The development and use of open-access, open-source material in mathematical sciences

By Bill Foster. e-Learning Team of the School of Maths & Stats, Newcastle University

This blog is on the use of e-assessment in all its forms for mathematics and statistics (mathematical sciences) at HE.

In a recent presentation at an IMA conference on barriers to learning in mathematics I recalled an earlier question I asked at yet another conference in 2011:

Why hasn’t HE with all its embedded technological and pedagogical expertise managed to organise a sustainable community-wide multi-platform system for mathematics and statistics with the required functionality and the concomitant savings and interoperability/dissemination via a database and good authoring system?

Of course I knew at the time that we were in the process of developing such a system at Newcastle University, so the statement was a tad self-serving!

Since then several such systems have been developed in the UK to the point of maturity in that they have well-established histories of use in HE institutions, not only in the UK but internationally. The US system WeBWorK is also widely used.

These include:

• DEWIS, developed at UWE used in several UK institutions.
• Numbas, developed at Newcastle University with a wide range of use nationally and internationally.
• STACK, developed at the University of Birmingham with contributions from the Open University and Aalto University. A wide range of use, nationally and internationally and is extensively used by the Open University. STACK is a Moodle quiz type.

The following questions immediately arise:

• Why has all this effort been expended in writing these systems?
• Don’t they all cover the same ground and use similar techniques?
• Who can trust a system written by academics not subject to commercial constraints?
• What advantages do they have over commercial systems, other than no upfront costs?
• What about sustainability? The history of the development of e-assessment systems in mathematical sciences in HE is littered with examples of failed or unused projects usually funded by some well-meaning but sometimes misinformed public funding body.
• What responsible educational institution is going to use such systems when they are subject to the selling pressure of commercial vendors? Most feel more comfortable in dealing with such vendors as some sort of support is built-in and there is a well-defined contract.
• Do such open-source systems have the necessary administrative tools to handle data-flow securely? It is noticeable that commercial systems emphasise their data-handling and reporting facilities and once again this may be an important factor in making choices of systems.
• Etc.

The aim of this blog is to stimulate discussion on these and other questions as there is growing momentum over the last 5 or so years in the development and use of open-source material. There is every likelihood that this will continue as course teams/directors, lecturers and administrators appreciate that not only do these open-source systems have equal and often better functionality, but also give local control back to the institution, both in pedagogy and administration.

However there are still significant barriers to the acceptance of open-source systems, especially at the higher levels of decision-making in HE institutions. At present the move towards their acceptance is “bottom-up”; departments and Schools adopt the systems as they are seen as being produced by the community for the community together with all the advantages in terms of transmitting and sharing good practice. This “community-spirit” within a discipline is not I believe appreciated at higher levels.

• How can this be transmitted and viewed as a good practice within the institution?

It is essential that any system used in the mathematical sciences allows the input of mathematical expressions together with the checking of such input and with good quality and timely feedback. However, there are many systems used in HE for e-assessment which do not meet thes
e requirements but have been adopted at the institution level which cannot be used for valid mathematical or statistical testing or learning. So another question that arises is:

• How do we signal the unique nature of e-assessment for mathematical sciences to those making decisions?