productivity wordsThis is the first of two final posts in the series on whether online learning can increase educational ‘productivity.’ In this post I discuss some of the issues in identifying how online learning could lead to productivity gains in post-secondary education. In the second post, I will try to pull the main conclusions together

A list of all the previous posts in this series can be found at the end of this post.

Why this is a difficult topic

I have to say that this has been a bit of a hard slog. It’s not a topic that seems to fire the interest of most of the readers of this blog (at least in terms of the number of hits each post gets.) Nevertheless, I do believe that a hard look at the potential and limitations of online learning for improving the productivity of higher education is important. Politicians and, as Alex Usher calls them in his excellent blog, ‘techno-fetishist windbags who tried to make us all believe that …. MOOCs were an unstoppable wave of the future‘, often make at best naive comments about the ‘productivity’ of online learning, without really understanding what productivity means in educational terms, or explaining how the claimed productivity gains will occur. ‘Free’ to students of itself does not guarantee productivity gains, for instance, if the students are learning nothing or very little.

Another reason why it’s a hard slog is because little serious work has been done up to now on the topic of the productivity of online learning. Tom Carey’s and David Trick’s’s report for HEQCO is one of the few works on online learning that even raises the issue of productivity, and I think they would be the first to say that their report is very much a beginning of the discussion.There is almost no systematic research on the costs or ‘inputs’ of online learning, and certainly no cost research on some of the new forms of online learning, such as hybrid learning, mobile learning, OERs, or MOOCs (just because they are free to students does not mean there are no costs involved). In other words, we lack some of the really basic data needed for us to talk about productivity intelligently.

More importantly, though, while there is a large literature on the productivity of commercial companies or nation states, we don’t have a clear or agreed conceptual framework or definition of what productivity means in education, especially in terms of ‘output.’

Fears around ‘productivity’ in education

One reason of course for the lack of interest in productivity in education among educators is that the topic has undertones of commercialization. Many educators see education as a public good, which should be appropriately funded by the state, i.e. taxpayers, because the whole of society benefits from well educated people. Looking at costs or return on investment is ideologically disturbing for many of us working in education. But the real fear is that policy-makers and computer scientists will try to use ‘productivity’ in online learning as a way of eliminating instructors or teachers.

However, for me productivity in education can be put in very simple terms: is our society collectively (learners, taxpayers, parents, faculty, and other key stakeholders) getting the best bang for the bucks we are spending on education? If we can increase productivity we can educate more students to a higher level for the same investment. Why wouldn’t we want to do that?

Second, as we have seen from the posts so far, there is little at least immediately to fear from computers replacing faculty or teachers. Demand for post-secondary education far exceeds supply, but more importantly, there are extremely strong educational reasons why at least into the foreseeable future computers cannot do many of the really valuable things needed in teaching at a post-secondary level. Nevertheless, there are some things that online learning can do that could ease somewhat the financial stresses on the system, as we shall see, and we should be paying more attention to this.

Some key concepts and principles

I do not pretend to be an expert on productivity, despite a background in economics. Nevertheless, there are some basic concepts or ideas that need to be considered when trying to apply ‘productivity’ to online education.

Inputs and costs

Key terms in discussion of productivity in other contexts are ‘inputs’ and ‘outputs’. Inputs in education can be measured in various ways, but cost is one convenient way of counting input. The time that faculty and learners spend on teaching and learning is also a key input factor, which could be converted into an equivalent cost figure. Investment in technology and the human support required to make it effective is another input measure. Although not without some problems, though, the measurement of input with regard to online learning is not conceptually difficult. It’s just that it’s not been done up to now.


It’s when we come to outputs that there are more conceptual difficulties. How do we measure output in education, and in particular how do we relate outputs to inputs? What is the best way to measure outputs in education? OECD and other agencies use measures such as participation rates, graduation rates, standardized tests, etc. but at a post-secondary level these measures are more difficult. For instance there is a lot of mobility across different sectors of post-secondary education, with almost as many students transferring out of universities into two year colleges or institutes of technology as the other way. Do we count people who transfer halfway as drop-outs? How is this tracked? Does online learning facilitate this mobility (I believe it does) and is this mobility a good thing (I believe it is)? Lifelong learners have different needs from high school leavers. How important are these groups relative to each other, and does online learning serve one group better than another?

Many of the conceptual issues around international national standardized testing such as PISA apply even more so to measuring outputs at a post-secondary level. For instance, how do you measure what students have learned and more particularly relate that to the mode of delivery such as classroom versus online? How for instance do you measure creativity, originality and critical thinking, and relate that to teaching method and/or mode of delivery? In particular, what if, as I believe, online learning lends itself to new or better learning outcomes that have not to date been targeted in classroom teaching? One example would be the development of 21st century skills, such as digital literacy embedded within a specific subject domain. Thus we need to be careful when measuring productivity through online learning that we are not under-measuring some of its unique advantages. Furthermore, if it could be shown that learners learn more quickly, or instructors have to spend less time teaching to get the same results, this too would be a valid way of measuring improvements in productivity.

At the same time, it has to be acknowledged that there is wide disagreement among educators about the nature of learning, so what may be a perfectly acceptable means of increasing productivity for one person may be totally unacceptable to another. In essence, there are likely to be strong disagreements about the definition of valid measures of output, and how to obtain such outputs, which makes the discussion of productivity in education all that more difficult. One way to deal with this of course is to ask people to state what they see as valid outputs, then try to relate the use of online learning to meeting such outputs in a more productive way.

Scale and scope

One way to increase productivity in industry is through economies of scale: a standardized product that can be replicated very cheaply. MOOCs are a classic example of this in online learning: the same content delivered to hundreds of thousands of learners, with almost no variation to take account of individual differences, tastes or requirements (and almost no delivery costs, for that matter, except where broadband Internet access is expensive).

Production line car engines

There are some difficulties though with this concept being applied in education, particularly regarding standardized products. Especially at a university level, there is an argument for variations in the curricula taught across institutions, for example. This is partly to do with freedom of expression and partly with the need for heterogeneity and diversity in the knowledge base. Many would agree that we would not want all Canadian universities teaching exactly the same content in all subject areas. However, there may be more agreement on sharing common core foundation courses between different universities or colleges, which online learning would enable, thus achieving some economies of scale. The HEQCO report in particular emphasised the opportunities for economies of scale that online learning offers, but then pulled back, arguing that scaling is more difficult if not impossible for the ‘learning that matters most’ (see below).

It is worth noting that ‘economies of scale’ are associated with an industrial economic model, but online learning is based on digital not manufacturing technologies. Economies of scope are more associated with post-industrial economies. Economies of scope enable many variations on a standard product to meet individual needs at a low marginal cost for each variation. An example from industry would be different models of cars built around a common chassis and/or engine. An example from online learning would be a core curriculum with many optional routes through the material, using adaptive technologies that respond to the inputs from individual students in different ways, depending on the needs of the learner.

Digital technology in particular allows for an almost unlimited range of ‘options’ at low marginal cost. Social media take this even further with the concept of the ‘long tail’ strategy of of selling a large number of unique items with relatively small quantities sold of each. Applied to online learning, it would mean mounting a course for which there was very little demand locally, but would have a large market world wide.  While there has been quite a lot of discussion about economies of scale in online and open learning, the potential for economies of scope have been less well understood or discussed.

© The Social Enterprise Blog, 2009
© The Social Enterprise Blog, 2009

Replacement of labour by technology

A key way to improve productivity in the business world is to replace high cost labour with lower cost technology. Post-secondary education certain has high cost labour, which constitutes a major part of total costs. There is great interest then among policy makers and politicians in the possibility of computer-based learning replacing high labour costs, which partly explains the excitement about MOOCs.

However, there will be no productivity gains by replacing labour (instructors) with technology (computer-based learning), unless outputs are maintained or improved. It is important then to examine carefully the current state of computer-based learning, to see which parts of teaching and learning can effectively replace human instructors. This means disaggregating the different tasks or roles of instructors, and looking at those outputs that can be achieved better through computer-assisted learning as it exists now or in the near future. There is also a deeper philosophical issue, which is whether certain aspects of teacher-learner relationships should be replaced by computers. Are there some parts of the educational process that needs to remain ‘human’?

Process design and management

Another productivity issue is the efficient design and management of the processes by which ‘inputs’ are turned into ‘outputs.’ In a car company this would be the manufacturing process. The more one can simplify and/or reduce the cost of ‘processing’, while maintaining or enhancing quality, the greater the productivity.

A key process in education is the method of teaching. If we consider for example the development of 21st century skills such as independent learning, this will be heavily influenced not just by the content being taught, but how the teacher designs and/or how learners conduct activities that enable learners to develop such skills. In online learning this is usually called course design. We will see that course design can be in fact a major factor in increasing the productivity of online learning.

Learning that matters most

Lastly, it’s important that any attempt to measure productivity in education takes account of the nature of learning, and the different views of what constitutes learning. For me academic learning is a developmental process through which learners develop increasingly deeper and more complex understanding, and above all a capacity for learning how to learn within a particular subject domain or discipline. This means pushing beyond the superficial presentation and reproduction of information to students understanding what a subject discipline is really about and behaving and thinking as a professional within that subject domain. This for me is the kind of learning that matters most, as Tom Carey has described it in the HEQCO report. My experience of teaching online leads me to believe strongly that this kind of teaching is not only possible but often achieved in online learning. The challenge though is to scale that kind of learning or rather to find ways of teaching online that enable such learning to be successful across large numbers at less cost.

Connectivism and the wisdom of the crowd

Lastly we need in any discussion of productivity in online learning to consider a newly emerging area, the impact of social media and in particular the impact of massive inter-connections and communications across the Internet, on learning and knowledge.

Some, such as George Siemens and Stephen Downes, argue that informal learning, through online communities of practice or ad hoc or informal online connections through social media, and self-learning through Internet searching and networking, has massive potential for reducing the costs of education by content becoming increasingly freely accessible on the Internet and by eliminating or dramatically reducing the need for professional teachers.

© Ann Helmond 2009
© Ann Helmond 2009

 The next step

These are all different ways to look at productivity within a system. It can be seen that it is not a trivial issue and given its complexity, focusing on the productivity potential of online learning  is unlikely to provide policy-makers with the clear, simple answers that they seek. But I still think it is a worthwhile effort to examine at least the potential for online learning to increase productivity, to clarify the issues in doing this, and to have a stab at defining those areas that look most promising, and perhaps even more importantly, identifying possible dangers in certain approaches to productivity.

So in the next and last post in this series, I will attempt to identify those areas where online learning offers the greatest potential for productivity gains in post-secondary education.

Your turn

I would really like to hear from readers about:

  • whether you agree or disagree with these concepts or whether you think I have misunderstood what productivity means in post-secondary education
  • whether this post is useful for you in thinking about the role of online learning in your institution

Previous posts in this series include:

There is also a CIDER webinar presentation on the HEQCO report available from here


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.


  1. Thanks for another very clear post, Tony. I look forward to your “last post in this series”…I have the sense that there will be many more posts on this topic, whether explicitly part of this series or not!

    Re your comment that our HEQCO report was very much a beginning to a deep discussion of productivity issues in higher education: that may be too strong, what we did is probably more like a preface to the beginning :). The real starting point has to be a recognition by the key players – faculty and students – of the need to take productivity more seriously, and the opportunity it provides for developing key student capabilities. More on the opportunity in my video interview at

    Re your comment about “the learning that matters most”: we noted in the report that our use of this phrase was adapted from Steve Gilbert of the TLT Group in Washington DC. Steve has an evolving set of Principles for Teaching, Learning and Technology which are worth tracking at, e.g., “Beware of those who most strongly advocate/reject ‘scalable’ educational improvements” which includes the subtitle “There are no educational ATMs”.

  2. Posted by Wendy Kilfoil:

    I think that one aspect of productivity relates to time on task. It is easy enough for a student to be invisible on campus, not to attend class, to submit only the bare minimum, to take only the tests that count towards a semester mark. The ability to track students’ time online and their tool use, to give regular short tasks that require output, to involve students collaboratively, to provide a variety of resources, forces students to spend more time on task.

    I actually think that some online courses and certainly MOOCs require too much of students who then fall behind and drop out. It’s part of the role of the lecturer and/ or the course designer to calculate the amount of time relative to the credit value of an online module to ensure productivity without over-burdening students.


Please enter your comment!
Please enter your name here