EVALUATING ENGAGEMENT TO ADDRESS
UNDERGRADUATE FIRST YEAR TRANSITION
A Case Study
Clive Holtham
1
, Rich Martin
1
, Ann Brown
1
, Gawesh Jawaheer
2
and Angela Dove
1
1
Cass Business School, City University, 106 Bunhill Row, London, EC1Y 8TZ, U.K.
2
School of Health Sciences, City University, Northampton Square, London, EC1V 0HB, U.K.
Keywords: Student Engagement, Learning Analytics, Learning Design, Moodle.
Abstract: Rapidly changing demands from employers of students of business meant substantial redesign of the first
year undergraduate experience whose underlying pedagogy drew on the concept of “high-engagement”
learning. This paper focuses on the question of how engagement can be evaluated. It is argued that a variety
of “sensors” are needed for evaluation, both quantitative and qualitative. Of particular interest is the use of
Moodle logs as an emerging powerful sensor.
1 INTRODUCTION
Nicol (2006) summarises the particular importance
of the undergraduate first year and relates it to the
use of formative assessment, which, he argues has
considerable potential to enhance students’
subsequent experience.
This case study examines an innovative
undergraduate first year core module which
explicitly aimed to create through its learning design
high levels of student engagement (Cass Business
School, 2010). These high levels were planned to be
achieved through the design of learning activities
which were both electronic and non-electronic.
During the development phase of the project,
consideration was given as to how engagement
might be measured and evaluated. It was planned at
that time to use a mixed method, including a weekly
meeting of tutors, and heavy utilisation of the
Reports feature of the Moodle virtual learning
environment. A pilot was carried out in a small
elective module, and ways were found to track
engagement using the standard Moodle reports. But
it was also found to be time consuming and to be
unlikely to scale.
The assumptions explicit in the design of the
module were:
(a) the transition from high school to university
was becoming more problematic
(b) the core theory of engagement was Chickering
and Gamson’s (1987) long-standing framework
(c) the approach should be based on high-touch as
well as high tech (Naisbitt, 1999) through much
more extensive and intensive use of the virtual
learning environment
(d) a cross-university initiative in 2010 in learning
analytics had highlighted the potential of a data-
driven approach to high-touch interaction, and
new Moodle analytic facilities specifically for
this module were commissioned from the
Health Science School learning analytics
research team.
2 RESEARCH METHOD
The main body of empirical work undertaken and
reviewed here took place over a one year period
(2010-2011). The approach taken is participatory
and collaborative action research (Stringer, 1996).
The data sources which form the empirical
evidence base and that have been used to generate
and interrogate theory includes: our own reflexive
narratives in response to the developing work; the
textual material contained in the online collaboration
forums of the module tutors, the Moodle log data of
student activities, an online survey of the module
tutors, a sample of classroom interaction using
personal response systems, and a student focus
group.
223
Holtham C., Martin R., Brown A., Jawaheer G. and Dove A..
EVALUATING ENGAGEMENT TO ADDRESS UNDERGRADUATE FIRST YEAR TRANSITION - A Case Study.
DOI: 10.5220/0003925002230228
In Proceedings of the 4th International Conference on Computer Supported Education (CSEDU-2012), pages 223-228
ISBN: 978-989-8565-06-8
Copyright
c
2012 SCITEPRESS (Science and Technology Publications, Lda.)
3 ENGAGEMENT
Although motivation is one important factor in
engagement, engagement also relates to the level of
achievement. Perhaps the clearest identification of
high engagement is from Csikszentmihalyi (2002),
who applied his concept of “flow” to the educational
process. Astin (1993) reported that student
engagement is a key predictor of success in higher
education. Krause (2003) in turn suggested that
effective first year engagement involved students in
self-reflection in their first year at university. The
decision in the case discussed to move from earlier
VLEs was connected with a move towards greater
student engagement (Holtham & Courtney, 2006).
Kearsley and Shneiderman (1999), taking a
technology-orientated perspective, argue for
engagement theory as a basis for the use of new
technology to make new approaches possible. In the
event, the high-level group gave most weight to
Chickering and Gamson’s (1987) principles of good
practice in undergraduate education. These are in
effect a manifesto for a high-engagement approach
to learning, as opposed to a scientific framework.
1. encourages contact between students and
faculty,
2. develops reciprocity and cooperation among
students,
3. encourages active learning,
4. gives prompt feedback,
5. emphasizes time on task,
6. communicates high expectations, and
7. respects diverse talents and ways of learning.
A decade ago, we anticipated that proactive use
of a virtual learning environment would naturally
promote high engagement. Sadly, as identified by
JISC Digital Media (2011), much use of virtual
learning environments, including Moodle, is simply
as a content repository and assignment uploading
facility (Lane, 2009).
This narrow use is perhaps particularly
disappointing in Moodle, whose espoused
philosophy is avowedly social constructivist
(Moodle.org, 2011), embodying a change in role of
teacher from away from purely being a source of
knowledge. A text on Moodle as a business
(Henrick, Cole and Cole, 2011) stimulated in us the
conception that a VLE such as Moodle also had the
potential to provide the engine for a workflow
system, which could be used educationally.
The development of the module, drawing
together three separate modules was a complex task
and a fluid working group structure was developed
to ensure that as transparent an approach as possible
was taken to design and implementation. The design
team included an experienced learning designer at
professorial level who operated as both coach and
technical developer throughout the module itself.
This was in addition to school and programme-based
expertise in e-learning, without which an enterprise
of this nature could not have been contemplated.
At the time of selection of Moodle, radical
alternatives to a VLE were considered, such as a
personal learning environment (PLE) and generic
social media. Both of these are still under
consideration, but would at the most represent
augmentation above the VLE, rather than its
replacement.
The technological dimension was deeply
embedded in the module design, and symbolised by
the phrase high-tech/high-touch (Naisbitt, 2009).
One of our ongoing areas of pedagogic research is
into generational dimensions of learning and
technology (Rich, 2008), and current first year
students expect to engage with contemporary
technologies within their learning experience.
More particularly, in a first year first term
module, there is a particular concern about
identifying “at risk” students, who may not in
practice be participating, and a strong emphasis was
placed on promoting physical attendance and on
monitoring participation.
4 LEARNING ANALYTICS
The generic importance of analytics in learning had
been brought home to two members of the
development team who were in parallel also
involved in researching a large scale adult education
informal learning project, which was entirely web-
based and made very heavy use of web analytics to
track engagement of its audience of learners. There
was also familiarity in the development team with
web analytics being used widely in business. So the
team became interested in the potential for moving
beyond the minimally featured Moodle Reports, and
contact was made with the Health Sciences School
of the university where there was expertise in
Moodle analytics and in the mining of very large
datasets (Jawaheer et al, 2011).
Learning analytics is a very fast growing field,
with a lively leading-edge community promoting the
sharing of experience and the collective acceleration
of both theory and practice (Macfadyen & Dawson,
2010, Brown, 2011. Romero; (2010) outlines eleven
distinctive domains of the learning analytics
CSEDU2012-4thInternationalConferenceonComputerSupportedEducation
224
literature; this paper only relates to three of those
areas: Providing feedback for supporting instructors;
detecting undesirable student behaviours; and
constructing courseware.
A substantial body of work on VLE analytics is
beginning to emerge, and some of this (eg Urwin,
2011) is as with ourselves, concerned not simply
with retrospective historical tracking, but with what
we call “Action Support” that is, with learning
quickly and then taking direct action as a
consequence with the current cohort of students.
With a primarily face-to-face module, much of
the assessment of engagement would need to be
based on the two personal observation sensors,
physical and digital. Log data has proved to be
enormously helpful, but it does not relate to the
actual content, eg what is asked or said within a
discussion forum. We found difficulties in trying to
develop measures for each of the 7 principles. In
some ways though tutors felt that they could assess
engagement as a whole for their groups, for the
individual teams, and to some extent for individual
students.
5 RESEARCH PROJECT
FRAMEWORK
As the start of the module came closer, it was
essential from a research point of view to articulate
the parameters of the research project. The final
expanded framework is a layered model (Figure 1,
next page), where engagement is measured through
a number of “lenses”. The lowest layer is the vast
mass of data which derives from unfolding everyday
experiences of both students and teachers. This takes
many diverse forms - hard and soft; objective and
subjective; physical, digital and mental; explicit and
tacit; text and non-text, and the amount of such data
readily available in digital form has increased
considerably. However this increase does not
necessarily lead to more information and particularly
to more knowledge and insight. Our layered model
is built around a number of questions:
What is engagement? We have already
indicated our own use of the Chickering and
Gamson framework, augmented by the idea of
"flow" as indicating an extraordinarily high level of
engagement.
About whom can we evaluate engagement? In
the context of the present case study, we clearly
identified four levels - the cohort as a whole, the 6
tutor groups, the 24 teams and the 120 individual
students.
What are the lenses through which we choose
to evaluate engagement? In our case, we had
identified three lenses - taking the temperature,
insight into individuals, and searching for stimuli for
process improvement. This is the layer where
Figure 1: The full framework.
EVALUATINGENGAGEMENTTOADDRESSUNDERGRADUATEFIRSTYEARTRANSITION-ACaseStudy
225
information gets transformed into knowledge, and
also where the issue of plausible outcomes is
discussed.
In the case study "the seven sensors" for
gathering information were identified:
1. Observation and dialogue - physical
2. Observation and dialogue - digital
3. Assessments of student work
4. Attendance records
5. Formal surveys of students
6. Reflective journals
7. Virtual learning environment logs
We also regard sensors 1 and 2, direct
observation and dialogue (whether physical or
digital), as the "primary" sensors, due to their being
able to offer both broader and deeper sensing than
the other "secondary" sensors.
In this case study, very extensive use was made
of reflective journals, particularly proactive use was
made of attendance records, and there was slightly
above average use of formal student surveys. For us,
the major new sensor were the Moodle logs.
By the start of the module, the three lenses for
evaluation of engagement had been decided. Two of
these, relating to "overall temperature" and
"individual insight" might be found on any module
anywhere. However the third, "stimulus to
improvement" was a very specific function of being
a wholly new module run operated using a variety of
features which were distinctive to those involved.
The improvement lens potentially applies to any
of the levels of measurement, while the other two
relate to overall and specific levels respectively.
Moodle’s constructivist philosophy and emphasis on
learning communities makes it relatively weak in
organising reports by tutor group and team, and
much of the measurement customisation effort
related to generating "temperature" level reports.
Even within a single module, student
engagement can and perhaps should be defined in a
wide variety of ways. Despite the intrinsic difficulty
of measuring engagement, the course team was able
to identify five broad categories representing levels
of engagement. The highest and lowest of these were
fairly straightforward to identify, the highest
drawing on the concept of "flow". Whether in group
or individual work, it is generally not difficult to
observe flow. It does not mean all those with the
highest marks achieve flow - flow relates also to
fulfilling potential. A modest student may more than
fulfil their talents if they can achieve flow. A strong
student may get excellent marks without flow.
At the lowest level, non-engagement is a student
who rarely if ever shows up physically, rarely or
ever contributes online where the contribution is
voluntary, and often shows a lack of understanding
about even when and where the module is taking
place or what resources need to be consumed.
6 ACTIONABLE INTELLIGENCE
Regardless of the type of institution, learning
analytics almost encapsulates or symbolises a move
from a medieval (or at best nineteenth century)
lecture-based transmissive approach, to one that
embodies the idea of the academic as a facilitator of
learning, using a breadth of media both physical and
digital.
Good data alone is not sufficient: it needs to be
disseminated to the right people and to feed into
decision-making. Learning Analytics cannot be
divorced from the ongoing organisational pressures
and time shortages, and is most likely to be used if it
feeds into worthwhile actions (Campbell et al,
2007).
We also need to recognise that a trace is not the
same as the object or experience that made the trace.
Furthermore, some students prefer a static version of
resources due to their learning styles and time
management approaches; adding new links and
resources is not seen as beneficial by all.
Figure 2: Custom learning analytics: individual student
cumulative graph.
Actionable intelligence – Taking the
temperature
This involved daily participation statistics that
showed low usage, which caused concern and led to
development of a new mandatory Moodle “lesson”.
Actionable intelligence – Continuous
Improvement
Even basic Moodle reports enabled us to track
CSEDU2012-4thInternationalConferenceonComputerSupportedEducation
226
the views of each resource. An “activity report” on
library resources showed there was a good
immediate take-up for some sources. But Business
Source Complete, the key database resource, had
hardly been looked at, and this was the impetus for
urgent creation of a library discussion forum.
Actionable intelligence – Individual insight
The course team was preoccupied with students
making zero or very low contributions, both in
general and for specific online resources. A Moodle
report was used to identify students who had never
accessed the FAQ Forum, which we regarded as a
key indicator of engagement. Tutors used this data to
follow up with their own tutees in the “at-risk”
group the question of non-participation. Figure 2
represents a custom report on individual student
activity, produced via direct access to Moodle logs,
rather than via the standard Moodle reports.
7 CONCLUSIONS
After the end of the module, we reviewed how far
the 5 levels of engagement (no, low, medium, high
and flow) might inter-relate with the 7 sensors. The
interest was in how different sensors were able to
support the evaluation of different levels of
engagement. This opens the possibility of a
dashboard to support learning design relating to
engagement. The results are shown in Table 1.
Table 1: Sensors related to engagement levels.
It is well understood that attendance records are
a very limited tool for assessing engagement. But
they are quite a powerful and easy to use tool for
picking up non-engagement. VLE logs are also
useful in measuring zero and low levels of
engagement. But simply accumulating clicks in the
VLE rarely related to the highest level of
engagement. Indeed some of the students with
extremely high levels of VLE used appeared over-
anxious in their approach generally. Some
concluding reflections were:
1. We combined both computer and non-
computer based evaluations of engagement eg tutor's
opinion, as done in any module
2. We have extended this, both by the design of
the VLE and then the use of basic plus enhanced
metrics using VLE activity logs and we have used
other electronic methods such as survey and
clickers.
3. Our focus on evaluating engagement has
helped us to redesign learning activities within the
module, better to address Chickering and Gamson’s
definitions of engagement.
REFERENCES
Astin, A., (1993) What Matters in College? Four Critical
Years Revisited. Josey-Bass: San Francisco, CA
Brown, Malcolm (2011) Learning Analytics: The Coming
Third Wave Educause Learning Initiative Brief,
Educause,
Campbell, J. P., DeBlois, P. B., & Oblinger, D. G. (2007).
Academic Analytics: A New Tool for a New Era.
Educause Review, 42(August 2007), 40-57
Cass Business School (2010) “Transition issues and
creating a student community stemming from
Undergraduate Review – 1 year on report: Issues for
Cass Undergraduate Programme” Cass Board of
Studies Report, 2010
Chickering, A. W., & Gamson, Z. F. (1987). Seven
Principles of Good Practice in Undergraduate
Education. AAHE Bulletin
Csikszentmihalyi, Mihaly (2002) Thoughts about
education. New Horizons for Learning, Seattle.
http://education.jhu.edu/newhorizons/future/creating_t
he_future/crfut_csikszent.cfm
Henrick, Gavin; Cole, Jeanne and Cole, Jason (2011)
Moodle 2.0 for Business Beginner's Guide Packt
Publishing, Birmingham
Holtham, Clive and Courtney, Nigel (2006) High-
engagement learning for business education Higher
Education Academy Annual Conference, Nottingham,
3rd-5th July
Jawaheer, Gawesh; Kostkova, Patty; Sultany, Ajmal &
Bullimore, Anise (2011) Moodle Analytics Project:
Final Report City University, Learning Development
Centre, London
JISC Digital Media (2011) Use of VLEs with Digital
Media Accessed 19
th
November 2011: http://www.
jiscdigitalmedia.ac.uk/crossmedia/advice/introduction-
to-the-use-of-vles-with-digital-media
Kearsley, Greg & Shneiderman, Ben (1999) Engagement
Theory: A framework for technology-based teaching
and learning Version: 4/5/99 Accessed 9
th
December
2011: http://home.sprynet.com/~gkearsley/engage.htm
Krause, Kerri-Lee (2003) On Being Strategic About the
First Year, Griffith Institute for Higher Education,
Griffith University http://www3.griffith.edu.au/03/ltn
/docs/GIHE-First-Year-Experience.pdf
Krause, Kerri-Lee; McEwen, Celina and Blinco, Kerry
(2009) E-learning and the first year experience: A
framework for best practice EDUCAUSE Australasia
Conference, Perth, Western Australia, 3-6 May 2009
EVALUATINGENGAGEMENTTOADDRESSUNDERGRADUATEFIRSTYEARTRANSITION-ACaseStudy
227
Lane, L. M. (2009). Insidious Pedagogy: How course
management systems affect teaching. 14(10). First
Monday http://firstmonday.org/htbin/cgiwrap/bin/ojs/
index.php/fm/article/view/2530/2303
Macfadyen, Leah P. and Shane Dawson, (2010) “Mining
LMS Data to Develop an ‘Early Warning System’ for
Educators: A Proof of Concept,” Computers &
Education, vol. 54, no. 2 (2010), pp. 588–599.
Moodle.org (2011) About Moodle: Philosophy Accessed
19
th
November 2011: http://docs.moodle.org/21/en/
Philosophy
Naisbett, John (1999) High Tech, High Touch Broadway
Books, New York
Nicol, D. J. (2006). “Increasing success in first year
courses: assessment re-design, self-regulation and
learning technologies.” ASCILITE Conference,
Sydney, Dec 3-6, 2006
Rich M (2008) “Millennial Students and Technology
Choices for Information Searching”. The Electronic
Journal of Business Research Methods 6 (1), pp. 73 -
76, http://www.ejbrm.com
Romero, Cristóbal and Ventura, Sebastián (2010)
“Educational Data Mining: A Review of the State-of-
the-Art” IEEE Transactions on Systems, Man, and
Cybernetics—Part C: Applications and Reviews, Vol.
40, No. 6, pp601 – 618
Stringer, E. (1996) Action Research: A handbook for
Practitioners, Thousand Oaks: Sage.
Urwin, J. (2011) “Engagement with virtual learning
environments : a case study across faculties”, Blended
Learning in Practice, 2011: January , pp. 8-21.
CSEDU2012-4thInternationalConferenceonComputerSupportedEducation
228