Inside Government - London

8/7/2015 Jisc Learning Analytics

2

Jisc’s Learning Analytics Project

»About Jisc»Learning Analytics»Jisc’s Open Learning Analytics Project»Finding out more

3

About Jisc…

MissionTo enable people in higher education, further education and skills in the UK to perform at the forefront of international practice by exploiting fully the possibilities of modern digital empowerment, content and connectivity

VisionTo make the UK the most digitally advanced education and research nation in the world

What does Jisc do?

Does 4 things…

Providing and developing a network infrastructure and related services that

meet the needs of the UK research and

education communities

Supporting the procurement of

digital content for UK education and research

Our network of national and regional teams

provide local engagement, advice and support to help you get

the most out of our service offer

Our R&D work, paid for entirely by our major

funders, identifies emerging technologies

and develops them around your particular

needs

Co-design challengesResearch at risk

(R@R)

Prospect to alumnus (P2A)

Learning analytics

Digital learning & capabilities

Implementing FELTAG

Business intelligence

Hosting platform Hosting platform

7

About Learning Analytics…

8

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 achievement (current project)» Improve employability (current project)»Personalised learning (future project)

9

Jisc’s Learning Analytics Project

Three core strands:

Learning Analytics Service

Toolkit Community

Jisc Learning Analytics

10

Jisc’s Learning Analytics Service and Open Learning Analytics Architecture

Learning Analytics Service

11

Student App

Dashboards

Alert and Interventio

n

Structured Data

Machine based

learning

Learning Records Store

Transformations/Mining

About the student Activity Data

Consent

Data Collection

DataStorageand Analysis

Presentation and Action

Architecture overview

12

Our project partners

13

Presentation and Action Layer

14

Presentation and Action Layer Pt1Dashboards

Visual tools to allow lecturers, module leaders, senior staff and support staff to view:

 » Student engagement» Cohort comparisons» etc…

Based on either commercial tools from Tribal (Student Insight) or open source tools from Unicon/Marist (OpenDashBoard)

15

Presentation and Action Layer Pt2Student App

Specification still underdevelopment, but first version will include:

 »Overall engagement»Comparisons»Self declared data»Consent management

Bespoke development by Therapy Box

16

Presentation and Action Layer Pt 3Alert and Intervention System

Tools to allow management of interactions with students once risk has been identified:

» Case management» Intervention management» Data fed back into model» etc…

Based on open source tools from Unicon/Marist (Student Success Plan)

17

Data Storage and analysis layer

18

Open SourceCommercial

Predictive Analytics/Machine learning

Transformations and mining

Structured Data/

Business Intelligence

‘Big Data’ Learning Records Store Big Data

Data storage and analysis layer overview

Presentation layer

19

Data Collection Layer

Title of presentation 00/00/2013 20

Learning Records Store

About the student

Activity dataTinCan (xAPI)

ETL Student

Record System

LibraryVLE Other

s

Data collection layer overview

Data Collection

DataStorageand Analysis

21

Activity data via Tin Can API• People learn from interactions with

other people, content, and beyond.

• These actions can happen anywhere and signal an event where learning could occur.

• When an activity needs to be recorded, the application sends secure statements in the form of “Actor, verb, object” or “I did this” to the Learning Record Store (LRS.)

from: http://tincanapi.com/

22

Activity Data (TinCan) examples

Actor Action Object

Result

Michael Accessed VLE

Sally Completed Basic Maths Test

85.0

Kim Module Comment

Added

https://registry.tincanapi.com

23

Example as JSON code{ "actor": { "name": ”Michael", "mbox": "mailto:michael.webb@jisc.ac.uk" }, "verb": { "id": "http://adlnet.gov/expapi/verbs/accessed", "display": { "en-UK": ”accessed" } }, "object": { "id": "http://example.com/activities/vle", "definition": { "name": { "en-UK": ”VLE" } } …

Actor

Verb

Object

Title of presentation 00/00/2013 24

About the student’ data

Personal (demographic) data

Birthdate, gender etc.

Course data

mode of study, level etc.

Grade data

Assignment, module etc.

(aligned with HESA data)

25

Jisc/Unicon Discovery

Jisc Learning Analytics

Implementation

Wish to explore readiness and products

Know you are ready and what you want

Want to get involved in tech work first

Blackboard Discovery

Unicon/Marist pre-

implementationTribal pre-

implementation

Other pre-implementation

Blackboard Trial

Moodle Trial

Other Learning Analytics

Implementation

Tech Trials Discovery Pre-implementation

Implementation

26

Jisc Learning Analytics Toolkit

Toolkit

http://www.jisc.ac.uk/guides/code-of-practice-for-learning-analytics

First Publication: Code of Practice

Deeper Dive

http://repository.jisc.ac.uk/5661/1/Learning_Analytics_A-_Literature_Review.pdf

Literature review – basis for the code of practice

Code of Practice

Privacy

Validity

Responsibility

AccessEnabling positive

interventions

Minimising adverse impacts

Transparency and consent

30

Community

Community

31

Project Blog, mailing list and network events

Blog: http://analytics.jiscinvolve.org

Mailing: analytics@jiscmail.com:

Next event: Bradford Oct 2015

michael.webb@jisc.ac.uk

One CastleparkTower HillBristolBS2 0JAT 020 3697 5800

info@jisc.ac.ukjisc.ac.uk

Michael WebbDirector of Technology and Analytics