collective-group. There is a presumption that the work of supervision is a research field with systematizations, as well as a space open to prescriptions about “how” to supervise. We found a tension between the external inductions (related to support and evaluation, multiple requirements) and the protagonism (autonomy) in the implementation of a supervision praxis, conceived as a process that goes beyond the relationship between a supervisor and a graduate student.

The review of the literature about supervision in graduate programs in education revealed a near absence of research and production until the mid-1990s. Before this period, we found a text by ^{Castro (1978}), as a professor and supervisor in the master’s in education program of the Instituto de Estudos Avançados em Educação of the Fundação Getulio Vargas of Rio de Janeiro (IESAE/FGV/RJ); an article by ^{Alves-Mazzotti (1992}), of the Universidade Federal do Rio de Janeiro (UFRJ), and an article by ^{Haguette (1994}), of the Universidade Federal do Ceará (UFC). The texts by Castro and Haguette conclude with prescriptions or “commandments” for a good supervisor, and the finding of a lack of and thus the need to establish a type of deontology for the supervision praxis.

Since the mid-1990s, in the wake of an unprecedented expansion of degree-oriented graduate studies in Brazil, a pleiad of authors/supervisors have reflected, conducted studies, issued statements and published articles about supervision. We can highlight those that provided an empiric base and theoretical support for our studies: ^{Saviani (2000}), ^{Machado (2000}), ^{Leite Filho and Martins (2006}), ^{Viana (2008}), ^{Mazzilli (2009}), ^{Alves et al. (2012}), ^{Baibich (2014}, 2016). For Haguette (2002) and ^{Follari (2002}), supervision is an “obscure function” that has been poorly codified; an “unevaluated activity” that is little discussed, and is part of the “hidden curriculum” of graduate programs. It is also understood as a “polyphonic process” (^{Garcia & Alves, 2002}, own translation); a “poorly evaluated space” (^{Santos & Maffei, 2010}, own translation); a locus characterized as “uncharted territory” (^{Hess, 2005}, own translation); a “complex task” (^{Hockey, 1997}, own translation); and a “chaotic pedagogy” (^{Manathunga, 2010}, own translation). According to ^{Park (2007}), it is a relationship developed based on the “secret garden’ model”. Thus, we can propose affirmations that support the idea that, given the lack of explicit concern and initiatives to train supervisors, a situation has materialized in which ad hoc supervisions are exercised by ad hoc supervisors, given that there are no previsions or initiatives for their training.

Since the middle of the decade of 2000, we found that supervision and graduate studies have become research objects, even for master’s and doctoral students. This work includes collections organized by ^{Moraes et al. (2005}), and ^{Schnetzler and Oliveira (2010}). These authors, working with a focus on graduate supervision, organized a collection with articles produced by graduate students based on questions prepared by one of them. One of the doctoral students who participated in this course asked; if supervision is so important “how come it is not a required subject in all the graduate programs in Brazilian universities?” (p. 204, own translation).

For ^{Corrêa (2012}, p. 392, own translation), the issue of supervision of dissertations and theses should be the “object of study of academic research”. Corrêa emphasizes that this requires studies about the theme, not only in relation to individual researchers or graduate programs, but “in the possible articulation between them (supervisors) and their supervisees, and in the interinstitutional partnership with Research Nuclei and Groups found in the country or abroad” (2012, p. 413, own translation). Based on research about the issue, Corrêa recommends that: “graduate programs must give greater visibility to the supervision of dissertations and theses, because this pedagogical moment has deserved little institutional attention”, either from “researchers individually or in research groups” (Corrêa, 2012, p. 392, own translation). This aspect is also highlighted in a study conducted by Mazzilli about current changes in the concept of supervision and the work of supervisors, which concludes:

In terms of the specificity of the work of supervisor, although it is found that supervisors continue to be the ultimate person responsible, in the graduate programs, for the final production of the work of their advisees, supervision, as a silenced practice, is being overtaken by the collaborative culture of the programs, broadly favored by the adoption of curricular activities that allow sharing the process of developing new researchers. (2009, p. 125, own translation).

Meanwhile, Nóbrega emphasizes the need to think of supervision in the twenty-first century and asks: “who supervises the supervisor?” (2018, p. 1058, own translation). She affirms that supervisors have used their experience being supervised, “acting intuitively, as a layman”, to conduct the work of supervision, which generates a “deprofessionalization of academic supervision”. She adds that the main problem with this situation is the waste of talents of the supervisors and the supervisees, ineptitude and discouragement, which she believes “harms the academic and economic incentives to the [graduate] programs”. These considerations support Nóbrega’s advocacy of the importance of training courses for supervisors that “develop skills based on a systematization of proposals, and an exchange of experience among beginning and experienced supervisors” (p. 1062, own translation).

In relation to research outside of Brazil, we highlight work that seeks to respond to problems and propose alternatives for supervision in the post-implementation period of the Bologna process (^{Walker & Thomson, 2010}; ^{Peelo, 2011}; ^{Trafford & Leshem, 2008}; ^{Murray, 2009}). Among these studies we find authors who discuss how to deal with the specificities of graduate students; those that are interested in researching the activity and education of supervisors; and others who focus their studies on the experience of the graduate education process for both students and their supervisors.

More recently, given institutional transformations in graduate studies related to changes in the form of evaluation and financing by official agencies, in parallel to conjunctural and structural issues - rankings, for example -, various strategies begin to be discussed, such as online supervision, co-supervision, collective supervision, collaborative networks and steering of the praxis of research groups to include supervision as a pedagogic component. And with increasing emphasis, arguments and proposals arose - mainly in Anglo-Saxon countries - to treat supervision as a “pedagogy or didactics” (^{Walker & Thomson, 2010}; ^{Halse & Malfroy, 2010}; ^{Kamler & Thomson, 2014}).

In this discussion, ^{Walker and Thomson (2010}) defend the implementation of strategies for training supervisors that conceive supervision as a pedagogy or didactics that “can be taught and learned”. In addition, they point to the need to consider, when supervising, the individual and collective specificities of graduate students: those who study fulltime and those part-time, immigrant or migrant students, those linked to professional master’s and doctoral programs or to academic programs, among other aspects, because they come to constitute subcategories that require special attention. In a recent publication, ^{Kamler and Thomson (2014}), working with the notion of writing as social practice, point to the need to locate academic work in personal, institutional and cultural contexts. In this sense, they suggest that supervisors think of “writers” of a doctorate - a level of education in which there should be a focus on text - as novices who need to learn new fluencies with words as they enter the discursive practices of academic communities. This process, according to Kamler and Thomson (2014), involves learning sophisticated writing practices with specific sets of textual conventions and characteristics.

Meanwhile, ^{Halse and Malfroy (2010}) discuss the need to articulate and clearly delineate the work of doctoral supervisors, recognizing higher education to be an increasingly competitive environment, and verify the need for organized training for doctoral supervisors. In this sense, they seek to respond to this imperative by examining the question: “in the contemporary university, what do doctoral supervisors do and how can their work be conceptualized?”(p. 79, own translation). To obtain responses, they analyze statements from interviews about the life histories of doctoral supervisors of five large fields of knowledge from a large metropolitan university in Australia. The analysis conducted allowed the authors to affirm that doctoral supervision is conceived as professional work that includes five facets: an alliance of learning, mental habits, academic expertise, technê³ and contextual expertise. The authors conclude that these aspects are key to a model of education of supervisors that “offers more precise discourse, language and theory to understand and prepare to work in doctoral supervision in the contemporary university” (Halse & Malfroy, 2010, p. 91, own translation).

We found that supervision is realized in institutional spaces permeated by hierarchization, and is characterized as a relationship in which the exercise of power can be implemented in various forms and intensities, depending on the conception and experience of the supervisor. We can say that, in terms of supervision, we are in a locus or a “scientific field” (^{Bourdieu, 1974}) in which a habitus or various habitus are constructed, powers are exercised, mobilized, and reconstructed, exacerbating the complexity of what is materialized in the supervisor-supervisee relationship and what is not explained or even fully addressed in “how to” guidelines found in manuals of methodology.

We highlight these authors and their research because we understand that they point to aspects that appeared in the study that we conducted and became foundational to the research process, from the review of the literature to the empiric research.

Research methodology

This section will present and discuss the data gathered from leaders of research groups who work as professors in graduate programs in education in doctoral courses. We focus on the organization and functioning of their research groups, as well as the perceptions and statements of these leaders about the potential of the group for training supervisors to work in degree-oriented graduate programs. Our objective was to identify the actions present in the research groups in which these professors work as leaders, to reveal and characterize their research strategies in general and for supervision in particular.

The data collection tool was an online questionnaire composed of 15 closed questions (about professional work and education in graduate studies) and 3 open questions (about the organization of their research group and the relation of the group to the education of supervisors). The criteria established to define the research participants was that their graduate program in education [programas de pós-graduação em educação] (PPGE) had conducted at least one doctoral defense, that is, at least four to five years had passed since the doctoral course was created and approved in the Capes system. The survey conducted in January 2018, period of organization and preparation for issuing the questionnaire, revealed that there were 87 PPGE with doctoral courses, although 13 of them had not yet conducted their first defense. Among the 74 programs that complied with the research criteria, we identified 1,971 professors/researchers, all of whom were participants or leaders of research groups, which act in these graduate programs.

The data was collected between February and March of 2018. We had returns from 562 respondents, 28.46% of the total, who represented 100% of the programs, that is, there were respondents from all 74 programs. Upon relating the total number of professors in a program with the number of respondents, we found: in 10 PPGE, more than 50% of their professors responded to the questionnaire; in 20 programs between 30% and 50% of the researchers responded; and in another 44 programs less than 20% of the professors responded.

The group of professors/researchers who responded to the questionnaire is composed of 380 women (67.6%) and 182 men (32.4%). In terms of age, the largest group was between 51 and 55, with 136 respondents, followed by the group from 56 to 60, with 98 participants. In sum, we can say that 392 (69.7%) were between 46 and 65.

The highest number of respondents were from the state of São Paulo (117), followed by Rio Grande do Sul (94), Rio de Janeiro (47), Minas Gerais (46), Bahia (22) and Pará (21). In relation to the graduate course completed by the respondents, 179 participants had studied Pedagogy, composing the largest group. Among the other courses most cited are: History (54), Psychology (53), Philosophy (49), Language and Literature (39) and Physical Education (34). Of the total number of respondents, 56 had completed more than one undergraduate course. Most had concluded their graduate studies in the 1980s (214), followed by the 1990s (149). Nearly all of the institutions of higher education in Brazil are represented by the locations of the graduate courses of the respondents, mainly those existing in the period from 1980 and 1999. We can highlight some of them based on the number of citations: Universidade Estadual Paulista “Júlio de Mesquita Filho” (Unesp) (31), Universidade de São Paulo (USP) (27), Universidade Federal do Rio Grande do Sul (UFRGS) (20), Universidade Federal de Minas Gerais (UFMG) (17), Universidade Federal do Paraná (UFPR) (16) and Universidade Federal de Santa Catarina (UFSC) (15). Only 16 respondents indicated that they conducted their graduate studies abroad.

The data collected allow affirming that 50.8% (285) of the respondents conducted their master’s in the field of education, while the others were in different fields of the social and human sciences. The Pontifícia Universidade Católica de São Paulo (PUC-SP) (45), UFRGS (40) and the Universidade Estadual de Campinas (Unicamp) (38) accounted for most of the master’s courses, cited by 237 respondents who took these courses in the 1990s. Only five participants conducted their master’s abroad: 3 in the United States, 1 in Italy and 1 in France.

In relation to their doctorates, the majority, 358 respondents, earned it in the field of education, in the period from 2000 and 2010 (52.9%), and the universities most cited were: UFRGS (52), USP (51), PUC-SP (36) and Unicamp (36). Only 15 respondents conducted their doctorate abroad, most in Spain and France. The question about having conducted a post-doctoral fellowship abroad had the following responses: 331 (58.9%) had conducted post-doctoral studies, and 231 (41.1%) did not. Among those who did, the largest group - 146 respondents, 44.1% of the total - did their post-doctorate in Brazil in 38 different universities.

In terms of the professional work of the respondents - specifically in degree-oriented graduate studies - we found that 158 (28.1%) work in more than one program, while 404 (71.9%) work in only one graduate program. In terms of the time of work in higher education, the majority - 326 (58%) professors - have been working for 16 to 30 years at this level of teaching. In relation to the time of work in graduate studies, 144 (25.6%) respondents - the largest group - indicated that they have been working for 6 to 10 years in graduate studies.

In the responses referring to the number of supervisees attended and defenses realized, we found the following data. Firstly, in relation to the current number of master’s supervised, 255 (45.4%) respondents declare that they have 2 or 3 supervisees, while 183 (32.6%) affirmed they have 4 or 5 supervisees. The highest number declared was 11, by 2 participants. In terms of supervisions concluded, 113 (20.1%) professors said they had concluded between 10 and 20 supervisions.

Upon looking at the data about supervision at the doctoral level, we found that the largest group of responses about the current number of supervisees included two subgroupings: four or five supervisees was the response of 193 (34.4%) professors, while another 145 (25.8%) responded 2 or 3. The highest number of doctoral supervisees was 10, for 6 respondents. In relation to the number of thesis defenses conducted, 205 (36.5%) indicated between 1 and 7 defenses, while 200 (35.6%) others said they had not concluded any doctoral supervisions.

We believe that this data allows delineating aspects that characterize education and action in graduate studies of the 562 respondents to our study. They are data that allow affirming that the participation was widespread among the 74 PPGE selected, with a good representation of professors/researchers who had recently entered the doctoral courses, along with a more experienced group who have been working for a long time at this level. We found a wide variety of educational backgrounds, mainly in the fields of the human sciences, but also a small number linked to the health and engineering fields.

The relationship between the research group and training supervisors

In discussing the relationship between the research group and training of supervisors, we sought to focus the analysis on aspects that we thought could offer contributions to the issue. At first, we sought to characterize the organization and functioning of the research groups through the statements of their leaders, a professor/researcher who works in a doctoral course of a PPGE. Based on this initial data, which allowed knowing and understanding the dynamic of operation of the group, we sought to reveal its research strategies in general, and the actions and initiatives that the group participants consider to have potential for training supervisors in particular, according to the leader. In the questionnaire, these issues composed two open questions about: a) the organization and operation of the research group; and b) the potential of the group to train supervisors. The next item presents the routines and activities indicated by the respondents, and discusses their ideas about the potential of the group for training graduate study supervisors.

Organization and functioning of the research groups

Upon being questioned about the regular work of the research group, the first data, provided by 107 respondents, was related to the frequency of its meetings: 56 leaders affirm that they hold weekly meetings, 35 said that the meetings are bi-weekly, 13 said they were monthly, 2 leaders said their meetings are every two months and 1 said they were once a semester.

The activity study encounter/discussion of texts was mentioned most: 310 responses. Table 1 presents the responses most mentioned by the 562 participants in the study.

There were fewer responses that mentioned activities, such as: field research with data collection, discussion of research methodologies, encounters with researchers from outside the research group, meetings to make decisions and divide tasks (planning), collective research, and others. In relation to the theme of interest of this study - supervision - there were 36 mentions of activities of “collective supervision” in the groups; 35 mentions about “supervision” activities; and 9 references to “individual supervision”, which totaled 70 statements indicating the presence of this activity in the research group.

One leader summarized the activities that the research group participants conduct: “the routine is composed of the actions of reading, writing and discussing in the group and in small groups”. She said that this dynamic takes place in the regular meetings, usually weekly, with advisees and researchers, “to support the interlocution between research projects; deepen conceptual approaches based on shared reading and writing; and support the systematization of the individual and collective studies in written productions”.

Another researcher, upon reflecting on the regular work of the research group in which she is the leader, reported that “the Group conducts at least one study of common interest among the participating professors and tries to maintain a frequency of meetings with the objective of strengthening the theoretical field that supports the study”. According to the leader, the meetings discuss “aspects related to the research methodology, forms of entering the field, forms of data analysis, and discussions about the common themes”. In addition to these encounters, each professor-supervisor organizes, with her advisees, “small groups for systematic meetings in an effort to integrate the new and older members and so that the group takes advantage of the group discussions”. The leader explains that at some meetings, “the group conducts, as an act of solidarity, the reading and discussions of the work of colleagues to qualify the texts”. She said that this has been “an interesting exercise in the sense of realizing the criticism of the others work in a respectful way, but it is also necessary to help them to advance”; because the reader “must learn to make respectful criticism and perhaps allow the colleague to construct her arguments, reconsider, advance in the knowledge”. She emphasized that the group “conducts the exercise of being exposed”.

In relation to the question of the writing of the supervisees, we emphasize the studies by ^{Kamler and Thomson (2014}), who understand research as writing and writing as research. The authors discuss the work of the supervisor and find that it is largely anchored in assisting students with writing. In the statements collected from the research groups we found a large amount of space given to writing, whether individual or collective. In this context, ^{Wadee et al. (2017}) note that - when they conduct workshops about the work of supervision - the perspectives of the supervisees can be summarized as: wanting comments from the supervisor about their written work and to encounter availability, cordiality, solidarity and encouragement. These aspects can be attained/met in the working dynamics of the research group, according to our observation in this study. The statement of one leader described strategies the research group used to help with writing: “members of the group, organized in pairs composed of a graduate and undergraduate student, take minutes of the meetings”. According to the leader, the minutes serve as a “register of the memories of the group by means of written narratives, and to improve writing”. They are corrected and complemented by the coordinators and later socialized with the other participants. In addition, the leader added, the group produces joint studies for presentation and publication at academic events, to “give visibility to the actions conducted”.

This range of statements offered a privileged look at the functioning and organization of the research groups of the respondents to our study. They expressed a clear concern about organizing the group activities, despite the intense work routines of these professionals. We found that the leader and other researchers in the research group collectively conduct the actions that are part of their work in the graduate program: research, writing, publication, and supervision. Work that is commonly undertaken by a researcher/teacher individually, came to involve collective actions. In more organized groups - with a larger number of experienced researchers - it is possible to find the organization of actions based on a division into subprojects of studies. Within this organization, collective supervision and co-supervision gain greater space. The responses point to expressive gains with research conducted collectively.

The potential of research groups to train supervisors

Considering that a research group can also be a space for training supervisors of dissertations and theses, the intention of the open question was to obtain data about how the leader of the research group analyzed the relationship between the routine actions of the group and the opportunity to train supervisors. Although the ability of the group to train researchers seemed evident, the objective of the questionnaire was to create a space to reflect on the work of supervisors in degree-oriented graduate programs. By raising this aspect, we were able to visualize the processes of organization, administration, planning, training, and social relations established in this space for the construction of knowledge.

This question was responded by 537 research group leaders, that is, by 95.5% of all the participants. Of these, 493 (91.8%) consider the group to be a space for training supervisors, because, according to one researcher, “the personal exercise for the selection of sources, analysis, data collection, selection of methodology; dividing the research for projects of scientific initiation students . . . is an experience that allows soundly supervising new researchers”. For 28 (5.2%) respondents, the research group is a space for training of supervisors - up to a point: “we have not had a specific or planned concern for the supervision process. The exchange of experience with the longer term participants winds up helping, but it takes place without an objective or planning”. These researchers emphasized the importance of individual supervision, which appears to have more relevance outside the realm of the research group. As one of them explained: “the supervision is individual, although some of what takes place in supervision can be fulfilled in the group. But as an individual moment of the supervisor with the supervisee, it involves particular affective and cognitive aspects”. A very small group - 10 (1.8%) researchers - responded negatively about the possibility for the research group to also be a space for training future supervisors. They argued that this is not an activity inherent to the research group, because, as one of the respondents said: “I understand that the research group has its [own] nature, which should not be confused or identified with the specific roles of the relationship between the supervisor and the supervisee”. He said there is a risk of discharacterizing the objectives of a research group.

The responses obtained involve issues that contemplate the discussion about training supervisors, but also advance to other aspects that involve what it means to be the leader of a research group (to create strategies for functioning and adhesion/participation) and to be a researcher professor who works in a graduate program (in production and supervision). Many of the statements emphasize the importance of belonging to the research group as a possibility for professional and academic growth for its participants. The statements are very positive about the potential of the group to consolidate the process of interaction, experience with research, collective work, and training of supervisors. In Table 2 we highlight the responses that were most mentioned among the participants, grouped by what we consider to be the organizational elements of the statements.

The statements highlight the understanding that the research group is a “laboratory of training to work as a supervisor”, as well as a space of “academic partnership”. While they highlight the importance of collective research and supervision, the researchers point to difficulties encountered when conducting collective research projects, which can require that they abdicate the role of the researcher/supervisor as one solely responsible for defining the theoretical-methodological knowledge valid for all group members. For many, it is also a challenge to share this function with their supervisees.

It is important to highlight the concern that many respondents had to express their opinion and reflection with clear details. This was present in a significant number of statements about the actions they take in their research group and that they consider to be important and ripe with possibilities for training supervisors. The related processes frequently mentioned include: socialization and debates about individual and collective projects (317 mentions); collective debate about research processes (142 mentions); socialization of theoretical sources (130 mentions); sharing of ideas and experiences (94 mentions); collaboration and support among peers (86 mentions); sharing of written work (55 mentions); sharing of difficulties (46 mentions); collective space for supervision (46 mentions); and joint production (40 mentions).

We found that the research group functions as a space that provides support, as “mitigator of the problems of the graduate student and supervisor”, in which they find a collaborative network of researchers and scholars, and more than anything else, a space for training, which is qualified through the activities and in a collective manner.

Final considerations

This study allowed deeper exploration of the current bibliography about the work of supervision in degree graduate programs. It also enabled the identification of the paths and fields of knowledge of the researchers who discuss and research the issue. We observed that supervision is discussed by researchers from various fields (highlighted by the fields of health, psychology, and administration) involving very similar focuses: relationship between supervisors and supervisees, difficulty of graduate students to successfully conclude their education, lack of institutional support, demands for evaluation made by the graduate program and others. The problems of those who work and study in graduate programs are not limited to specific fields of knowledge. The statements of the supervisors, found in the studies that served as a basis for this research, pointed to very similar main challenges: difficulty in the relationship with graduate students, lack of preparation of the supervisees, excess of institutional demands, loss of status associated to work at this level of education, difficulty in reconciling work in graduate studies with activities of teaching and extension at the graduate level, and others.

In this plunge into bibliographic production about graduate studies and, more specifically, about the work of supervision, we found a small number of researchers in the field of education interested in the theme. Among them, the priority focuses of attention are analyses of public policies - related to support and evaluation - for graduate studies. The issue of supervision occupies less space among the projects of the researchers dedicated to higher education in the research groups.

It is among the publications in a foreign language that we found the highest concentration of researchers who discuss the work of supervision, and in doing so mainly analyze the space of education of researchers: doctoral programs. Most of the studies conducted about the issue directly consider two aspects about doctoral students and their supervisors: the collection of data that allow understanding the dynamic of this relationship and the proposition of referrals and guidelines that can improve the relationship so that degrees can be earned through good processes.

Based on the particular study that we conducted with leaders of research groups in the field of education, we can list aspects that we believe can help reflect on how they organize their academic life - as teachers, researchers, and supervisors - and all the activities stemming from each of these work fronts. The statements of the 562 professor-supervisors who responded to the questionnaire, who belong to 74 PPGE, helped us to delineate a representative chart - and data for identification - about the functioning of the research groups, their potential for training supervisors and the indicators and actions that can support this training. To summarize, the most relevant aspects include:

To summarize, in the absence of specific institutional spaces/times for the training of supervisors, the actions of and in research groups and networks, considering their organization and operations, serve as loci for teaching and learning the métier of supervision, according to the statements of the respondents. Thus, support for models developed by supervisors, as well as exchanges in the realm of research groups, appear as aspects that offer very strong possibilities for training new supervisors.