A Research Supervision Framework for Quality and Scalability

Paul Stynes

and Pramod Pathak

School of Computing, National College of Ireland, Ireland

Keywords: Research Supervision, Taught Masters Programmes, Timetabled Group Supervision, Co-supervision.

Abstract: Academic staff provide research supervision based on the one on one approach or the apprenticeship model.

Current student recruitment policies are enrolling larger numbers of students on taught master’s programmes.

The current research supervision approach is not sustainable with growing numbers of students and

conducting research supervision that is scalable is a challenge. Increasing the number of supervisors leads to

difficulties with the consistency in the quality of the supervision. This research proposes a research

supervision framework that scales with increasing numbers of re-search students and ensures consistency in

the quality of research supervision among faculty. The framework combines teaching practices, timetabled

group supervision, co supervision, coaching and scaffolding. The research supervision framework was applied

in timetabled group research supervision sessions in May to August2020 with 15 students, September 2020

to January 2021 with 10students and May to August 2021 with 12 students. Results demonstrate an increase

in the quality of research as demonstrated by the publication of 4 book chapters, 3 peer reviewed international

conference papers and 3 invention disclosures. These publications occur during a period of growth in student

numbers by approximately 1000% from 2012 to 2020. This research is of interest to both Deans and faculty.

Deans will gain insight in how to ensure quality of supervision with growing student numbers on taught

master’s programmes. Faculty will gain insight in how to effectively supervise students in order to increase

their academic publications using alternative supervision approaches.

1 INTRODUCTION

Research supervision typically takes place on a face-

to-face basis in individual sessions. Research

supervision may take place at PhD level, Masters by

Research or through a taught Masters programme.

Although research is available on doctoral research

supervision, a gap in the literature on master’s

research dissertation supervision has been

highlighted by Macfadyen, et al., (2019), Cornelius &

Nicol (2015) and Drennan & Clarke, (2009). At

taught and research Masters level, both approaches

have a research dissertation as a final task to obtain

the qualification. A further difficulty is that a taught

masters would have a lower period of time based on

credits to spend on the research dissertation. The

contribution of the research dissertation on a taught

master’s programme typically varies between 25-30

credits. The duration of the dissertation on a taught

https://orcid.org/0000-0002-4725-5698

https://orcid.org/0000-0001-5631-2298

masters is typically 3 months full-time or 6 months

part-time in the last phase of the master’s programme.

Taught Masters programmes characteristically

have larger numbers of students studying on the

program. As student numbers grow the traditional

research supervision model known as the

apprenticeship model with one on one supervision is

not sustainable. The supervision of master’s students

is very difficult to define as there are many variables

based on the programme discipline, duration and

diversity (Pilcher, 2011). The aim of this research is

to investigate to what extent a research supervision

framework scales with increasing numbers of

research students and ensures consistency in the

quality of research supervision among faculty. The

major contribution of this research is a framework

that combines teaching practices, timetabled group

supervision, co supervision, scaffolding and

coaching. This research is of interest to both Deans

and faculty. Deans will gain insight in how to ensure

quality of supervision with growing student numbers

Stynes, P. and Pathak, P.

A Research Supervision Framework for Quality and Scalability.

DOI: 10.5220/0011093900003182

In Proceedings of the 14th International Conference on Computer Supported Education (CSEDU 2022) - Volume 2, pages 395-403

ISBN: 978-989-758-562-3; ISSN: 2184-5026

395

on taught masters programmes. Faculty will gain

insight in how to effectively supervise students in

order to increase their academic publications using

alternative supervision approaches.

2 RELATED WORK

This related work is organised around the

apprenticeship model of supervision, group

supervision, co-supervision, teaching practice and

scaffolding. The role of the supervisor as an assessor

is also re-viewed as this is a factor that has workload

implications for master’s supervision as applied to

group supervision approaches.

The British single supervisor model uses an

apprenticeship model of training whereby a student

registers to study on an independent piece of original

research under the supervision and guidance of an

experienced academic researcher who advises them

on the conduct and publication of their research.

Higher education institutes are under increased

pressure from growing national and international

student numbers enrolling on postgraduate

programmes (Bitzer & Albertyn, 2011; Engebretson,

et al., 2008). As a result, there is increased pressure

on the supervisor workload (Bitzer & Albertyn, 2011;

Usher, 2002). The increasing workload of supervisors

of postgraduate students threatens the quality of

research (Deuchar, 2008). The apprenticeship model

is seen as being inappropriate to meet the needs for

supervising students and alternative supervision

approaches involving combining supervision groups,

student colloquia and individual supervision are

proposed by Dysthe, Samara, & Westrheim, (2006).

Bitzer & Albertyn (2011) present alternative

supervision approaches such as group supervision

and the team/panel approach to relieve the burden of

increased supervisor workload arising out of

increased student numbers. Groups may be comprised

of students with similar levels of experience in the

research process or mixed supervisor-and-student

groups (Samara, 2006). Group supervision has the

benefit of students producing higher quality

dissertations particularly when students interact with

their peers and supervisor (Lovitts, 2008). The

structure of the group supervision process reported in

Malfroy (2005) included a one-to-one follow-up

session with the supervisor after a group session to

build on the intellectual discussion and to help set a

new direction.

Supervisory Teams can involve a co-supervisor

that participates occasionally (Nulty, Kiley, &

Meyers, 2009) which can assist with the continuity of

supervision or they can bring additional expertise to

strengthen the supervision. Team approaches provide

a more holistic approach to problem solving, less

interpersonal issues, and enhanced quality of research

(Lee, 2009).

Dysthe, Samara, & Westrheim (2006) propose

that a supervision approach should combine

supervision groups, student peer groups and

individual supervision to remove the dependency on

one supervisor interaction and to manage the

academic workload. Macfadyen et al. (2019)

indicates that a supervision approach should prepare

for growing numbers of master’s students and be able

to adapt to individual student’s needs.

Marnewick (2020) proposed a supervision

approach to support the learning process of master’s

students during research supervision. The structured

approach includes self-learning, peer learning,

individual supervision, and existing teaching

practices. Teaching practices can be used by less

experienced supervisors to structure students’

supervision.

Manathunga (2005) indicates that experienced

supervisors use scaffolding of activities through the

research process. Dysthe, Samara, & Westrheim

(2006) and Crossouard (2008) indicate the

importance of having guidelines to ensure that the

supervisory team fulfil their roles and responsibilities.

Another issue that affects research supervision is

the role of the supervisor as the assessor of the

research, with a second independent assessor (de

Kleijn, Mainhard, Meijer, Pilot, & Brekelmans, 2012;

Cornelius and Nicol 2015) or it is done by two

independent assessors.

In conclusion, growing student numbers

(Engebretson et al. 2008; Bitzer & Albertyn, 2011)

and maintaining the quality of the research

supervision (Deuchar, 2008) are challenges faced by

institutes of higher education. Alternative forms of

supervision such as group supervision and co-

supervision, (Bitzer & Albertyn, 2011) and structured

supervision that includes self-learning, peer learning,

individual supervision, and existing teaching

practices Marnewick (2020), are shown to reduce the

supervisor’s workload and lead to students producing

higher quality dissertations (Lovitts, 2008).

Scaffolding of activities and research supervision

guidelines support inexperienced supervisors and

assist the supervisory team with clear roles and

responsibilities to create a supportive research

environment for the students.

CSEDU 2022 - 14th International Conference on Computer Supported Education

396

3 METHODOLOGY

This research investigates the challenges around

growing student numbers and maintaining the quality

of the research supervision.

The environment in which the framework was

developed was in a School of Computing at an Irish

institute of higher education. Students attended taught

master’s programmes in the area of Data Analytics.

The students in full time were diverse and

predominantly international. The taught masters

programme consisted of taught modules and a

research theme. The research theme consisted of a

Research in Computing module of 5 credits in

semester 2 and a Research Project of 25 credits in

semester 3. The research theme represents 33% of the

qualification. The students would typically engage in

applied research. Supervision was based on the

apprenticeship model involving one on one

supervision. We developed the framework over

several years from 2017 to date following an action

research approach. Action research seeks

transformative change through the simultaneous

process of taking action and doing research, which are

linked together by critical reflection (Lewin, 1946).

The framework was applied to the research project in

the MSc in Data Analytics We applied the research

supervision framework in timetabled group research

supervision sessions in

 May-August 2020 to 15 students,

 September 2020-January 2021 to 10 students

 May – August 2021 to 12 students

Students first complete a research proposal that

consists of an abstract, literature review and proposed

methodology as part of research in computing in

semester 2. The students are assigned to group

supervision as part of the research project.

The delivery of the research project in full-time

over a 12-week period is as follows: -

In week 0, I would get the students Research

Project Proposal from Research in Computing and the

research paper that most influences their work from

the state of the art. I would provide detailed

comments on Abstract, Introduction, Literature

Review and Methodology in order to bring the

sections up to a standard that is publishable. I would

also ensure that the influential research paper has

been published recently in a reputable journal or

Scopus linked conference.

In week 1, I would go through the slides “Module

Overview, Research Handbook, and slide decks on

writing an Abstract, Introduction and literature

review. I encourage the students to identify a quality

conference that they would like to publish in with a

focus on Scopus linked international conferences.

In week 2, I go through slides on Methodology,

Design and Implementation. On a weekly basis, I

check the students’ progress in terms of what they

completed in the last week. I set goals with the

students that they have to complete in the following

week. Week 2 goals involve starting to scope out the

aim of the experiments that they will complete.

Where the first experiment is to replicate the state of

the art from the paper that most influences their

research. I also focus on the environment and

implementation of experiment 1.

In week 3, I help the student to sharpen their

research question in the problem domain and focus on

goals around the implementation of experiment 1.

In week 4 I flesh out each experiment that will be

documented in the results section with signposting

around "The aim of this experiment is to …".

In week 5, I go through slides on Results. I also

focus on completing experiment 1 and updating the

report on results.

In week 6 I go through the slides on Discussion

and Conclusion. I focus on getting the students to

update their conference style report with a discussion

of experiment 1. I also focus on getting the students

to implement experiment 2 based on making one

change to experiment 1.

In week 7 the co-supervisor joins the meeting. The

co-supervisor meets with each student and discusses

progress.

In weeks 8, 9, 10 and 11 we focus on completing

each experiment and making one change for the next

experiment.

In week 12, we check that the conclusion restates

the research question, objectives and the work done.

We check how successful student has been in

answering the research question and achieving the

objectives, that the key findings are restated, and any

proposals for future work or potential for

commercialisation are identified.

After the examination of the master’s research

project, we work with the student to refine the

research paper and submit to a Scopus linked

international conference. If the research has

commercial potential we submit an invention

disclosure form.

The research output quality is evaluated based on

publications and invention disclosures that are

recognised at a National and International level.

Publications are in the form of Book Chapters and

Scopus linked conference papers.

A Research Supervision Framework for Quality and Scalability

397

4 RESEARCH SUPERVISION

FRAMEWORK

The research supervision framework combines

teaching practices in section 4.1, timetabled group

supervision in section 4.2, co supervision in section

4.3, coaching in section 4.4 and scaffolding in section

4.5.

4.1 Teaching Practice

Teaching practices was applied to the research project

by allocating 5 of the 30 credits into semester 2 as part

of a module Research in Computing that consisted of

1-hour lecture and 2 hours tutorial. The Research in

Computing module introduces students to the

requirements of academic writing, including carrying

out a review of the state of the art in their chosen

domain, how to create a research proposal, the

importance of reproducible research, scientific

writing including citing and referencing, and the

presentation of results. 25 credits are allocated to the

Research Project module in semester 3. Students have

to submit a portfolio that consists of a research paper

style report, an artefact/product, a user configuration

manual and a presentation to be defended in a viva.

The traditional dissertation of 10,000 to 15,000 words

is replaced with a research paper style report shall

comprise up to 20 pages of between 4,000 to 6,000

words. The report describes the individual research

and production of an ICT solution. The format of the

paper follows a conference/journal style with

abstract, introduction, related work, methodology,

design, implementation, evaluation, conclusion and

references. The motivation for this change was to

reduce faculty workload in assisting students with

publishing their dissertation in an international

conference.

4.2 Timetabled Group Supervision

Timetabled group supervision is a structured activity

based on assigning 10 to 12 students from the same

thematic area to a scheduled class with a research

supervisor. The class is scheduled in a block of 4

hours. The first hour is a group session where the

supervisor goes through a slide deck on the research

process covering aspects of the research paper style

report in a timely fashion to the sections of the report

that students are working on. The remaining three

hours are based on a one-to-one follow-up session

between the supervisor and students to build on the

intellectual discussion and to advise the student on the

research.

4.3 Co-supervision

Co-supervision was applied to research groups to

ensure continuity of the supervision during the

summer holidays. The co-supervisor participates

occasionally and provides very different but

complementary support. They usually attend half way

through the research process. The students have to

give an update on their research progress, issues they

are encountering and the next steps. Co-supervision

is also applied to new full-time faculty as a form of

direct coaching the goal of which is to go through the

research supervision framework and create a

publication in a peer reviewed Scopus indexed

international conference/journal. This co-supervisor

participates for the full supervision.

4.4 Coaching

Coaching is codified in the form of weekly supervisor

guidelines that highlight the responsibilities that a

supervisor has in a temporal form for example before

meeting the student for the first time in the research

project the supervisor should critique the research

proposal from Research in Computing. The critique

should align the proposal to the sign posting in the

template and provide the student with comments on

the abstract, introduction, literature review, and

methodology.

4.5 Scaffolding

Scaffolding is in the form of a slide deck and a

template with signposting of the research vernacular.

The slide deck was created to provide consistent

instruction to the students independent of supervisor

experience. In addition, tips are provided to support

students with writing sections of the research paper

style report. The slide deck consisted of separate

slides on the abstract, introduction, literature review,

methodology, design, implementation, results and

conclusion. The template is based on a published

paper in an international peer reviewed Scopus

indexed international conference. The template

highlights signposting of the research vernacular and

is based on Oluwatobi, Murphy, Pathak, & Stynes,

(2021).

The template guides the student in creating a title

around 3 aspects namely the novel contribution,

classification of that contribution and the story. An

example from the template where the contribution is

CSEDU 2022 - 14th International Conference on Computer Supported Education

398

“An on-device Deep Learning”; the classification is a

“Framework” and the story is “to Encourage the

Recycling of Waste”.

The slides on the abstract discusses the structure

of the abstract in terms of background, objectives,

methodology, results and findings. The first sentence

in the abstract of the template discusses the story from

the title to provide context of the research. The second

sentence emphasises wording such as use of

“challenge” to bring out the research problem. The 3

sentence articulates what the research is proposing

and that it is aligned to the title of the research with

phrasing such as “This research proposes an …

framework (or model and so on) to …”. The fourth

sentence further describes the classification of the

research with phrasing around “The proposed

framework (or model and so on) combines…”. The

fifth sentence describes the methodology. The sixth

sentence starts with wording “Results …” to ensure

the student describes the metrics applied and results

achieved. The final sentence uses phrasing “This

research shows promise for …” to ensure the student

describes the benefit of this research to a stakeholder

that is interested in this research.

The slides on the introduction discuss the

background to the research, the motivation for the

research, the variables or factors that affect the

outcome of the research, research question, and

contribution. The template reinforces the slides with

phrasing “The aim of this research is to …” and “The

major contribution of this research is a novel …”.

The slides on the related work provide a guideline

for writing a critique, writing style and referencing.

There is a focus on identifying future work from a

journal or Scopus linked conference paper in order to

ensure that students work on a novel contribution that

is publishable. If a research paper is important to their

research that they will use or improve on, then they

should describe it in a paragraph in the literature

review. The paragraph should describe the purpose of

the research paper, methods used, main results and

strengths or limitations. Strengths are good practice

and something the student should aim to emulate in

their research. Limitations provide an opportunity for

students to improve on or extend that work. The

concluding paragraph of the literature review sums up

the strengths and weaknesses of the related work and

by assessing its contribution to the advancement of

knowledge, theory, or practice. The template provides

guidance in the first paragraph students should start

with a sentence that is directly related to the title of

their research. Phrasing for the concluding paragraph

to ensure students focus on summing up the strengths

and weakness “In conclusion, the state of the art

indicates …”.

The slides on methodology emphasise that

students experiment need to be verifiable by other

researchers, so that they can review the results by

replicating the experiment and guaranteeing the

validity. The slides describe the sections of a

methodology based on the steps followed in the

research; describing the materials and equipment

used in the research; explaining how the samples

were gathered, any randomization techniques and

how the samples were prepared such as cleaned and

transformed; explaining how the measurements were

made and what calculations were performed upon the

raw data; and describing the statistical techniques

used upon the data. The template provides phrasing

such as “The research methodology consists of five

steps namely … as shown in Fig. 1.”, “The first step,

…”, “The second step, …” and so on. Students must

ensure the steps are labelled in the figure and they

should name the step for example “The first step,

Data Gathering involves combining six waste image

datasets”.

The slides on design discuss presenting the

techniques and/or architecture that underlie the

implementation. The template provides phrasing

where the students must discuss the name of their

contribution from the title of their research paper and

the main architectural components of their design in

the following sentence “The … framework (or model

and so on) architecture … as shown in Fig. 2.”. The

design section is split into a description of each

component. The second sentence in the template

ensures the students discuss all components in the

diagram starting with phrasing “The components of

the …”.

The slides on implementation describe the outputs

produced such as transformed data, code written,

models developed, questionnaires administered. The

description should also include what tools and

languages you used to produce the outputs. The first

sentence in the template ensures students focus on

implementation with phrasing “The … framework (or

model and so on) was implemented …”. For

publications a screen shot of the graphical user

interface works well in this section.

For slides on results, discusses the presenting of

only the most relevant results using visual techniques

such as graphs, figures and tables. Presenting an in-

depth and rigours analysis of the results and how they

relate to answering the research question and the

current state of the art. Students are requested to

replicate the state-of-the-art as the first experiment.

This is to ensure students really understand what a

A Research Supervision Framework for Quality and Scalability

399

research methodology is in terms of replicating a

study and they really understand the state-of-the-art

approach. In addition, this provides an opportunity for

students to be able to discuss the results they get in

further experiments as a comparison to the state of the

art. The second experiment is based on making one

change to the first experiment. This may involve

implementing the future work identified in the state-

of-the-art research paper. A rich discussion can take

place in terms of the results of experiment 1 and how

they relate to previous research from the state of the

art. Experiment 2 may involve

changing/optimising/enhancing the framework (or

model and so on) and seeing if the results improve.

By following this approach there is clear and

systematic evidence of novelty and contribution to the

state of the art which is publishable in a conference or

journal. The template for the results starts each

experiment to ensure students are clear on the

experiment with phrasing “The aim of this

experiment is to …”. To ensure that students discuss

the meaning of the results the template uses phrasing

“This result indicates that …”. The template

emphasises that students should describe the

diagrams and tables in detail. To make clear the most

relevant results the template uses phrasing “The

results show promise for …”.

The slides on the conclusion discuss the

conclusion in terms of restating the research question,

your objectives and the work done; stating how

successful the student has been in answering the

research question and achieving the objectives;

restating the key findings and finally describing any

proposals for future work or potential for

commercialisation. The template provides phrasing

that reemphasises the discussion in the slides with

phrasing “The aim of this research was to …”. “This

research proposes an … framework (or model and so

on) that ...”. “Results demonstrate that …”. “This

research can potentially enhance …”. This work can

be improved by …”. “This research benefits …”.

5 RESULTS

The research supervision framework was designed to

scale the supervision based on handling the growing

number of students registering on taught MSc

programmes. In 2012, 69 students were enrolled.

Enrolments reached a peak in 2019 with 814 students,

representing a growth of approximately 1000% since

2012.

From 2012 to 2017 faculty provided one on one

research supervision in the research project. An

observation was that there was a break in the

continuity of supervision when faculty went on

summer holidays. MSc surgery sessions were

provided to ensure continuity of supervision. The

MSc Surgery involved one academic checking all

postgraduate students’ progress. Progress was

reflected in the student’s blog, documenting the

chapters they have updated, the academic answering

research questions and identifying work that needs to

be completed.

Table 1 shows the interventions that were made in

creating the research supervision framework.

Table 1: Interventions in the Research Project.

Year Intervention

2017 -

2018

Group Supervision (6 students) and One on

One Supervision. MSc Surgery Session.

2018-

2019

Scaffolding in the form of a slide deck.

Timetabled group research supervision (15

students) and Co-supervision.

2019-

2020

Scaffolding in the form of a Slide deck.

Timetabled group research supervision (15

students) and Co-supervision.

2020-

2021

Coaching in the form of weekly supervisor

guidelines. Scaffolding in the form of a Slide

deck and template. Timetabled group

research supervision (10-12 students) and

Co-supervision.

Difficulties in recruitment of faculty to supervise

based on one to one supervision indicated that this

approach of scaling up research supervision was not

sustainable. A model of group supervision of up to 6

students in a thematic area with an increased focus on

Associate Faculty supervision was trialled to one

group of data analytics students in September 2017.

Associate faculty are part time faculty and would

predominantly have a PhD. An observation was that

student attendance during the research project was

problematic. Timetabled group research supervision

was implemented in 2018 in order to address the issue

of attendance. Group supervision increased from 6 to

15 students in order to address growing student

numbers. However, the supervisory model was that

supervisors would have to examine their students and

second mark an equal number of students over a two-

week period. A supervisor would have 30 students to

examine, comprising of 15 students that were

supervised and 15 students that were 2

marked. At

2 hours per examination this would take around 60

hours over a two-week period. As associate faculty

are part time and may have a full-time job, this could

add an additional 30 hours per week for two weeks

leading to workload issues. The number of students

CSEDU 2022 - 14th International Conference on Computer Supported Education

400

assigned to timetabled group supervision was reduced

to 10 students in order to address the workload issue.

Whilst 10 students in a timetabled group supervision

is the right balance, this number could increase if

supervisors are not part of marking or they did not

have to complete second marking of other student’s

research style papers.

MSc Surgery sessions were poorly attended and

as such co-supervision was introduced to replace the

MSc surgery sessions.

We introduced scaffolding of activities around a

slide deck, template and research supervision

guidelines in order to create a supportive research

environment for the students.

Table 2: MSc research output quality recognised at National

and International level.

Date

Numberof

Students

Supervised ResearchOutput

May‐20to

Aug‐20

15

2BookChapter‐Lecture

NotesinComputerScience

Springer(Kumar,Pathak,&

Stynes,2020)and

SciTePress(Palani,Stynes,

&Pathak,2021)





3InternationalConference

Papers(Wagh,Pathak,

Stynes,&Nardin,2020),

(Kanhere,Sahni,Stynes,&

Pathak,2021),(Agughalam,

Pathak,&Stynes,2021)

  1InventionDisclosure

Sep‐20to

Jan‐21

10

1BookChapter‐Lecture

NotesinComputerScience

Springer(Ekundayo,

Murphy,Pathak,&Stynes,

2022)

  2InventionDisclosures

May‐21to

Aug‐21

12

1BookChapter–European

AllianceforInnovation

(EAI),EuropeanUnion

DigitalLibrary(Padalkar,

Pathak,&Stynes,2021).





Submitted‐Awaiting

Acceptance1Journal

 

Submitted‐Awaiting

Acceptance1International

ConferencePaper

Inpreparation1

InternationalConference

Papers

We applied the research supervision framework in

3 timetabled group research supervision sessions in

May-August 2020 to 15 students, September 2020-

January 2021 to 10 students and May – August 2021

to 12 students as described in Table 2. The 3 groups

of students were supervised and co-supervised by the

authors.

Results show promise with an increase in the

quality of research as demonstrated by increased

research publications and invention disclosures at

National and International level. We supervised 37

students on a taught master’s programme in

timetabled group supervision with co-supervision.

The research output of this supervision consists of

2 book chapters as part of lecture notes in computer

science with Springer, 1 book chapter with

SciTePress, 1 book chapter as part of the European

Alliance for Innovation and stored in the European

Union Digital Library, 3 peer reviewed international

conference papers and 4 invention disclosures. At the

time of writing one paper has been submitted to a

journal awaiting acceptance, one paper has been

submitted to an international conference and 1

conference paper is in preparation. This quality

research output as a result of the Research

Supervision Framework occurs during a period of

growth in student numbers by approximately 1000%

from 2012 to 2020.

6 CONCLUSION

The aim of this research is to investigate to what

extent a research supervision framework scales with

increasing numbers of research students and ensures

consistency in the quality of research supervision

among faculty. The framework combines teaching

practices, timetabled group supervision, co

supervision, scaffolding and coaching.

Results demonstrate that students we supervised

in a small group with co-supervision have been

successful at a national and international level in

writing 4 book chapters as part of lecture notes in

computer science with springer and SciTePress and

the European Alliance for Innovation, 3 peer

reviewed international conference papers and 3

invention disclosures. The framework has scaled with

growing student numbers during a period of growth

of approximately 1000% from 2012 to 2020. Pass

rates have also increased from 29% in 2012 to 88% in

2020.

Future work could investigate the effective use of

student peer groups to reduce the work load of the

supervisor. In addition, future work could explore the

A Research Supervision Framework for Quality and Scalability

401

role of a second independent assessor being an active

second supervisor in order to increase the likelihood

of publications at an international level.

REFERENCES

Agughalam, D., Pathak, P., & Stynes, P. (2021).

Bidirectional LSTM approach to image captioning with

scene features. Thirteenth International Conference on

Digital Image Processing (ICDIP 2021), 11878, pp.

81–88. doi:10.1117/12.2600465

Bitzer, E. M., & Albertyn, R. M. (2011). Alternative

approaches to postgraduate supervision: A planning

tool to facilitate supervisory processes. South African

Journal of Higher Education, 25(5), 875-888.

doi:10.33130/AJCT.2021v07 i01.019

Cornelius, S., & Nicol, S. (2015). Understanding the Needs

of Masters Dissertation Supervisors: Supporting

Students in Professional Contexts. Journal of

Perspectives in Applied Academic Practice, 4(1), 2-12.

doi:10.14297/jpaap.v4i1.161

Crossouard, B. (2008, March 1). Developing alternative

models of doctoral supervision with online formative

assessment. Studies in Continuing Education, 30(1),

51-67. doi:10.1080/01580370701841549

de Kleijn, R. A., Mainhard, M. T., Meijer, P. C., Pilot, A.,

& Brekelmans, M. (2012, December 1). Master's thesis

supervision: relations between perceptions of the

supervisor–student relationship, final grade, perceived

supervisor contribution to learning and student

satisfaction. Studies in Higher Education, 37(8), 925-

939. doi:10.1080/03075079.2011.556717

Deuchar, R. (2008, August 1). Facilitator, director or

critical friend?: contradiction and congruence in

doctoral supervision styles. Teaching in Higher

Education, 13(4), 489-500. doi:10.1080/135

62510802193905

Drennan, J., & Clarke, M. (2009, August 1). Coursework

master’s programmes: the student’s experience of

research and research supervision. Studies in Higher

Education, 483-500. doi:10.1080/03075070802597150

Dysthe, O., Samara, A., & Westrheim, K. (2006, June 1).

Multivoiced supervision of Master’s students: a case

study of alternative supervision practices in higher

education. Studies in Higher Education, 31(3), 299-

318. doi:10.1080/0307507060068 0562

Ekundayo, O., Murphy, L., Pathak, P., & Stynes, P. (2022).

An On-Device Deep Learning Framework to

Encourage the Recycling of Waste. In Intelligent

Systems and Applications, Lecture Notes in Networks

and Systems (pp. 405–417). Cham: Springer

International Publishing, ISBN: 978-3-030-82199-9.

doi:10.1007/978-3-030-82199-9_26

Engebretson, K., Smith, K., McLaughlin, D., Seibold, C.,

Terrett, G., & Ryan, E. (2008, February 1). The

changing reality of research education in Australia and

implications for supervision: a review of the literature.

Teaching in Higher Education, 13(1), 1-15.

doi:10.1080/13562510 701792112

Kanhere, S., Sahni, A., Stynes, P., & Pathak, P. (2021).

Clustering Based Approach to Enhance Association

Rule Mining. 2021 28th Conference of Open

Innovations Association (FRUCT), (pp. 142–150).

doi:10.23919/FRUCT50888.2021.93 47577

Kumar, A., Pathak, P., & Stynes, P. (2020). A Transfer

Learning Approach to Classify the Brain Age from

MRI Images. In L. a. Bellatreche, Big Data Analytics

(Lecture Notes in Computer Science ed., pp. 103–112).

Cham: Springer International Publishing.

doi:10.1007/978-3-030-66665-1_8

Lee, A. (2009, July 1). Some implications of European

initiatives for doctoral supervision. Political Science.

Lewin, K. (1946). Action Research and Minority Problems.

Journal of Social Issues, 2(4), 34-46.

doi:10.1111/j.1540-4560.1946.tb02295.x

Lovitts, B. E. (2008). The Transition to Independent

Research: Who Makes It, Who Doesn't, and Why. The

Journal of Higher Education, 79(3), 296-325.

doi:10.1353/jhe.0.0006

Macfadyen, A., English, C., Kelleher, M., Coates, M.,

Cameron, C., & Gibson, V. (2019, July 29). ‘Am I

doing it right?’ Conceptualising the practice of

supervising master’s dissertation students. Higher

Education Research & Development, 38(5), 985-1000.

doi:10.1080/07294360.2019.1597024

Malfroy, J. (2005, May). Doctoral supervision, workplace

research and changing pedagogic practices. Higher

Education Research & Development, 24(2), 165-178.

doi:10.1080/072 94360500062961

Manathunga, C. (2005, May 1). The development of

research supervision: “Turning the light on a private

space”. International Journal for Academic

Development, 10(1), 17-30.

doi:10.1080/13601440500099977

Marnewick, A. J. (2020). A supervision approach to

facilitate learning during the master’s research journey.

Teaching in Higher Education, 1-16.

doi:10.1080/13562517.2020.1811223

Nulty, D., Kiley, M., & Meyers, N. (2009, December 1).

Promoting and recognising excellence in the

supervision of research students: an evidence‐based

framework. Assessment & Evaluation in Higher

Education, 34(6), 693-707.

doi:10.1080/02602930802474193

Oluwatobi, E., Murphy, L., Pathak, P., & Stynes, P. (2021).

An On-Device Deep Learning Framework to

Encourage the Recycling of Waste. In Intelligent

Systems and Applications - Proceedings of the 2021

Intelligent Systems Conference (IntelliSys) Volume 3

(Vol. 296, pp. 405-417). Springer Nature Switzerland

AG. doi:978-3-030-82199-9

Padalkar, A., Pathak, P., & Stynes, P. (2021). An Object

Detection and Scaling Model for Plastic Waste Sorting.

In M. Z. Philipp Wicke (Ed.), Proceedings of the 1st

International Conference on AI for People: Towards

Sustainable AI, CAIP 2021. Bologna, Italy: EAI, ISBN:

CSEDU 2022 - 14th International Conference on Computer Supported Education

402

978-1-63190-326-7, ISSN 2593-7642 Series CORE.

doi:10.4108/eai.20-11-2021.2314204

Palani, K., Stynes, P., & Pathak, P. (2021). Clustering

Techniques to Identify Low-engagement Student

Levels. In Proceedings of the 13th International

Conference on Computer Supported Education (Vol. 2,

pp. 248–257). CSEDU, ISBN 978-989-758-502-9;

ISSN 2184-5026, SCITEPRESS.

doi:10.5220/0010456802480257

Pilcher, N. (2011, February 1). The UK postgraduate

Masters dissertation: an ‘elusive chameleon’? Teaching

in Higher Education, 16(1), 29-40.

doi:10.1080/13562517.2011.530752

Samara, A. (2006, May). Group Supervision in Graduate

Education: A Process of Supervision Skill

Development and Text Improvement. Higher

Education Research and Development, 115-129.

Usher, R. (2002, July 1). A Diversity Of Doctorates: Fitness

for the knowledge economy? Higher Education

Research & Development, 143-153.

doi:10.1080/07294360220144060

Wagh, V. K., Pathak, P., Stynes, P., & Nardin, L. G. (2020).

An Evacuation Route Model for Disaster Aﬀected

Areas. AICS 2020, (p. 12).

A Research Supervision Framework for Quality and Scalability

403