Led by Rob Wyn Jones, consultant and Shri Footring, senior co-design manager - enterprise, both Jisc.
With contribution from Rebecca Davies, pro vice-chancellor and chief operating officer, Aberystwyth University.
Connect more in Wales, 7 July 2016
2. Session outline
»Introductions
»Overview of the Learning Analytics Service
»Next steps / get involved
»The user voice – Prifysgol Aberystwyth University
»Overview of the Business Intelligence project
»Next steps / get involved
7/8/2016 2
4. What do we mean by Learning Analytics?
»The application of big data techniques such as
machine based learning and data mining to help
learners and institutions meet their goals:
»For our project:
› Improve retention (current project)
› Improve attainment (current project)
› Improve employability (future project)
› Personalised learning (future project)
5. Effective Learning Analytics Challenge
Rationale
» Universities and colleges wanted help to get started and have access to a
standard set of tools and technologies to monitor and intervene.
Priorities identified
» Code of Practice on legal and ethical issues
» Develop a basic learning analytics service including an app for students
» Provide a network to share knowledge and experience
Timescale
» 2015-16 -Test and develop the tools and metrics
» 2016-17 -Transition to service (Freemium)
» Sept 2017 – Launch. Measure impact on retention and achievement
6. Toolkit and Community
»Blog: http://analytics.jiscinvolve.org
»Reports
› Code of Practice for Learning Analytics
› The current state of play in UK higher
and further education
› Learning Analytics in Higher Education:
A review of UK and international practice
»Mailing: analytics@jiscmail.ac.uk
»Network Meetings
8. Current Engagement
»Expressions of interest: 85
»Engaged in activity: 35
»Discovery to Sept 16: agreed (28), completed
(18), reported (17)
»Learning Analytics Pre-Implementation: (12)
»Learning Analytics Implementation: (7)
9. Technical Progression to date…
• Deployment of our LA Data ProcessingAgreement (DPA)
• Latest student data specification (UDD v1.2.4)
• https://github.com/jiscdev/analytics-udd
• 12-36 months of UDD (student) data + aligned Activity data (VLE,Attendance etc)
• OngoingTechnicalTrials: Learner RecordsWarehouse(s), Bb & MoodleVLE
plugin(s), UDD data validation/ APIs
• Predictive Model development – service pilots start Q4 2016
• BlackBoard Learn –VLE activity data plugin evaluation for historical data capture,
with Moodle equivalent
• Student App – Beta v1.0 due for release July 2016 (iOS/Android)
• Student App evaluations – currently being formulated for 3 HEIs
11. Future Engagement/ get involved
From Sept 2016
»“ReadinessToolkit” with a diagnostic set of
questions and support materials leading to
implementation.
»Start-up guidelines to get ready for learning
implementation.
Further details will be announced via
analytics @jiscmail.ac.uk
12. The value of Learning Analytics
LearningAnalytics at Aberystwyth University
13. Monitoring Attendance
»Recording attendance prior to 2014 involved registers
and manual entry onto SAMS (in-house Student
Attendance Monitoring System)
»SAMS simply recorded attendance and produced
limited reports on an individual student’s attendance
»The Computer Science department developed and first
ran MOPS (Monitoring of Performance System) 2013/14
› MOPS used data from SAMS to produced reports of
students with poor attendance
› MOPS then managed intervention workflows (typically
meetings) and recorded outcomes
7/8/2016 13
14. Initial Focus
»Initial focus was on
› Monitoring attendance - seen as the key indicator of
engagement
› Enabling early intervention
› Focus on student retention
»Systems were developed in-house (SAMS, MOPS)
»An accurate personalizedTimetable was an essential
pre-requisite of this work
7/8/2016 14
15. Automatic Attendance Monitoring
»But - high manual overhead of recording attendance
»Investigated card readers. Problem = high unit cost
(£500+)
»2014 Computer Science dept developed prototype
proximity card reader
»2015 CS worked with AU IS to turn this into production
»2015-16 Deployed across core teaching rooms
»By Sept 2016 will be deployed across all teaching
rooms
7/8/2016 15
16. Attendance Stats (up to end Semester 1 2016)
7/8/2016 16
2015 2016
Semester1 Semester 2 Semester 1 Semester 2
Auto = collected via card swipe. Manual = entered via SAMS. Auto Manual Auto Manual Auto Manual Auto Manual
Attendance recorded - 285,963 - 203,228 343,815 48,437 - -
Not attended - 85,811 - 100,238 108,095 14,099 - -
Statistic Details
17. Attendance Stats
»More data now collected on attendance & less work
»reporting on attendance at the student, module,
scheme, and department level – new insights
»Also being used for
› Monitoring international students to ensure compliance
withTier 4 visas
› Monitor exam attendance
› Time and attendance to record some employee working
hours - replacing time-consuming timesheets.
7/8/2016 17
19. 3 Pillars to Successful Implementation
»A University-wide attendance policy with the
expectation that students should attend all timetabled
sessions (introduced 16/17).
»Universal automated attendance monitoring points
to capture attendance data with minimal manual
intervention.
»Systems to monitor and report attendance, alongside
managing intervention workflows.
7/8/2016 19
21. Why the JISC Learning Analytics Project?
»Investigated commercial suppliers. Issues with
specification and price
»JISC solution designed by the sector and for the sector
»Main motivators:
› Continuous quality improvement
› Improve retention (reduce drop out, improve completion)
› Improve the student experience / satisfaction
› Improve student outcomes (degree class)
› Personalization / Student engagement with their learning
7/8/2016 21
22. Work with JISC Project so far
»Providing data from
› Student record system – in-house ASTRA
› VLE – Blackboard
› Attendance monitoring – from SAMS
»Feeding in to JISC’s UDD data specification
»Plan to pilot the JISC tools 16/17 & provide feedback:
› Test predictive model
› Unicon Learning analytics processor and dashboards
› Student Success Plan (intervention management tool)
› Student Learning Analytics App
7/8/2016 22
23. Other Plans 2016/17
»The following are being deployed University-wide from
September 2016:
› Attendance monitoring points in all teaching locations
› Consistent attendance policy
› New personal tutor system – incorporating use of analytics
in discussions
› New version of MOPS – “Tutor dashboard” showing
attendance &VLE use
› Display of attendance data in the Student Record system
(web-based app) – see next page
7/8/2016 23
25. Future Considerations
› Post Brexit
– will we need more formal monitoring for more of our
students?
– Data Protection legislation = EU law
› HE Bill in England – delay?What does this mean for the
metrics we need?
› Relationships with Student Unions / Representatives – will
they continue to support or does political uncertainty =
uncertainty in their support?
7/8/2016 25
30. Heidi Plus
The new business intelligence service for UK Higher Education
Replaces Heidi (which will be decommissioned in November 2016)
Launched in November 2015 offering:
Improved data content and functionality
Delivery of data sets through commercial data explorer tool
New visualisations and dashboards
New training programme and support materials
Available to HE institutions with a full HESA subscription
Over 80% of current Heidi subscribers have started the Heidi Plus
application process (40% completed)
32. Secure data processing environment
Technical infrastructure bound by legal agreements to ensure data and dashboards are secure
33. Information improvement manager UEL with;
Kent, Middlesex, Brunel, Royal Holloway
Strategic planning and BI manager Sunderland with;
Glasgow, Glasgow Caledonian, St Andrews, Sunderland
Director of planning, Kent with;
Birkbeck,Cardiff, Oxford, Southampton, Southampton
Strategic Planning Manager, MMU with;
Leicester, Leicester,Cambridge, Bishop Grosseteste
Winter teams
34. Upskilling of staff resource across sector
Opening up of collaborative relationships
across other organisations
Value, saving and efficiency gains from the
creation and delivery but also the actions
subsequently taken due to the insights gained
across research, student, staff and estates and
possibly internationally
Opening up access to disparate data sets and
making sense of them in an HE context
Possible national licensing deals for paid
access to data
Team member experiences
35. Team Laura – Q & A
Claire Daniells – University of Plymouth
Frances Leach – MMU
Natalie Butler - Leeds Beckett
NicolaWitts – MMU
Rhodri Rowlands – Sheffield Hallam
Shri Footring – JISC
ScottWilson – JISC
Laura Knox – St Andrews
36. User Stories
I want to: Understand the
destinations of my students post-
graduation (in particular further
study and employment)
So that I can: ensure the credibility
and sustainability of our curriculum
I want to: understand the
demographics of students
who progress on to further
study
So that I can: better
understand the quality and
demography of students
applying to PGT level study
at my or competitor
institutions
I want to: understand the
geographical locations of my
graduating students who enter
employment
So that I can: ensure the curriculum
is adding value and is credible in the
context of the relevant labour
market (local/national etc).
I want to: understand the gaps in
the labour market (local,
national and international)
So that I can: ensure the
curriculum is adding value and is
credible in the context of the
relevant labour market
(local/national etc).
37. White Paper andTEF
Graduate employment
Highly skilled employment
LEO dataset
New DHLE
Considering the need to “understand graduate migration
in greater depth, including the wider social impacts of
graduates and travel to work patterns”.
Linked data
Additional Context
38. DHLE Data
NOMIS (Official Labour Market Statistics)
Possible addition:
IPPR Burning Glass -Wheretheworkis.org
Association of Graduate Recruiters
Data Sources
39.
40.
41.
42.
43. Dashboard: Course Market Analysis for Institutions
What is it? An Overview Movie
Purpose:
This dashboard is designed to support a university’s strategic planner in
designing course by allowing comparison across the sector.
Use case:
As a Strategic planner when working out which courses to teach I want
to examine competition to my course offerings to ensure I target
recruitment activity most effectively.
Data sources:
National Pupil Database: http://bit.ly/224CU8I
Key Information Sets: http://bit.ly/1ZYnG5z
National Pupil Database: http://bit.ly/224CU8I
HESA Data
What needs to be done and issues Time and Effort to Market
Where there is scope for improvement:
• Generally very polished
• Some work on the interface required perhaps to sign-post the features
• Licencing issues for league table data need to be negotiated.
• Data sources would need updating each year – particularly the school data.
44. Dashboard: University Finder for Students
What is it? An Overview Movie
Purpose:
This dashboard is targeted at students who are looking for a university course to
fit their needs. By needs we don't only mean course but also: cost,
employability, location and entry tariff.
Use case:
As a student when working out which university course offers best fit my needs,
I want to understand factors of relevance to me (course, cost, employability,
location, cost of living, rural/urban and entry tariff) to compare and match
offers to my circumstances.
Data Sources:
Key Information Sets: http://bit.ly/1ZYnG5z
HESA Data
What needs to be done and issues Time and Effort to Market
This dashboard supplies a unique perspective on data and services that are already available to
students. In some ways this is a crowded marked. So the unique selling point of this product
would need to be promoted – that is that the data already available to students is amalgamated
and drawn together to create a” wizard like app” for students to find courses.
What would need to be done:
• Identify appropriate vehicle for delivery
• Market uniqueness of the the product
• Negotiate data licences for league table data
45. Dashboard: Finding Comparable Institutions
What is it? An Overview Movie
Purpose:
This dashboard can be used to identify a university’s relative performance
against a benchmark of similar institutions.
Use case:
As a Planning Manager I want to select similar institutions based on metrics I
choose so that I can determine the best institutions to compare with my own
university to understand if our performance is relatively good or bad
Data Sources:
HESA data from Heidi
Key Information Sets: http://bit.ly/1ZYnG5z
League Table Data – will require licensing
What needs to be done and issues Time and Effort to Market
Where there is scope for improvement:
• Data – a relatively narrow data set was used for prototyping; a production version could
accommodate a far more comprehensive data set.
• Filters – searching and filtering could be enhanced
• Licencing – Makes use of some league table data to benchmark against entry tariff.
Licence for this need to be negotiated.
46. Library Data Labs
Teams working on Library BI Stories at 0.2 FTE, total estimated
effort 15 days from July - Oct 2016
Both Product Owners and Sector Data Experts invited:
Product Owner from the sector to steer which stories are of interest
Sector Experts to understand what data sources are available & what is in the data
Jisc Contracted Data transformation specialist
Jisc Agile Scrum Master &Tableau User
Teams receive experience and guidance of Agile working
Option forTableau Desktop training to help with creating
visualisations
7/8/2016 46
47. Analytics academy – a Jisc beta service - October 2016
» Business intelligence offers value, savings and efficiencies to
Universities through data informed enhanced planning / decision
making
» Many problem spaces are commonly felt, while the data landscape to
support insights is vast.
» Some universities have little access to good BI at all, while those with
capability are often duplicating effort.
» There is no higher education focused CPD offer to train up BI expertise.
» Analytics academy addresses these problems by providing expertise
and tools for analysts (planning officers and others) to identify suitable
problem areas (student, staff, research, estates etc), exploring the data
landscape for insights and producing interactive dashboards for the
sector
7/8/2016 47
Myles 10.30 – 10.35
Overview of LA service – Shri (5 mins)
User voice – David Matthews (10 mins)
Overview of BI - Myles (5 mins)
The user voice - James (10 mins)
Group exercise - Shri (20 mins)
What’s coming next - Shri and Myles (5 mins)
Shri – 10.35 – 10.40
What do we mean by learning analytics. The service we are developing will collect data and undertake statistical analysis of historical and current data derived from the learning process to create models that allow for predictions that can be used to improve learning outcomes.
Models are developed by “mining” large amounts of data to find hidden patterns that correlate to specific outcomes
E.g. Mine VLE event data to find usage patterns that correlate to course grades
The service will provide predictive models initially for retention (identify students at risk of failing) and attainment (identifying students at risk of not achieving a specified level of attainment).
In the future we will look to offer predictive models to support employability and personal/adaptive learning.
Shri
The effective learning analytics challenge was initiated from consultation with stakeholders, senior manager and practitioners who felt the sector need support to get up to speed with learning analytics. They prioritised three main areas, a Code of Practice to address legal and ethical issues of using learning analytics; a set of basic learning analytics tools to allow institutions to get started and make informed decisions; and a network to allow institutions to share practice and learn from each other.
The current project has procured suppliers to provide a learning analytics service which are currently being tested by several institutions. This will be developed into a full service next year and provided as a new Jisc service from Sept 2017.
The project consists of the learning analytics architecture (next slide), a toolkit and community.
These consist of a blog with reports and information to assist institutions with readiness to implement learning analytics and technical implementation of the Jisc service.
There are three reports all linked from the blog a Code of Practice for Learning Analytics, A report from 18 months ago that reviewed current state of learning analytics in the UK and a more recent report on the evidence base for the effectiveness of learning analytics with 12 international case studies.
If you want to be involved and keep informed about the development of the service then join the analytics jiscmail list
We also hold quarterly network meetings which are promoted via the blog and jiscmail list
Overview of learning analytics architecture.
Red items are components that will include the tools in the project (Tribal student insight, Unicon/Apereo LAP and Student Success Plan, Student App) but also alternative third party or institutional tools.
We have ~400 people on the Jiscmail list and a pipeline of interested institution's (50+ HE, 20+FE). We are actively engaging with 35 institutions, 28 in discovery institutional readiness and 12 in beta implementations.
Overview of learning analytics architecture.
Red items are components that will include the tools in the project (Tribal student insight, Unicon/Apereo LAP and Student Success Plan, Student App) but also alternative third party or institutional tools.
From Sept 16 we’ll be introducing a new institutional readiness process to help institutions get ready for implementing learning analytics. This will consist of an overview workshop to introduce the service and an diagnostic assessment tool, institutions will complete the assessment tool and then undertake appropriate actions to address recommendations.
For institutions who are ready to start implementation there will be set of guidelines to get set-up with data collection and visualisations, ready to implement a predictive analytics solution and the student app.
Details will be announced via the jiscmail list – so join it to participate.
From Sept 16 we’ll be introducing a new institutional readiness process to help institutions get ready for implementing learning analytics. This will consist of an overview workshop to introduce the service and an diagnostic assessment tool, institutions will complete the assessment tool and then undertake appropriate actions to address recommendations.
For institutions who are ready to start implementation there will be set of guidelines to get set-up with data collection and visualisations, ready to implement a predictive analytics solution and the student app.
Details will be announced via the jiscmail list – so join it to participate.
Myles 10.50 – 10.55
Jisc and HESA are collaborating to develop new national shared services for business intelligence, making better use of the national data landscape, reducing repetitive activities across universities, brining the benefits of BI to all Univerisits regardless of capability / expertise
Myles
HESA is a not for profit subscription organisation, so similar to Jisc in that sense. As well as a mandatory subscription, members are mandated to provide data collections covering the broad themes of Student, Staff, Destinations (of graduates) and Estates data. This is annual but in year collection is under consideration. HESA cleanse the data and provide back full data sets, published statistics and undertake bespoke analysis. Jisc and HESA membership is similar.
Myles
Heidi Plus is depicted on the left – highlight the trucks driving in to the HESA data warehouse. HESA mandates that all publicly funded HEPs provide performance data on students, destinations of leavers, staff, finance and estates. Currently an annual collection they are moving to more frequent in year collections. The data is cleansed and a new team undertake dashboard development. Quality is assured as the dashboards are offered throught the radio mast in the middle – a new national BI dashboard delivery service offered to all HESA customers (currently 180 HEPs and associated organisations and departments). Built with Jisc and launched as a HESA service in November. Includes legal framework and national training programme. Replaces a system with 6.5K users. Lowers the bar to usage through the interactive dashboards so could take BI to a woder range of staff than is currently possible.
Heidi Lab is depicted to the right. A Jisc led alpha July 15 – July 16. Highlight the trucks again and note it’s a two way street – a data sharing agreement allows HESA data into the Heidi Lab secure data processing environment. Agile analysis teams are created from multiple universities and given access. They identify commonly felt problems spaces, explore the wider national data landscape, acquire non-HESA data and cleanse, link and transform it creating new proof of concept dashboards. Highlight the trucksa driving from the Lab to the Radio Mast. Successful dashboards will be branbded produced by Jisc and delivered via Heidi Plus.
Piece in the middle is the beta service – what comes next – Heidi Plus is sustained by HESA as a service. We have proved there is real merit in Heidi Labs and will launch a beta service July 16 – July 17.
James 10.55 – 11.00
HESA’s current data delivery service is known as HEIDI (Higher Education Information Database for Institutions) developed in house in 2007. Jisc and HESA collaborated to replace this with a more up to date service. We procured Tableau, market leading data exploration software and now offer Heidi Plus
Feedback has been extremely good across the sector
Myles 11.00 – 11.05
Heidi Lab as a Jisc Alpha project (proof of concept) engaged with 290 individuals from 130 universities to develop a successful model of agile analysis. 50 analysts (planners, directors of planning from 44 universities volunteered to join cross institutional agile analysis teams for three Heidi Lab cycles of 3 months each at just 0.2 FTE. Teams were supported as they identified and refined widely felt problem areas (see example on the slide – covered student, staff, research, estates etc) linked to national policy. They explored the data landscape for supportive insights, recording the issues encountered in our data catalogue. Finally they produced interactive dashboards using Tableau software as proofs of concept to offer through Heidi Plus
Led by a senior staff member with knowledge of the information needs of a wide range of staff and institutions as well as national policy and what is ‘up stream’
James – 11.05 – 11.15
13.00 – 13.20
There were more but these give a high level overview
In addition to general state of play for HE in terms of student perceptions of value for money, increased competition and squeezed public funding…..
White paper: LEO 'we hope this will also be used by providers evaluating their provision and considering how they can tailor it to better deliver relevant skills for the labour market'
New DLHE: understand graduate migration in greater depth, including the wider social impacts of graduates and travel to work patterns. And 'there may be linked data options that could be explored to obtain additional depth and quality of information, while minimising the costs of data acquisition. This might include utilising a geospatial data system such as the Unique Property Reference Number (UKPRN) to derive contextual information about the location where a graduate is living or working'.
Give overview of content of each…..
DLHE
NOMIS employment by sectors (industry and geo)
WTWI: IPPR Burning glass data set (LEP, SOC, employability potential/ave salary)
Explain potential opportunity with ARG for HESA to pick up: aggregate vacancies and hires sliced by career areas, industry types, employer size, and applicant numbers per role.
Note/make plea here in relation to lack of coherent/comprehensive data set
Demand data
Burning Glass has collected more than 1.5 million jobs posted online by employers in the UK since 2012. Burning Glass uses advanced natural language analytics to turn the information in each job posting into usable data. This allows Burning Glass to describe employer demand for specific roles or skills.
The demand for entry-level (< 2 years of experience) talent is compared with the available supply of new graduates or trainees. Burning Glass postings data is normalised against vacancy data published by the Office for National Statistics (ONS) and Jobcentre Plus. The data is further validated against the Annual Survey of Hours and Earnings (ASHE) from the ONS.
Supply data
We have used the numbers of learners leaving higher and further education (programme finishers by subject area) as a proxy for the ‘supply’ of entry-level talent.
Supply data are sourced from the following agencies:
Higher Education Statistics Agency (UK wide)
Skills Funding Agency (England)
Scottish Funding Council
Skills Development Scotland
Department for Employment and Learning of Northern Ireland
StatsWales
Occupations
We use the standard occupational classification 2010 (SOC2010) by the ONS, which is the official classification of occupational information for the UK. Within this system, jobs are classified in terms of their skill level and skill content. In this tool, the occupations are shown at ‘minor group’ or 3-digit level.
James to continue with these (as many as time permits)
James to continue with these (as many as time permits)
James to continue with these (as many as time permits)
Myles – just to note we are running a set of teams from the library area to prove the concept transfers
Myles – a new Jisc offer to explode whether there is a sustainable service in this