time bacsich final final pdf
DESCRIPTION
The cost- and time-effectiveness of online learning (e-learning, technology-enhanced learning): providing a perspective on Microlearning and the differences between academic and corporate viewsTRANSCRIPT
Paul Bacsich January 2012
The cost- and time-effectiveness of online learning: providing a
perspective on Microlearning and the differences between
academic and corporate views
Paul Bacsich
Abstract
Time “is strangely under-examined in the literature of e-learning” (Goodyear 2006). This paper aims
to build on this observation to provide a researched starting point for a new synthesis of time
aspects of e-learning in general, and microlearning in particular
Acknowledgements
I am indebted to the Microlearning organisation and specifically to Professor Peter Bruck for
facilitating the later phases of this work via an invitation to speak at the Microlearning 5.0
conference in Innsbruck, Austria in July 2011. My researcher Sara Frank Bristow was responsible for
the first draft of the bibliography on “time”.
Time has been a key factor in this paper, and surrounding its writing. From time to time, work on this
topic has been supported by JISC via its support of my work on costs of e-learning. I am grateful to
my wife and colleagues for their support and flexibility in allowing me to eventually finish this paper.
Even Time3 did not help (read on...)
Microlearning
As is common with concepts in learning, there is a range of meanings of the word “microlearning”.
Wikipedia (2011) in its informative article starts off its treatment with the crisp statement
“Microlearning deals with relatively small learning units and short-term learning activities”. The
educational technologist Schneider (2010) in his EduTech wiki article places this definition alongside
two others derived from the Microlearning project:
[a] term used in the e-learning context for a learner’s short interaction with a learning
matter broken down to very small bits of content. At present this term is not clearly
defined. Learning processes that have been called “microlearning” can cover a span
from some seconds (e.g. in mobile learning) to 15 minutes (learning objects sent as e-
mails).
in a wider sense [microlearning] is a term that can be used to describe the way more
and more people are actually doing informal learning and gaining knowledge...
especially those that become increasingly based on Web 2.0 and Wireless Web
technologies.
The literature references from these sources all point to the work of Hug and in particular to his
analysis of the dimensions of microlearning (Hug 2006). Of these dimensions, time is the one
mentioned first. It is also implicit (rather than explicit) in the concepts of informal learning and Web
2.0 that time is a key factor.
The cost- and time-effectiveness of online learning
Paul Bacsich 2 January 2012
For the links to communities of practice (informal learning among experts) see in particular
Kahnwald and Köhler (2006) and for the links to my favourite Web 2.0 technology, the wiki, see
Langreiter and Bolka (2006). Both these papers also reflect on appropriate definitions of
microlearning.
What is clear from all these papers is that time is a key dimension of microlearning.
Time – and its cost
I have been interested in the costs of e-learning since the early 1990s and carried out a series of
studies on this for JISC in the late 1990s and early 2000s – the CNL studies (Bacsich 1999; Bacsich and
Ash 1999, 2000; Bacsich, Ash and Heginbotham 2001). It rapidly became clear in discussion in the
late 1990s between myself and other costing experts, in particular Stephen Ehrmann, that the cost
of student time (spent studying) tended to dominate the other costs in the calculations, provided
that one regarded student time as an eligible cost! (For an overview of Stephen’s work see
Chickering and Ehrmann 2008.) However, in those days, in higher education and in schools, the time
spent by students was not taken into account in any planning-oriented way – completely at variance
with the approach in the training world. Thus the time issues for students were elided away, in a way
which some said was similar to the renormalization of infinities in quantum theory.1
However, a few experts kept worrying away at the issue, while devoting the majority of their
research in costing to more tractable issues of costs of equipment, support, training, content
development and so on. An interesting recent development is the modelling work of Laurillard
(2007) on “benefits-oriented costs” which includes some aspects of time.
I used to give presentations on the time issue every year or two, in particular at my inaugural
professorial lecture, but in the “days of abundance” before the recent recession, governments were
not much interested in costs of education and even less in any time aspects. But the world has
changed now – and students’ attitudes also.
What learners want
Actually, even in the 1990s, the avoidance by university planning staff of time issues did not mean
that students were unaware of such issues. And since that time students (and their parents) have
become (a) concerned (in many countries though not all across Europe) about the cost (to them) of
higher education and (b) alleviated these concerns by working while in theory studying full-time (in
every country in Europe and most of the rest of the world).
From our own work (Bacsich passim) over many years on learners (mostly at university): students
believe that e-learning increases costs to them – but they behave as though it saves costs and saves
time (or at least expensive time).
1 As noted in Wikipedia, as Lewis Ryder has put it, “In the Quantum Theory, these [classical] divergences do not disappear;
on the contrary, they appear to get worse. And despite the comparative success of renormalisation theory the feeling remains that there ought to be a more satisfactory way of doing things.” See http://en.wikipedia.org/wiki/Renormalization
The cost- and time-effectiveness of online learning
Paul Bacsich 3 January 2012
From other work (including from the USA and UK OU) it is clear that students want flexibility (but not
too much). In some countries like the US, some students want “overclocking” (i.e. faster progress
through their degree, in particular to complete a 4-year degree in 3 years). However, in other
countries like the UK there is little interest in this aspect. On the other hand, most so-called full-time
students work while they are studying. Thus the concept of a “full-time” course lasting a fixed
number of years is increasingly invalid.
As noted, students seem to have a good informal understanding of cost and time issues inasmuch as
it affects them. (This is well documented in the literature as we shall show later.) It is thus all the
more surprising, in an era when universities are supposed to be listening to students, that this is not
the case among institutions – and this appears to be true worldwide.
Summary of research on costs of e-learning
In the 1990s a number of research teams in a number of countries took up the challenge of
establishing the cost-benefits of e-learning (online learning on- or off-campus). My team was one of
these and had the benefit of coming along a little later in the cycle since the UK was rather late in
realising that this was an important topic. The main research activities world-wide on “costs of e-
learning” of that era are summarised in the CNL Phase 1 Report (Bacsich et al 1999), which has an
extensive and comprehensive bibliography of all prior work on costs of e-learning and comparative
analyses of all the main methodologies of the time used in education and in training.
An ultra-summary of all our research in the 1999-2005 period is that it was not possible to truly
understand the issue of costs of e-learning without introducing aspects of Activity-Based Costing
(Bacsich and Heginbotham 2005) – ABC for short. The key reason was that one had to understand
how the pattern of use of time by teachers was different when they engaged in developing content
and teaching via e-learning as opposed to conventional face to face teaching
However, research and pilot trials by our team and others showed that to subject the teachers in an
institution to the full rigours of Activity-Based Costing and software was intractable, unpopular with
lecturers and did not garner support from senior management. Thus various more or less
unsatisfactory compromise approaches were proposed – which might be called ABC-lite – in order to
gain the benefits of Activity-Based Costing without the pain of a fully-fledged ABC approach (with
specialist software support). In the UK my team’s CNL approach and the INSIGHT approach which
emerged a little later from the University of Strathclyde (Nicol et al 2002) became popular – for
some time, but not for long.
Amazingly the matter largely rested there, until the next recession – deemed by some to have
started in late 2007. A study in 2008 by myself for JISC (Bacsich 2008) indicated that progress since
the earlier work had been negative! In particular, all of the earlier schemes had lapsed into
irrelevance – and several were not known even to all the (few) remaining active researchers in the
area.
This was particularly strange, even in Europe, for three reasons:
1. A global recession was starting (even if in Europe we were very slow to realise this – unlike in
the Gulf States and US)
The cost- and time-effectiveness of online learning
Paul Bacsich 4 January 2012
2. There was a substantial growing together of higher education and training, particularly in
the area of continuing professional development – and the training world remained expert
in analysing the cost-effectiveness of online learning
3. Even in universities, time recording of staff effort was not unknown, especially in non-
academic departments (especially libraries and IT departments), mandated by a variety of
influences including the European Commission for its funded projects, some national
governments and commercial partners, and the increasing use of hourly-paid temporary
staff. However, little of this work was published. (For an Australian university ABC study see
Reich and Abraham 2006.)
“No Significant Difference”
Even in the mid 1990s, it was regarded as surprising that there appeared to be large returns on
investment in the training world from some deployments of what would now be called e-learning,
yet little evidence of any similar return in the university sector.
A 700% return at Royal Bank of Scotland was reported by one commentator and it was routine to
expect 10-25-40% learning gains using e-learning compared with using a conventional classroom for
the same time, and some situations where 60% learning gains were reported. Very little of these
results came out into the public domain because the results were so spectacular that the companies
regarded them as trade secrets, and the culture and technology of the time (just at the start of the
Web) did not encourage or facilitate open publication.
On the other hand, it is believed by many academics, and apparently validated by the seminal work
of Thomas L. Russell (1999), that the mode of delivery makes no significant difference to the grades
or other performance indicators of students. The Web site http://www.nosignificantdifference.org,
now supported by WCET, rounds out and updates this earlier work. There have been several
impassioned critiques of this work (see for example Oblinger and Hawkins 2006) and an increasing
number of recent studies – Sweet and Sealover (2007) is one of many, relevant to us since it focuses
strongly on time issues) – and meta-analyses (US Department of Education 2010) do suggest
significant differences in some cases, but the main thrust of the NSD argument is still accepted by
many analysts.
But not by all – and by fewer and fewer institutions. KC Green (2010) has documented the
inexorable rise in online learning at US university and college campuses each year since the No
Significant Difference results first came out – use keeps growing. Carol Twigg at the National Center
for Academic Transformation talks about “moving beyond” No Significant Difference (Twigg 2001)
with an increasing focus on reducing costs (Twigg 2003).
I have contended since my 1999 work on costing that the No Significant Difference work has several
major flaws – in particular many key variables were not controlled. Putting it at its crudest, if
students are offered a course, in the majority of cases they will study as much on the course as they
think is justified, however well or badly the course is presented: many will aim to pass but not
necessarily excel, so that when their effort-grade curve begins to tail off (in their mind, as well as
they can predict) they will stop studying. In a situation where students were studying “full-time”
(often rather less than a working week at many universities in many countries) and time was thus
The cost- and time-effectiveness of online learning
Paul Bacsich 5 January 2012
elastic (even if there was a counter-pull from sports, meals, bars, clubs and so on) the amount of
effort they spend is up to them, not the lecturer.
Saving time
So how can the use of e-learning save time? (Even if the institutions do not notice, or care to notice,
this saving of time by students.) There are three main ways:
1. Travel
Just as in training, students can save travel time by studying via e-learning. Bacsich et al (1999, p.1)
noted that:
The student survey showed that there is a disjunct between student beliefs – in
essence, students believe that Networked Learning increases costs to them – and
student behaviour – time has an opportunity cost to them
Many academic thinkers still seem stuck in a mythical era where students walked (or at worst,
cycled) to a campus-based university where they spent all day, either in lectures or the library – or
laboratories. Somewhere like KU Leuven or Oxford comes to mind. Yet having lived in Oxford and
worked at the university, even 30 years ago many students had long bus journeys into Oxford and
the traffic was nightmarish. The reality of travel to many city-based universities, be they in London
or Hong Kong, is far worse, with commute times in excess of an hour. At many other universities,
even though students may be “supposed” to live nearby, many do not. When teaching in Sheffield a
few years ago, I was surprised to find several of my students living in other cities, with long commute
times, and subtly they implied that this commute time was the university’s fault – despite our
brochure clearly stating where our university was situated! So every journey saved – to a lecture, or
(much more common) to the physical library – is time saved – and indeed money directly saved as
travel is no longer free for students. The situation is far worse for part-time students, an increasing
minority in many universities, since they on average live further away. And all kinds of students
complain (Bacsich et al 1999, section 6.6) about other wastes of time:
Students reported that the main cost to them was time: including time to gain
familiarisation with software, time delays at log on, time queuing for printing, and time
lost due to system crashes. The telephone bill, hardware and software were next most
important – other items such as insurance were fairly low on the list.
That paper also reports that other projects have reached similar conclusions.
2. More rapid learning
Even in universities, students can also reduce “time on task”. This is not only possible by using
sophisticated e-learning systems (intelligent teaching systems, etc) – much more commonly students
can do this by studying only the parts they do not know, and thus avoiding lectures and other
activities where they have already got the knowledge – in other words, personalised learning.
The cost- and time-effectiveness of online learning
Paul Bacsich 6 January 2012
But there is no need to provide a personalised curriculum – students can do this within the standard
curriculum. We still tend to assume that students come to university with no relevant knowledge –
this is just not the case – and not only at Masters level.
3. The third way
To make sense of this we need to undertake some more analysis.
Bologna
The fundamental credentials of the university are defined by time, not by content or competence. In
Europe (and the countries influenced by it), each course is specified not only by its syllabus but also
by its “points”. A similar situation pertains in the US and the countries it influences. The European
Credit Transfer and Accumulation System – always now called ECTS – is defined as follows in the
ECTS Users’ Guide (ECTS 2009). The Guide is a complex document but the essence of ECTS is
described in just three short paragraphs.
ECTS credits are based on the workload students need in order to achieve expected
learning outcomes. Learning outcomes describe what a learner is expected to know,
understand and be able to do after successful completion of a process of learning. They
relate to level descriptors in national and European qualifications frameworks.
Workload indicates the time students typically need to complete all learning activities
(such as lectures, seminars, projects, practical work, self-study and examinations)
required to achieve the expected learning outcomes.
60 ECTS credits are attached to the workload of a fulltime year of formal learning
(academic year) and the associated learning outcomes. In most cases, student
workload ranges from 1,500 to 1,800 hours for an academic year, whereby one credit
corresponds to 25 to 30 hours of work.
This is the “Bane of Bologna” – but in fact there are several linked banes.
The first bane is that there is a 16% jitter in the definition of ECTS credits – anywhere between 30
and 25 study hours. Suppose a teacher worked very hard to ensure that he could teach his course in
such a way that students could take 10% less study time. An evaluation would probably not pick that
up. Worse than that, his rival could just drop 10% of the syllabus (keeping the same ECTS points) and
achieve the same result!
The second bane is that the figures for study hours are not realistic. In particular a “low end” student
workload of 1500 hours for an academic year equates to 37.5 hours per week, on the assumption of
a 40-week academic year (much longer than in many universities) and this figure of 37.5 hours per
week is far higher than survey figures suggest is typical in many countries. The well-known HEPI
student surveys in 2006, 2007 and 2009 produced some embarrassing figures:
[surveys] revealed an average of just 20 hours of total study per week in mass
communications, and not much more in business studies. This is just two thirds of the
amount required on average in, say, engineering.
The cost- and time-effectiveness of online learning
Paul Bacsich 7 January 2012
Even more surprising, even within a subject there were found to be large differences
between institutions. In historical and philosophical studies, for example, the loads
ranged from 39.5 hours in the most demanding institutions to just 14 hours in the
least. (HEPI 2009 paras. 7-8)
The third bane is that the figures are not comparable across nation states. The same studies noted
that:
students in other European countries undertook on average 15 per cent more study
each week than English students (HEPI 2009 para. 17)
There are differences even within countries. A degree course in Scotland, the northern part of the
United Kingdom, lasts four years of full-time study. In contrast, the degree courses in England, Wales
and Northern Ireland, the southern and western parts of the UK, lasts three years. There is a
considerable amount of evidence, including from international employers and universities admitting
graduates of such courses to doctoral programmes, that the 3-year and the 4-year systems in the UK
produce the same “quality” of graduates. The conventional reason adduced (though rarely in public)
for the different lengths of programmes is that in Scotland the output of college entrants from
schools is not as well qualified as that from England and so the degree course needs to “top up” the
competence of Scottish students. Even if this were true (and that is much disputed) such an
admission would open up a whole new can of worms about comparability of school systems across
the UK and Europe and why some universities in some countries can claim to be universities not
colleges yet be teaching school-level topics. If it is not true then Scottish universities – who admit
many UK students from outside Scotland – might be accused of wasting scarce UK money (from
government or students or parents) delivering a programme 1/3 less efficient than in other parts of
the UK. More likely is that the Scottish degree is different from the England/Wales/Northern Ireland
in some way not captured by the ECTS system (which is designed ab initio to capture comparability).
Whatever the truth, major issues loom around the issue of time-based measures for university
education.
Reaction from government and associations of universities in England to such studies has
understandably been negative, tending to challenge the assumptions or the methodologies, not deal
with the implications. However, a meta-analysis (CHERI 2009) carried out by the UK Open University
for the Higher Education Funding Council for England confirmed in broad ways the HEPI conclusions,
though phrasing its conclusions in more cautious language.
Competency measures
The classic approach to resolve the issues of time-based measures of education is to move to a
competency-based approach. This has been much talked about since the 1970s but with one notable
exception (discussed below) little has happened within universities. This paper is not an overview of
competency-based learning – merely an introduction to the topic designed to tease out the issues of
time. An introduction to early work on the topic can be found in Voorhees (2001).
A competency-based approach to learning is one where in order to pass a course a student need
only demonstrate competency in the syllabus – the demonstration can be via exam and/or some
kind of practical assessment, depending on the topic. There is no need to attend any specific classes,
The cost- and time-effectiveness of online learning
Paul Bacsich 8 January 2012
either face to face or even virtually. Thus there is no need to spend any specific minimum amount of
time on studying the course (other than the small amount of time spent on taking the
exam/assessment).
In Europe, such an approach would be fundamentally different from the time-based approach of
Bologna. Perhaps because Europe has spent so much effort on propagating Bologna across the
European Higher Education Area (a region much broader than the EU) there has been little energy to
spare for development of competency-based approaches, although there have been a few research
projects
Even in the United States, at the university level the only well-known exponent of a competency-
based approach is Western Governors University (WGU). In a recent overview of the topic, Kolowich
(2011) described the WGU model as follows:
Western Governors’s pedagogical model... focuses not only on teaching new skills but
also on awarding credits for existing ones. No classes, no lectures, no fixed academic
calendar. For what students can prove they already know, they get credit. For what
they can’t, they are given learning materials and some light guidance. Students are
charged tuition every six months and take exams whenever they feel they are ready.
The model, called “competency-based learning,” has a number of champions in the
policy world. The Council for Adult and Experiential Learning has advocated for “prior
learning assessment” as a way of making sure students – especially low-income ones –
do not pay any more to complete a degree than they have to. The Center for American
Progress has praised Western Governors specifically as a beacon of innovation in an
industry entrenched in the tyranny of the credit hour. The Stanford Research Institute
recently rose to the defense of a competency-based associate degree program in
nursing at Excelsior College, noting in an extensive study that its graduates are on par
with those of more traditional programs.
Competency-based learning is of course much better known in the training field and many aspects of
this have been taken up in the Vocational Education and Training (VET) sector – the “colleges” – in
several countries. This has been the case for many years in Australia where it is deeply embedded in
the TAFE sector (TAFE Queensland 2011) and extensively researched (MacKay 2004, MacMillan
2007), but also now to an extent in Europe – see for example Wesselink et al (2007). This makes it all
the more surprising that there is so little uptake in European universities, even the vocational ones.
Interestingly in the US there is also little uptake of competency-based learning in the community
college sector, perhaps because of their close links with universities.
In contrast, US schools (unlike European ones) are now, in several US states, getting interested in
competency-based learning in schools – the “K-12” sector. Nagel (2011) gives an overview of recent
K-12 studies from iNACOL and points in particular to the iNACOL (2011) aptly-named report It’s Not
a Matter of Time, and related studies.
Thus maybe at last, in several countries, this is now more focus on time issues in all sectors of
education – even if it will take years for this focus to translate into widespread deployment of
alternative non-time-based approaches.
The cost- and time-effectiveness of online learning
Paul Bacsich 9 January 2012
The dimensions of time
So maybe it is timely to look more deeply into “what is time”. By this I mean a practical human-
focussed not a philosophical or cosmological analysis. (Usefully, Goodyear (2006) observes that
“time and space are not containers for activity, but social constructs”.) I shall accept the
conventional wisdom that a year of time is divided into 365 days (with 366 in just under ¼ of the
years) and each day into 24 hours. Days are divided into hours and hours into minutes and then
seconds. International travel causes some complications, but for all but a few of us (and all but a
very few students) the majority of days in our year take place wholly within one time zone.
An accountant or project manager would say “time is what staff put down on their time sheet” (so
that it can be charged, directly or indirectly, to a client – such as the European Commission – or be
used to provide a basis for paying such staff). In such time sheets there are two sorts of time – two
dimensions:
1. Time1 – on duty – which used to be called “at work” – and thus charged
2. Time2 – off duty (not at work) – and thus not charged.
But is it really so simple? In my inaugural professorial lecture after joining Sheffield Hallam University
I proposed “time of the third kind” by analogy with the film Close Encounters of the Third Kind. This
topic appeared in presentations I gave from time to time (some of which appeared for a while on the
web, but later disappeared) but it was much later (Bacsich and Pepler 2008) before anything made it
into the public domain in a lasting way.
Time3 is “in-between” time – in other contexts (Bacsich 2009) I use the word “liminal” for this. Such
time can be in longer chunks which are:
on duty (on timesheet?) but less productive (though somewhat productive – e.g. travelling
(especially on train or plane, less so in car or bus)
off-duty but somewhat productive – e.g. in the bar with colleagues or at dinner with
customers.
In timesheet terms, these longer chunks of Time3 are often the kind of time period where one
agonises about how to code it on the timesheet – and sometimes the kind of time period which it
can be hard to charge out to clients – some national funding agencies in some countries refuse to
pay for travel time (or travel, even when necessary).
But Time3 can also emerge from the cracks, and this is where the link to microlearning is very clear.
In “interstitial time” – gaps between meetings, odd moments of reflection (or distraction), whether
on-duty or off-duty, and other short periods – access to a personal mobile device (and one’s brain)
can allow one to make profitable use of these time fragments – to send emails or texts, look up a
blog posting or Wikipedia entry, etc. This is all possible and effective, without resorting to
contestable theories about “net gen” multitasking, since in such moments one is only handling one
foreground activity, the other activities being autonomic (like sitting or walking).
Thus Time3 has at least two subdimensions:
The cost- and time-effectiveness of online learning
Paul Bacsich 10 January 2012
1. a “larger” subdimension carved out of the liminal regions of Time1 and Time2 – “dark
matter”
2. a compact subdimension emerging from the insterstices in Time1 and Time2 – “dark energy”
harnessing the power of microlearning
The power of Time3 is that it allows one to expand the working day without requiring more than 24
hours in the day.
The value of time
The typical timesheet records time in amounts but it does not directly record its cost – and even
indirectly the best it will do is record the cost (or fee) to be charged to the client, not the cost to the
individual. Yet in situations where people have a free choice as to when they can work they will
choose to work at some times and not others. For example a landscape artist will choose to work
during the day – and during daylight hours – and a concert pianist will want (or need) to work in the
evening as that is when most concerts are. Many writers and some consultants like to get up really
early in the morning to write, so that later on in the day they can relax or make client calls;
notoriously many programmers claim that their best software development can be done only in the
evening hours. In the early days of computers, many junior researchers had to work in the middle of
the night since that was the only time of the day that they could book time on the mainframe, the
daytime hours being reserved for senior professors.
The theoretical insight is that each instant of time has an opportunity cost to an individual. This cost
is affected by their employer, their work requirements and their domestic circumstances. Greater
attention paid to this allows analysts to understand otherwise obscure situations.
Thought experiments on time
To make this “theory of time” come alive it is useful to consider three “thought-experiments”
(Gedankenexperimenten) – two concerning staff, one concerning students. These are based on real
examples from my academic career. There is only space to give outline solutions – all three are very
rich case studies. Some related examples are covered in Goodyear (2006, pp. 91-93).
1. Attracting lecturers to teach in face to face courses
Why did “The Captain” manage preferentially to attract many lecturers to teach on his postgraduate
face to face courses when on these courses they had the same teaching load as they would have had
on other face to face courses?
2. Attracting lecturers to teach online
A distance teaching university (DTU) can attract tutors (adjunct professors) even in times of high
employment, yet how can it do this while paying labour rates lower than equivalent positions at
other universities?
3. Good and bad lecturers – no significant difference?
A course is taught on campus to undergraduate students. A version of the course with the same
syllabus is also taught at a distance to part-time students. The same lecturer teaches both courses.
The cost- and time-effectiveness of online learning
Paul Bacsich 11 January 2012
One year she left and she was replaced by another lecturer who by general agreement was a much
less competent lecturer. That year the overall results on the face to face course were very similar to
the previous year (although the first assignment results were poor) but on the distance course the
results were catastrophically bad. Why was there such a difference?
Outline solutions
1. Attracting lecturers to teach in face to face courses
The Captain was a charismatic personality and treated his lecturers well but this cannot fully explain
the strong willingness of many lecturers to teach on his courses.
In this UK university, as in many, the teaching load is made up of a fixed number of “hours” (say 400)
per year. The teaching load for each lecturer is the sum of the teaching load on each module they
teach. For each module the teaching load is typically made up of two 1-hour sessions per week at
the same times each week. This means that the typical lecturer has a diary which is very much cut up
by teaching commitments each week of each semester. That makes it very hard to get uninterrupted
spells on research or to travel to the capital city to liaise with funding bodies and win more grants.
In contrast, the Captain’s postgraduate courses were taught in blocks of 2 or even 3 hours at a time.
Thus the disruption effect was considerably less. Indeed, some of the modules were taught in
intensive mode over 4-day sessions, and for those students and staff who found it appropriate, a few
of these were taught over the weekend in “residential” mode. This bunching meant that in a few
weeks of such intensive mode teaching lecturers could “burn off” (as they called it) a substantial part
of their semester teaching. Interestingly, since they had a lesser teaching load (typically 200 hours)
this proposition was even more attractive to the externally-funded and professorial staff, leading to
another bonus that the postgraduate students got the most research-active staff of all to teach
them. All within the same effort envelope – but a very different pattern of time!
2. Attracting lecturers to teach online
The solution again lies in the different pattern of time the university requires and some gender
biases in use of time and the opportunity cost of various time slots. Typically the DTU requires its
tutors to teach online during the evenings (when the DTU’s students, most of whom work in the day,
are online). Occasionally the DTU also requires its tutors to attend Saturday sessions presented face
to face in various cities. Many of the tutors are already employed (often at other universities) and
this evening and occasional weekend work is useful extra income. (There are secondary benefits too
– the DTU’s material may be relevant to their other teaching.) Other potential tutors are not
employed or not fully employed, due for example to childcare responsibilities – but evening work
(and given planning with their partner, the occasional weekend) is often easier to sustain.
3. Good and bad lecturers – no significant difference?
On the face to face course the students are mostly young and in theory studying full time. Even
though many do undertake some part-time work, it is easier for them to expand their time on just
one particular course when a lecturer fails to teach it effectively. Also being on campus most
students have an effective social network which straddles different years (a fact unknown to many
lecturers) and so after the shock of their poor results on the first assignment, contact was made with
the students from the previous year of the course (it was not a final year course) who supplied good-
The cost- and time-effectiveness of online learning
Paul Bacsich 12 January 2012
quality course notes and indeed answers to the previous years’ assignments. (Being not very
competent, the lecturer had made the classic mistake of merely changing some numbers in the new
assignments rather than changing the underlying approach.)
In contrast, the students on the part-time course were mainly in full-time employment and many
had family responsibilities also – thus they were very short of time and could not easily expand their
study time. Also, being older, less computer literate, rarely on campus and isolated even online from
the previous years’ students due to the way the learning environment was set up, they had much
less opportunity to do any social networking – and of course, not much time to do it in.
It was not surprising that the results were better in the second year that the bad lecturer taught the
course – the students, as typical at distance universities would have reached out one way or the
other to the previous cohort, but in the off-campus situation this requires more effort and takes a
longer elapsed time.
(The real-world case study would go on to show that the good lecturer had taught in a good but
minimal-content way which did not make much use of pre-prepared course notes on the learning
environment – and in terms of the material she did use, she took the Word files away with her,
leaving just some hard to update web pages. The less competent lecturer did not have the time or
skills to rapidly create effective online course material for his very first year. So there were some
issues of lecturer time, lecturer skills and learning design also to be learned.)
Measuring and researching use of time
Measuring use of time
In Europe there is sufficient experience with timesheets in EU projects that one can be reasonably
confident that staff in universities, even lecturing staff, can record time on projects – if they can be
motivated to (e.g. by EU funding). General experience including from industry is that it would be
hard and intrusive for them to monitor at a more detailed quantum than one hour on any kind of
regular basis, though for occasional intensive days or even a week this might be possible down to
the 15-minute level.
Students are reluctant to monitor time unless incentivised – and of course for staff and students the
monitoring is likely to distort the situation. My informal experience over many years is that moving
the monitoring quantum from one half-day down to one hour increases the amount of time
recorded – and I predict that moving the quantum from one hour down to 15 minutes – essentially
the cusp of moving into Time3, the threshold of microlearning – would again increase the total time
recorded. Nevertheless there are a number of reports over the years of useful results being gained
from student monitoring of time from as early as 1999 (Taraban et al 1999) to more recent studies.
These include such as Allan (2004), backed up by content analysis, with an interesting focus on time
management aspects of student learning, and a similar one by the same author and Lewis (Allan and
Lewis 2006) with a focus on workforce training.
For many years the great hope was that the system would do the time monitoring. There was a brief
period, maybe up to around 2000, when the majority of e-learning took place online via the Virtual
Learning Environment, when this might have been feasible, except that the systems did not support
The cost- and time-effectiveness of online learning
Paul Bacsich 13 January 2012
it. Some experiments using special tools produced useful results (Hwang and Wang 2004, though
perhaps overly mathematical in its analysis for some tastes).
However, since then the advent of web 2.0, multiple devices and off-line learning, means that this
seems infeasible if that is the only way of monitoring. However, a combination of student time
sheets, correlated to automatic monitoring, is a sound approach (Spector 2005).
Building on this, one should not underestimate the enthusiasm and talent of the Learning Analytics
community and it would no doubt be fruitful to join forces with them on ensuing that “Time
Analytics” can be supported – see in particular the proposals in Siemens (2011).
Conclusions
There are two headline conclusions:
1. We need better and more accepted ways of thinking about time in e-learning. At present
there is no consensus even among the few researchers active in the area.
2. We need better – less intrusive, more reliable – ways of measuring time use in e-learning –
and in working life! Closer liaison with the Learning Analytics community is indicated.
A remark on the time aspects of the IT underpinning e-elearning
I have one parting thought. If time is so important, have we really engineered IT systems to be time-
efficient, now that many users are power users? Every click counts!! I think there is in fact a
considerable amount of time optimisation to do even on so-called web 1.0 systems.
References
These are split in two: on “Time” and then on “Wider issues”.
On time
Allan, B. (2004). E-learners’ Experiences of Time. Networked Learning Conference 2004. Available at
http://www.networkedlearningconference.org.uk/past/nlc2004/proceedings/individual_papers/
allan.htm
Allan, B., & Lewis, D. (2006). Virtual learning communities as a vehicle for workforce development: a
case study. Journal of Workplace Learning, 18(6), 367-383. doi:10.1108/13665620610682099
Brown, A. H., & Green, T. (2009). Time Students Spend Reading Threaded Discussions in Online
Graduate Courses Requiring Asynchronous Participation. The International Review of Research in
Open and Distance Learning, 10(6), 51-65. Available at
http://www.irrodl.org/index.php/irrodl/article/view/760/1432
Goodyear, P. (2006). Technology and the articulation of vocational and academic interests:
reflections on time, space and e-learning. Studies in Continuing Education, 28(2), 83-98.
doi:10.1080/01580370600750973
The cost- and time-effectiveness of online learning
Paul Bacsich 14 January 2012
Hwang, W.-Y., & Wang, C.-Y. (2004). A Study of Learning Time Patterns in Asynchronous Learning
Environments. Journal of Computer Assisted Learning, 20(4), 292-304. doi:10.1111/j.1365-
2729.2004.00088.x
Laurillard, D. (2006). Modelling Benefits-oriented Costs for Technology Enhanced Learning. Higher
Education, 54(1), 21-39. doi:10.1007/s10734-006-9044-2
Spector, J. (2005). Time demands in online instruction. Distance Education, 26(1), 5-27.
doi:10.1080/01587910500081251
Sweet, S. and Sealover, P. (2007). The Affect of Decreased Classroom Time on Course Outcomes
Association for University Regional Campuses of Ohio. AURCO Journal, 13, 55-64.
Taraban, R., Maki, W. & Rynearson, K. (1999). Measuring Study Time Distributions: Implications for
Designing Computer-based Courses. Behavior Research Methods, Instruments & Computers,
31(2), 263-269. Available at http://www.ncbi.nlm.nih.gov/pubmed/10495808
On wider issues
Bacsich, P. (1999). The Hidden Costs of Networked Learning: The Consequences for University
Administrators. Paper presented at the 12th International Meeting of University Administrators
(IMUA); 5-9 September 1999, Edinburgh University, Scotland. Available at http://www.matic-
media.co.uk/publications/IMUA99.doc
Bacsich, P. (2008). Costs of E-Learning Scoping Exercise: Report. Report to JISC, October 2008.
Available at http://www.sero.co.uk/assets/celse_report.doc
Bacsich, P. (2009). Liminality. Page on Re.ViCa wiki, last updated 28 August 2009. Available at
http://www.virtualcampuses.eu/index.php/Liminality
Bacsich, P. & Ash, C. (1999). The hidden costs of networked learning: The impact of a costing
framework on educational practice. Proceedings of the ASCILITE Conference 1999, Queensland
University. Available at
http://www.ascilite.org.au/conferences/brisbane99/papers/bacsichash.pdf
Bacsich, P. & Ash, C. (2000). Costing the lifecycle of networked learning: documenting the costs from
conception to evaluation. ALT-J Volume 8 Number I, 2000. Available at
http://repository.alt.ac.uk/324/1/ALT_J_Vol8_No1_2000_Costing%20the%20lifecycle%20of%20
netwo.pdf
Bacsich, P. & Heginbotham, S. (2005). The Trial of Activity-Based Costing at Sheffield Hallam
University. First published April 2002 with additional footnotes added September 2005.
Available at http://www.matic-media.co.uk/CNL2-overview.doc
Bacsich, P., & Pepler, G. (2008). Organisational Change: First Report. CAPITAL Horizon Scans, July
2008. Available at
http://www.lsri.nottingham.ac.uk/capital/Yr1/HorizonScans/HS1%20Organisational%20Change.
The cost- and time-effectiveness of online learning
Paul Bacsich 15 January 2012
Bacsich, P., Ash, C. & Heginbotham, S. (2001). The Costs of Networked Learning – Phase Two. Report
to JISC, Sheffield Hallam University. Available at http://www.matic-media.co.uk/CNL-2-
handbook.doc
Bacsich, P., Ash, C., Boniwell, K. & Kaplan, S. (1999). The Costs of Networked Learning (Phase 1).
Report to JISC, Sheffield Hallam University. Available at http://www.matic-media.co.uk/CNL-1-
handbook.doc
CHERI. (2009). Diversity in the student learning experience and time devoted to study: a comparative
analysis of the UK and European evidence. Centre for Higher Education Research and
Information, The Open University. Available at
Chickering, A. & Ehrmann, S. (2008). Implementing the Seven Principles: Technology as Lever.
Reprint and update of Implementing the Seven Principles: Technology as Lever, AAHE Bulletin,
October, pp. 3-6, 1996. Available at http://www.tltgroup.org/programs/seven.html
European Commission. (2009). ECTS Users’ Guide. Available at
http://ec.europa.eu/education/lifelong-learning-policy/doc/ects/guide_en.pdf
Green, K. (2010). The 2010 Campus Computing Survey: The 2010 National Survey of Information
Technology in U.S. Higher Education. Campus Computing Project, 2010. Available at
http://www.campuscomputing.net/sites/www.campuscomputing.net/files/Green-
CampusComputing2010.pdf
HEPI. (2009). The Academic Experience of Students in English Universities: 2009 Report. Higher
Education Policy Institute. Available at http://www.hepi.ac.uk/466-1393/The-Academic-
Experience-of-Students-in-English-Universities-(2009-report).html
Hug. T. (2006). Microlearning: A New Pedagogical Challenge (Introductory Note). Proceedings of
Microlearning 2005. Available at http://www.microlearning.org/ml_files/microlearning-
conference2005_hug.pdf
iNACOL (The International Association for K-12 Online Learning). (2011). It’s Not a Matter of Time:
Highlights from the 2011 Competency Based Learning Summit. iNACOL/CSSO, July 2011.
Available at
http://www.inacol.org/research/docs/iNACOL_Its_Not_A_Matter_of_Time_full_report.pdf
Kahnwald, N. & Köhler, T. (2006). Microlearning in Virtual Communities of Practice? An explorative
analysis of changing information behaviour. (Changing Patterns of Learning: Schools,
Universities, Vocational Training). Proceedings of Microlearning 2006. Available at
http://www.microlearning.org/ml_files/microlearning-conference2006_kahnwald-koehler.pdf
Kolowich, S. (2011). Model of the Moment. Inside HigherEd, 9 May 2011. Available at
http://www.insidehighered.com/news/2011/05/09/western_governors_university_and_online_
competency_based_learning_model_gain_traction
Langreiter, C. & Bolka, A. (2006). Snips & Spaces: Managing MicroLearning (On Microlearning and
Microknowledge in a Microcontent-based Web). Proceedings of Microlearning 2005. Available at
http://www.microlearning.org/ml_files/microlearning-conference2005_langreiter-bolka.pdf
The cost- and time-effectiveness of online learning
Paul Bacsich 16 January 2012
MacKay, H. (2004). Locating the Fault Line: The Intersection of Internationalisation and Competency-
based Training. International Education Journal Vol 4, No 4, 2004 (Educational Research
Conference 2003 Special Issue). Available at
http://ehlt.flinders.edu.au/education/iej/articles/v4n4/mckay/paper.pdf
MacMillan, G. (2007). 30 Years on from Kangan: An Analysis of the current Policy Position of TAFE
Queensland. Dissertation for Doctor of Education, Queensland University of Technology, August
2007.
Nagel, D. (2011). Towards a Competency-Based Learning System. THE Journal, 26 July 2011.
Available at http://thejournal.com/articles/2011/07/26/toward-a-competency-based-learning-
system.aspx
Nicol, D., Kay, N., Gordon, G. & Coen, M. (2002). INSIGHT: A Model for Evaluating the Costs and
Benefits of ICT in Teaching and Learning, February 2002,
http://www.jiscinfonet.ac.uk/Resources/external-resources/strathclyde-insight-case-study
Oblinger, D. & Hawkins, S. (2006). The Myth about No Significant Difference. EDUCAUSE Review, vol.
41, no. 6 (November/December 2006): 14–15. Available at
http://www.educause.edu/EDUCAUSE+Review/EDUCAUSEReviewMagazineVolume41/TheMyth
aboutNoSignificantDiffe/158103
Reich, F. & Abraham, A. (2006). Activity Based Costing and Activity Data Collection: a Case Study in
the Higher Education Sector. Available at
http://ro.uow.edu.au/cgi/viewcontent.cgi?article=1220&context=commpapers
Russell, T. (1999). The No Significant Difference Phenomenon. IDECC, 1999. For ordering information
see https://www.arello.org/store/default.cfm?prodlistid=16-17
Schneider, D. (2010). Microlearning. Edutech wiki. Available at
http://edutechwiki.unige.ch/en/Microlearning
Siemens, G. (2011). Open Learning Analytics: A proposal. elearnspace, 10 December 2011. Available
at http://www.elearnspace.org/blog/2011/12/10/open-learning-analytics-a-proposal/
TAFE Queensland. (2011). Flexible learning options. Last updated 8 August 2011. Available at
http://www.tafe.qld.gov.au/courses/flexible_study/index.html
Twigg, C. (2001). Innovations in Online Learning: Moving Beyond No Significant Difference. A report
to The Pew Learning and Technology Program 2001. Center for Academic Transformation.
Available at http://www.thencat.org/Monographs/Mono4.pdf
Twigg, C. (2003). Improving Learning and Reducing Costs. Educause Review, 38 (October), 28–38.
Retrieved from http://www.thencat.org/PCR/Rd1Lessons.pdf
US Department of Education. (2010). Evaluation of Evidence-Based Practices in Online Learning: A
Meta-Analysis and Review of Online Learning Studies. Office of Planning, Evaluation, and Policy
Development: Policy and Program Studies Service, Revised September 2010. Available at
http://www2.ed.gov/rschstat/eval/tech/evidence-based-practices/finalreport.pdf
The cost- and time-effectiveness of online learning
Paul Bacsich 17 January 2012
Voorhees, R. (2001). Competency-Based Learning Models: A Necessary Future. New Directions for
Institutional Research, no.110, Summer 2001, pp. 5-13. John Wiley & Sons. Available at
http://www.medbev.umontreal.ca/gtea/Competency-Based%20Learning%20Models.pdf
Wesselink, R., Biemans, H., Mulder, M. & van der Elsen, E. (2007). Competence-based VET as seen by
Dutch researchers. European Journal of Vocational Training, no. 40, issue 1, 2007. Available at
http://www.cedefop.europa.eu/etv/Upload/Information_resources/Bookshop/468/40_en_wess
elink.pdf
Wikipedia. (2011). Microlearning. Wikipedia English Edition. Available at
http://en.wikipedia.org/wiki/Microlearning