There has been phenomenal growth in government data on the internet, with an estimated more than a million datasets posted online by governments. The boom has major implications for university governance – but different types of data have varying degrees of impact on diverse levels of governance.
The open data boom – driven by advocacy groups, and public funding agencies, as well as hackers and developers – opens up opportunities for improving the transparency of higher education governance and providing evidence to inform policy, says Francois van Schalkwyk, a researcher at the Centre for Higher Education Transformation, or CHET, in Cape Town.
Van Schalkwyk, in a presentation drawing on in-progress research into open data usage in South African higher education governance governance, was speaking at a Carnegie Corporation of New York-sponsored convening in Nairobi on “Higher Education Policy, Leadership and Governance”, before an audience of African vice-chancellors, heads of commissions and think-tanks. (See the Open Data in Developing Countries Project website.)
The objectives of the convening were to share research, practices and findings on how grantees are influencing higher education policy and leadership, to identify gaps, develop a common vision and map the future.
Van Schalkwyk said little was known about the actual uses and impacts of open data, especially in the developing world. Research tended to focus on open data supply-side issues such as technical platforms and standards, and to be based on cases from the developed world.
Open data in South African HE
South Africa’s public university system of 23 institutions can draw on two large datasets, the first belonging to the Department of Higher Education and Training, or DHET, the other produced by CHET.
Paradoxically, both draw from the same source – the Higher Education Management Information System, or HEMIS, to which universities are required by the DHET to submit data. HEMIS contains student and staff records, and direct public access is restricted for privacy reasons, since the records contain personal data.
This ‘apparent duplication’ of the datasets is historical. CHET started providing data at a time when DHET was not yet making HEMIS data publicly available in a consistent and discoverable manner, and it has persisted because the two providers are ostensibly responding to different demands from ‘governance consumers’.
It is possible that the DHET supply of data is driven by the South African government’s commitment to public transparency and accountability.
The South African government is a founding member of the Open Government Partnership launched in 2011 ‘that aims to secure concrete commitments from governments to promote transparency, empower citizens, fight corruption, and harness new technologies to strengthen governance’. Greater accountability of actors in decision-making and in the application of rules, are all to the public good.
CHET’s provision of data is driven by the need of universities to meet governance targets set by the DHET, using evidence-based decision-making. It is especially targeted at university councils.
CHET goes further, beyond ideology, theorising that university governance is best improved by informed data-driven decision-making.
As a result, the DHET open dataset is made available as primary (anonymised) data, while CHET provides shaped open data that can be used to develop performance indicators that can be used in comparative analysis.
Consequently the CHET open data allow the generation of graphs on the fly, annual updates across some 20 indicators, and comparisons between up to four institutions – and the data are downloadable as spreadsheets or as images.
Van Schalkwyk’s preliminary findings, articulated in an upcoming paper, are that different types of open data are directed at different ‘governance consumers’, and that the format of open data varies along a continuum from primary to shaped “depending on how the supplier of the open data perceives change to take place by means of open data provision”.
The implication of this for governance is that different types of open data supply have, firstly, varying degrees of impact on governance and-or, secondly, “may impact at different levels in the governance ecosystem”.
Regardless of the differences in specific configuration, Van Schalkwyk told the Nairobi convening, open data came with the promise of improved governance at systemic as well as institutional level, greater research efficiency and the bonus of ‘new, unexpected’ knowledge.
It was also increasingly a requirement of funders.
Receive UWN's free weekly e-newsletters