=Paper=
{{Paper
|id=Vol-2348/paper02
|storemode=property
|title=Student-Centric Learning Analytics: Putting Analytics into the Hands of Students
|pdfUrl=https://ceur-ws.org/Vol-2348/paper02.pdf
|volume=Vol-2348
|authors=Craig Banyard,Ismini Vasileiou
|dblpUrl=https://dblp.org/rec/conf/cerc/BanyardV19
}}
==Student-Centric Learning Analytics: Putting Analytics into the Hands of Students==
https://ceur-ws.org/Vol-2348/paper02.pdf
Big Data, Warehousing and Data Analytics
Student-centric Learning Analytics: Putting
analytics into the hands of students
Craig Banyard[0000−0002−0589−3894] and Ismini Vasileiou[0000−0001−6174−3586]
University of Plymouth, Drake Circus, Plymouth PL4 8AA, United Kingdom
craig.banyard@plymouth.ac.uk ismini.vasileiou@plymouth.ac.uk
https://www.cscan.org/
Abstract. Learning Analytics is growing rapidly across Higher Edu-
cation institutions across the globe. There are some software solutions
to learning analytics which are mainly focusing on intervention, reten-
tion and resource allocation, driving institutions to collect and analyse
data at an institutional level. Such data only provides a small window
into a student and how they spend their time learning. Institutions link
progression to specific interactions and neglect students’ self-study, any
social factors and any individual learning needs. Therefore, this paper
is looking at Learning Analytics and how data can help each student
individually. To achieve this, the data should examine and analyse a
student providing intuitive data that can be used to reflect and alter
learning behaviour. The proposed solution is more than just contribut-
ing to a data-driven education system allowing students to add learning
activities, set personal targets and interact with members of staff during
sessions; this enhances students self-regulated learning skills and their
engagement while providing a clearer picture of a student’s learning be-
haviour.
Keywords: Learning Analytics · Student Engagement · Higher Educa-
tion · Big Data · Analytics
1 Introduction
In recent years various learning pedagogies have emerged and suggested that
there are changes in the way students learn. Personalised learning is becom-
ing core amongst Institutional Education Strategies, and the use of technologies
spreads rather fast. Learning, whether it takes place online, face to face, or as
flipped classroom approach, changes the group dynamics and puts the learner
in the heart of the education system. For a student, engaging in the process
of learning has been an area of research at the University of Plymouth. With
aims and goals around student motivation, student engagement and participa-
tion, learning becomes more profound, the learner becomes a collaborator, and
the institution needs to participate in instructional planning. With Learning
Analytics, institutions can enhance their reputation by increasing student learn-
ing retention and success. By using Learning Analytics, institutions value their
27
Big Data, Warehousing and Data Analytics
2 Craig Banyard and Ismini Vasileiou
learners, and the latter can shape their experiences and learn according to their
goals and ambitions.
Learning Analytics refers to those applications that analyse educational data
in an attempt to provide both the learner and the tutor with patterns of be-
haviour to improve learning and its related activities. The UK Higher Education
sector is trying to respond to forces and changes that either stakeholders or the
government is pushing and introducing. Moreover, there are great opportunities
that are arising where current practices can be examined and new solutions to
emerge.
Kolb’s traditional experiential learning cycle[1] which was based on Piaget
and Dewey theories[2,3], which has been there as a foundation for all recent
developments of teaching and learning. In addition Lewin’s suggestions of how
a student can learn via the route of feedback[4], making a stronger case of visu-
alising the student data, and the need for generating data that will be able to
measure and guide any intervention needed. Being open and transparent with
the Institutional data, we develop greater social acceptability[5]. By seeing the
students as partners, by understanding what student engagement means for each
individual, we understand better as educators the challenges we are facing[6,7].
This, of course, is undoubtedly not a straightforward scenario and it can be af-
fected by data protection legislation. Using personal data is becoming more sen-
sitive nowadays. In this research, Learning Analytics were used fundamentally
to try and bridge the gap between student data, measuring student engagement
and institutional policies. The research, is proposing strategies on understanding
the student data, how the data can be used at the local level and proposes a
conceptual framework for achieving such strategies to improve desired outcomes.
2 Student Engagement in Higher Education
Student Engagement is becoming a core area of research across many UK Insti-
tutions. Gradually students are invited to suggest views and often make changes
to the curriculum and/or research. The growth of roles, responsibilities and op-
portunities that arise from such potential collaborations, increase the student
involvement but at the same time it creates a lack of clarity over the defini-
tion of ”Student Engagement”. Common questions we often hear at conferences,
policy and curriculum meetings are ”when is a student engaged?”, ”how do we
measure student engagement?”, and ”what stops students from engaging?”.
In 2016 the UK Higher Education faced the introduction of the Teaching
Excellence Framework (TEF)[8]. This framework aims to measure the retention
rates of each institution, student satisfaction and employability. The UK HE
sector is making an attempt to analyse and understand the various data that it
is holding and extract common patterns for which policies can be implemented.
In order, though to achieve this, the HE sector needs to fund the area of learning
and data analytics in order to understand the complex nature of the data already
28
Big Data, Warehousing and Data Analytics
Putting analytics into the hands of students 3
collected but also the future data. The figure below shows how student experience
is being influenced these days and the main areas students can affect.
Fig. 1. The student experience that influences institutional NSS result and ranking.
The authors, reinvestigated the current UK Higher Education climate, analysing
the current Learning Analytics process and it was apparent that the institutional
focus across the UK was mainly on data collected by the establishment and what
the latter can do with such data. For the purpose of this study, the authors ap-
proached the data from the student perspective. Thus, the data collected and
analysed had to tell a story where the student can use the data for their con-
tinuous personal development and reinforcing their learning. The questions that
started formulating were around attendance, module grades, overall course grade
and of course how all these define student engagement. Some argue that when
face to face interaction is present, then the students perform better.
29
Big Data, Warehousing and Data Analytics
4 Craig Banyard and Ismini Vasileiou
Fig. 2. Learning Analytics application cycle.
3 The Case Study
Higher education institutions are collecting vast quantities of data; however, the
data collected only shows a narrow insight into students learning and devel-
opment. The institutions can only analyse the data that has been produced on
university systems or while a student is on campus, this can produce a bias in the
data which during analysis can lead to misinterpretation. The literature suggests
that student engagement is closely linked to the students’ desired outcomes from
university and can attribute to students persistence, educational attainment and
their learning and development [9].
The authors designed an application to provide students with a broad pic-
ture of their learning behaviours while fostering student engagement. It was
identified that student participation is essential to gain a comprehensive view
as there are many learning activities for example ”Writing up Notes” that with-
out participation from the students it would be impossible for universities to
collect and analyse. Ifenthaler and Schumacher[10] conducted a study on ”Stu-
dents perceptions of privacy principles for learning analytics”. The results show
students are more willing to share their data with a learning analytics solution
that also provides them with detailed and meaningful information. The aim of
the application was not merely to get students to add data without anything in
return; the application provides students with intuitive data that can be used in
self-evaluation.
30
Big Data, Warehousing and Data Analytics
Putting analytics into the hands of students 5
The application has three main components; the first allows the student to
view and add learning activities, giving the student the ability to monitor how
long they are spending on each learning activity they are also able to compare
themselves with the average of his or her peers. The second component of the
application allows users to set personal targets for learning activities and modules
which can keep them on track or to push themselves to spend more time on a
particular activity. The final component of the application facilitates interaction
with a member of staff during sessions. Students can ask anonymous questions
which the lecturer can see in real time, giving students the ability to engage in
a dialogue without fear of asking a question that might seem irrelevant to their
peers. The lecturer is also able to poll the room they can define the question and
what each of the three responses are, allowing them to quickly gauge students’
understanding of the material and make their sessions more interactive.
Fig. 3. User interface of proposed application: A: Overview of students progress, B:
Detailed overview of a target and C: Lecture registration
4 Discussion
As the authors reflected on the outcomes of the app, it was evident that there
is a high need to use the available institutional data not only for the benefit
of improving HE policies, but also to offer the students direct access to their
profiles, give them the opportunity to identify areas of development and self
31
Big Data, Warehousing and Data Analytics
6 Craig Banyard and Ismini Vasileiou
awareness. Big Data is becoming increasingly important research across the so-
ciety. In Education, the collection of data over the years has been massive, but
it has not been used in an attempt to provide definitions and terminology that
will be useful to both educators and students. During this research, by reversing
the focus of data and the outcome of them we have defined the below definitions
of Learning Analytics[11].
Table 1. Definitions of various types of Analytics suitable in HE
Term Definition
Academic Analytics This is where the focus lies with processing the analysis of assess-
ment and perform comparisons across individuals, institutions
and programme/courses. This helps Schools and Faculties to de-
fine the career pathways they are offering and the development
of them. In addition, it helps to evaluate teaching and learning
approaches and identify areas where intervention is needed.
Learning Analytics Interpreting the data that are being gathered on behalf of the
students and identifying their learning objectives and how that
feeds back to the module and its allocated assessed element is
an aspect that at least UK establishments are not implementing
fully. The desire though to produce and predict student progress
is vital; this can be achieved by observing learning behaviours
and data associated with it.
Predictive Analytics In this category the need to identify patterns of reliable conclu-
sions in order to lead to actions that demonstrate impact and
change across one or more educational establishments.
Action Analytics Recognising the high need for academic productivity can be
achieved by focusing on encompassing practices, and measur-
ing the innovation and performance by cultivating and shifting
behaviours and cultures.
The development of the proposed app has led the authors to propose a con-
ceptual framework of how an establishment can use the data collected to support
students’ personal and continuous development. Since it is acknowledged that
data are collected at different levels and for different needs, it is proposed that
Higher Education should start focusing on using Learning Analytics for the ben-
efit of the individual learner and not just for benefiting the institution’s goals[12].
The figure below shows how decision making can be influenced, informed and
reinforced by the process of applying analytics to a Higher Education establish-
ment.
32
Big Data, Warehousing and Data Analytics
Putting analytics into the hands of students 7
Fig. 4. Conceptual framework of Learning Analytics and Decision Making in Higher
Education
5 Conclusion
Learning in Higher Education is increasingly becoming a large area of research. It
is mostly common that Institutions develop and research this field for the benefit
of the tutor. However, in our research we have placed the focus on the student
and it is aiming at developing such an approach where learning analytics will be
used by the students. This is the beginning of our research and case study and
we intend to investigate in a variety of future research questions such as what
institutional resources do we need? How can we integrate analytics and learning
to design teacher inquiry? What is the technical, financial, social and cultural
impact on institutions?
Data can change everything. Learning Analytics in Higher Education is an
evolving process, with no clear start point and no clear end point. Predictive
analytics should help not only individual students, but in the bigger picture, it
could help institutions to identify recommendations around realities and impli-
cations. Analytics is becoming an important part not only for Higher Education,
but also businesses. It is an emerging technology and the interactions behind it
can lead to successful summarised reports and displays. Business intelligence
techniques assist the analytics techniques to continue to grow and to enable
better decision making. The responsibility of the researchers investigating and
manipulating such data is great and this is an area that still lacks guidance in
the modern world. With the 4th Industrial revolution the philosophical approach
and personal beliefs of the researchers can impact on efficacy. Researchers often
focus mostly on exploring the various dimensions of learning analytics. What it
tends to be missing is the analysis of learning outcomes, how those are measured,
what patterns can be formulated and how the latter can enforce and reinforce
decisions. Transparency is a key aspect of any form of analytics and especially
33
Big Data, Warehousing and Data Analytics
8 Craig Banyard and Ismini Vasileiou
when it comes to education, institutions and their direct relationship to the
students and the student experience.
Social conditions change rapidly. Social networks, forms of communication,
the Internet of Things and so many other things influence but at the same time
inspire every individual and in this case the students. Student demand is grow-
ing. There is emerging evidence that students wish to be provoked, challenged
and pushed in their learning. Moreover, they want to be more informed about
their learning progress. The possibilities of expanding the Learning Analytics
process for every Higher Education establishment are huge. Institutional deci-
sions should be carefully informed by using Learning Analytics in such a way
that is a collaborative approach between the educators and the students. This
way, Higher Education will be in a good position to deliver but also to enable
success.
References
1. Kolb, D.: Experiential Learning: Experience as the Source of Learning and Devel-
opment. Englewood Cliffs: Prentice-Hall. (1984)
2. Piaget, J.: The Language and Thought of the Child. New York: Meridian (1955)
3. Dewey, J.: Experience and education. New York: Collier Books. (1963)
4. Lewin, K.: Field theory of learning. In The psychology of learning. The forty-first
yearbook of the National Society for the Study of Education: Part II, The psychology
of learning., pp.215-242. (1942)
5. Clow, D: The learning analytics cycle: closing the loop effectively. In: Proceedings
of the 2nd International Conference on Learning Analytics and Knowledge - LAK
12, p. 134. (2012)
https://oro.open.ac.uk/34330/1/LAK12-DougClow-personalcopy.pdf
6. Lowe, T: Data Analytics A critique of the appropriation of a new measure of
Student Engagement. Student Engagement in Higher Education Journal 2(1), 2–6.
(2018)
https://journals.gre.ac.uk/index.php/raise/article/view/Lowe/636
7. Shaw, C., Sims, S., King, S., Paddison, A. and Lowe, T.: The Development of
Contemporary Student Engagement Practices at the University of Winchester and
Winchester Student Union. 1(1). (2017)
https://winchester.elsevierpure.com/en/publications/
the-development-of-contemporary-student-engagement-practices-at-t-3
8. Teaching Excellence Framework: analysis of metrics,
https://www.gov.uk/government/publications/teaching-excellence-framework-analysis-of-
Last accessed 10 Nov 18
9. Kuh, G., Kinzie, J., Buckley, J., Bridges, B. and Hayek, J.: Piecing Together the Stu-
dent Success Puzzle: Research, Propositions, and Recommendations ASHE Higher
Education Report, 32(5), pp.1-182. (2007)
https://doi.org/10.1002/aehe.3205
10. Ifenthaler, D. and Schumacher, C.: Student perceptions of privacy principles for
learning analytics. Educational Technology Research and Development 64(5), 923–
938. (2016)
11. Van Barneveld, A., Arnold, K. and Campbell, J.: Analytics in Higher Education:
Establishing a Common Language. Educause Learning Initiative. 1. (2012)
34
Big Data, Warehousing and Data Analytics
Putting analytics into the hands of students 9
12. Picciano, A.: The Evolution of Big Data and Learning Analytics in American
Higher Education. Online Learning, 16(3). (2012)
https://files.eric.ed.gov/fulltext/EJ982669.pdf
35