Tuesday, April 18, 2017

Cassells is wrong! University can be cheaper!

This is a slightly longer version of an article of mine in the Irish Times today: http://www.irishtimes.com/news/education/technology-is-key-to-reducing-college-education-costs-1.3045100 - it relates to a panel debate that I will be taking part in on the Cassells Report at the HECA conference on the 27th April, 2017 - http://www.heca.ie/heca-conference-2017/

In twenty years time, much fewer school leavers will be going to college. This will be simply because there will be so many more options available to them. Many of the options will be based on technology and the Internet and will be attractive for many reasons, not least that they will be much less expensive for the school-leaver. The strongest economies in Europe already have a balanced academic/vocational delivery model and England is following with an apprenticeship levy. This rebalancing of the academic and vocational is happening worldwide, including Ireland. As well as the increased availability of apprenticeship courses in many more fields, colleges will deliver more of their own courses in an apprentice style or work-based learning mode and these will include a significant amount of online learning. These will reduce costs for learners by allowing them to live at home for longer and earn while they complete their courses. Much more, if not virtually all, university courses will become available in distance-learning modes over the Internet and this will allow school leavers to go straight into jobs and gain their qualifications more gradually, even taking more time to select the right course. In addition to cutting the cost of their education through working and living at home, fees will decrease simply because the number of people simultaneously taking these courses will reduce unit costs. You can, right now, get a four-year degree in Computer Science from the University of the People, an accredited online university for $4,000. Indeed the claim that students need personal interaction with their lecturers and peers will be counteracted by the claims from employers that full-time students are not prepared for the workforce and that those who work and learn at the same time, learn more efficiently and are much better prepared for the world of work. To top it all, employers may become less interested in degrees. Even now, major employers are announcing that they no longer require degrees from recruits where they once did. This is because learners now have many more ways, including free courses on the Internet, to gain knowledge and competencies, and employers are becoming better at assessing these competencies before hiring.

In March of 2016, the expert group on future funding for higher education released the report "Investing In National Ambition: A Strategy For Funding Higher Education", also known as the "Cassells Report" after Peter Cassells who chaired the expert group. As an invited participant to one of the consultative meetings informing the report, I was intrigued that as we discussed possible funding mechanisms for higher education, nobody seemed interested in discussing the possibility of reducing the cost of providing higher education. To some extent, I accepted that this was probably beyond the brief of the expert group. However, the report that emerged did briefly mention the suggestion that information technology might reduce the cost of higher education, but quickly dismissed the idea, suggesting that it is not a "quick fix". This was despite having a statement in the executive summary that the purpose of the report was to "consider issues related to long-term sustainable funding of higher education".

Now rather than just disagree with the statement that technology cannot be an immediate solution to the funding problem (which is arguably false), I would like to suggest that the report is significantly flawed in addressing its objectives by not considering how costs can be reduced, in the medium term if not the short term, and particularly through the use of information technology. Information technology is revolutionising almost all information and communication businesses so why should it not have a significant impact in higher education? George Bernard Shaw once suggested that all professions are "conspiracies against the laity". Now, I don't want to accuse the higher education community, (of which I am a member of and in which I have many admirable colleagues), of conspiracy, but there is a natural tendency to support the system you are part of, even when times have changed and there may be less need for it. So it is not surprising that if you gather together a consultative group of higher education professionals they will tend to tell you they need more rather than less money.

However, there may be another, less self-serving, explanation for the seeming lack of interest in cost-cutting. Perhaps the expert group did not have the full range of expertise required to consider this option. One clue that this might be the case, is a statement in the report that the recent reduction in funding for higher education is resulting in less time to "accommodate diverse learning styles". It is well known in the educational research community for some time that "learning styles" do not exist, as such, and the idea that improved learning outcomes can be achieved by facilitating diverse "learning preferences" is essentially a myth. The issue here is not that the report might advocate for an expensive ineffective teaching method, but that such a well-known educational myth slipping into the report suggests that the group might not have had the full range of expertise that it required.

As it happens, my personal opinion is that the group has recommended the best of the three funding options available to it, namely student loans. However, because public institutions do seem to have an insatiable appetite for funding, it still could end badly if there are no attempts made to reduce unit costs. In the U.S. there are a large number of people who cannot afford to repay their loans, either because they never completed their studies or because the qualifications they earned have not significantly improved their earning capacity. This is a significant cause of personal suffering for many and also considered to be a significant risk to the economy as a whole. For that reason, even if we do improve access to student loans, it is important to try to reduce costs in order to minimise the level of debt incurred by students.

Although I have suggested that a major mechanism for reducing costs in the future may involve having more students studying off campus in distance learning modes while working, many of the same principles can be used immediately to reduce on-campus costs. Campus students can be given access to online modules designed for distance learners, or many of the free online courses on the web. Online modules can be created specifically to be shared by several campuses or colleges. The technology exists now to create such modules quite cheaply, for little more than the cost of delivering to a local group of students. If used in a "flipped classroom" mode, where the students interact with course materials online before attending tutorials or workshops on campus, not only can it reduce unit costs but also improve the learning experience of the students. Such shared online courses will necessarily have large numbers of students enrolled in order to gain economies of scale and so may need a certain amount of more personalised tuition added. However, even that can be made more efficient with tools that are available now. Computerised quizzes can provide feedback to learners on their progress and can be used to provide timely information to lecturers in order to see who most needs their help thus allowing them to use their time more effectively. The use of computer-aided peer assessment systems as well as rubric based assignment grading, often with standardised feedback, allows lecturers to get important timely feedback to large groups of students with much less effort.

Even more sophisticated tools based on artificial intelligence are now emerging that will allow lecturers provide a higher quality learning experience to larger numbers of students. Adaptive systems using deep learning techniques are able to analyse the behaviour and performance of large numbers of students using the system to determine what learning materials to present to individual students next as well as when to do so. Recently, as an experiment, in Georgia Tech university, a "chatbot" similar to those in computerised help systems, was added to the available human teaching assistants and was so successful that one student nominated it for a teaching award. Interestingly, some other students spotted that it was a chatbot because it responded so quickly to queries.

Spending more money, although often required on a temporary basis, is both the least ingenious and least sustainable way of solving a problem. Asking lecturers to work longer hours is not a very clever, or sustainable, way to improve productivity either. With technology, a small amount of ingenuity and possibly a significant amount of courage, we can improve quality, improve access and reduce costs in higher and further education in Ireland. And if we don't, someone else will.