I recently attended this event in London, which provide some great speakers, and useful networking opportunities, as well as showing what others are doing with HE data, and where we might want to do more. These notes taken at the event provide an insight for colleagues; I also have the slides from the presentations for those who want to look in more detail.
The event was opened by Sir Tim Wilson, former VC of University of Hertfordshire, who referenced the 2011 white paper which had asked to lessen the burden on information provision, but noted the level of complexity and diversity due to different providers in a more heterogeneous sector
Paul Greatrix (Registrar at University of Nottingham)
Paul introduced the ideas behind redesigning information landscape. He raised his concern about regulatory landscape also and government requirements. He identified that provision of more information does not necessarily mean better decision making
Scale of challenge for HEIs was the need to respond to 550 different external reporting requirements in addition to any internal reporting
In reference to league table providers, Dr Greatrix identified plenty if objections, but HEIs care because they have impact on potential students and the wider public, even though league table results can lead to perverse behaviours such as VCs and senior managers focusing on the wrong things.
In conclusion, he identified an uncertain future but with grounds for optimism. The fundamental issues were around regulation and the need for proper data, organised in the right way. It was not that there is not too little information re HE but that it needs to be underpinned with proper IAG for those with no previous family HE participation.
Malcolm Scott ( BIS Digital economy directorate)
Everyone is talking about data but there is nothing new about big data. The reason we a etalking about if now is due to the actual volume of data, massive increase in volume, growth of technology to sort data and bring it together, and the ability to get value out of it. Data can provide value to existing and new industries.
Government t has designated big data as one of 8 great technologies and has Looked at skills, infrastructure and hygiene factors that must be right to be able to exploit data. Major investments have already been made such as the Turing Institute and large km array telescope
Importantly he raised this issue: How do we get managers to realise data can make org better?
This can be alternatively expressed as the organisation will lose advantage of it doesn’t use data properly.
He suggested that on the supply side there is shortage of big data skills, expecting 13-23% rise in demand for big data staff in UK by 2017, noting that the people needed are not just computer scientists
Johnny Rich (push.co.uk)
Johnny pointed out that the information landscape for students is confusing and that Students don’t know what they need to know, Eg what’s it really like to study x at y?
As students go through jungle they will latch on to things that they recognise, eg league tables, names of courses. These may not actually be the useful things, as they will only spend a maximum of 1/3 of their time studying
He proposed that many information requirements are an unnecessary burden but a necessary evil. Sometimes data or information provided can be a part of marketing, but not what we want to know about- eg traffic light health guidelines on a sandwich wrapper. Regulation and the manufacturer might want this, but the customer isn’t likely to make a purchase decision on this. Is KIS is like this?
He proposed that KIS…
- ignores the information needs of the disenfranchised
- only tells them what they think they want to know
- is better than nothing
and thta the NSS was about:
- Satisfaction is not quality
- Enhancement not choice
The NSS however indirectly affects choice as it feeds into league tables.
He proposed a Marketing 101 approach – find your point of difference eg shampoo adverts, and start from there.
Graeme Wise (NUS assistant director policy)
Graeme introduced the policy context in which HE data is used and alignment of interests between data provides, collectors and users.
He was anticipating research on financial outcomes of HE from different institutions, using combined data from SLC, ,tax records as these linked data sets will provide model of earnings and outcomes from different unis and subjects which would be a driver for further marketisation
Looking at other public policy agenda- public sector data and fashion for analytics- he proposed the provocative thought of industries moving from production to service (eg software was previously a product, now a subscription service, eg Microsoft, media industries) and will education become that kind of industry, with a move from students as consumers to students as producers.
He looked at the ecology of data collection that spans student life cycle with the consolidation of data at output end, whereas an area for most attention is on the daily experience of students.
Fashion for analytics is not a passing fad and this is a challenge to sector – it is possible to get itvery wrongly not being there, or by doing it badly..
He proposed that in ideal world, everything collected for external purposes should be available for internal purposes, and need people with insight and experience as well as analysis skills to apply to real world student experience issues, eg retention, learning space utilisation, curricula. He suggested we involve student representatives to legitimise and shape the work
Phil Richards (JISC)
Phil had 2 main messages – Overcoming barriers to sharing, and making data a priority for senior management
The barriers to sharing were proposed under three headings: Co opetiton, Compulsion and Coherence.
Making data a priority for senior management came under 4 Rs:
- Reputation– league tables, Unistats, research metrics, Which? Guide,
- Recruitment and retention-this is an area for investment decision for software- for example the ability to identify at risk students either before they arrive or how they behave when on campus
- Risk mitigation – in a new HE paradigm we need detailed scenario planning (however who had considered removal of SNC in last year’s planning?)
Andy Youell (Director HEDIIP)
Andy started by considering the future of data and information, noting that many organisations were not designed to keep up with technology meaning that ad hoc solutions emerge, frequently in silos, which provide a sort term result only.
He asked- What is “here” like? There are Over 500 data collections which lead to duplication, inconsistency, lack of data sharing, lack of comparability across collection eg in sometjign simple such as the definitional difference between a course and a programme
Also highlighted were data management and governance issues,,eg security, quality, accessibility. There is often low awareness of where data is held in institution, low awareness of where is being supplied from and to whom.
The HEDIIP vision was one of new systems that reduce burden for data providers and improve quality, timeliness and accessibility of data and info about HE
The benefits are to: reduce the cost of data (duplication, inefficiencies); increase value of data (analytical capability, quality and timelines linking using standard identifiers), and improve information (clarity).
John Gledhill (Tribal – supplier of SITS)
John pointed out that we tend to sum up 3-4 yrs of education in snapshot data and that student data collection is low resolution and low frame rate currently. In future we might need to capture data that we think might be worthless, for instance working in areas of unstructured data eg Facebook, Twitter, RSS, as well as structured databases and file systems.
Steve Egan (HEFCE)
Steve talked of the need for accurate data definitions to protect those who want to play the game properly, but questioned how we can produce timely data, eg HESA? For example, for widening participation, the data is 2 years out of date, and this has implication for funding.
Since students make decisions on range of information some of which is influenced by data, then they need to be able to trust the data, for example claims for employability, noting the weaknesses of DLHE data.
Government also needs good data to be able to identify what is happening with part time students, SIV subjects and accountability. Better and more timely information will lead to better decision making
Summary by Sir Tim Wilson
Are we using external data internally?
Is data collection and analysis a cost or an investment?
Have to change because if we don’t it won’t get better. Some people enjoy being victims and complaining. Has to change to make things better for students and all stakeholders
Willingness to move to a common good, which is not the same as uniformity.
We have the power and knowledge to do data analysis which needs transformational leadership, vision and innovation.