3. Agenda
— Introduction to UNSW and theAustralian School of
Business
— Description of the business problem
— 4 step methodology
4. University of New South Wales
— UNSW
— Formed in 1949
— More that 50,000 students
— Member of the Group of Eight (Go8)
— Ranked 52 in the QSWorld University Rankings
https://www.unsw.edu.au/sites/default/files/documents/UNSW4009_Miniguide_2012_AW2_V2.pdf
5. University of New South Wales
— Australian School of Business
— Over 12,000 students
— Currently ranked 12th in the world for Accounting
and Finance degrees
— Top ranking MBA inAustralia
— MBA ranked 48th in the world
http://en.wikipedia.org/wiki/University_of_NSW
6. Julia Enterprise Data Warehouse
— Developed by the Institutional Analysis and Reporting
Office at UNSW with support from UNSW IT
— SAS was installed at UNSW in 2004 as a proof of
concept
— 2009 Migrated from SAS 9.1 to 9.2
— 2010 Julia in its current form commenced in SAS
Enterprise BI Server using Kimball dimensional
modeling techniques
— Flagged for replacement by an EDW being developed by
UNSW IT using SAS Enterprise BI Server 9.4
7. Business Problem
— Identification of students potentially at risk
— Widespread, automated and earlier student
advisement related to engagement and performance
— Student engagement in courses via Learning
Management System (LMS) access and activity
— How do you identify two or three hundred students
out of 12000 needing support?
— Students are often shy in asking for help
8. Methodology
Step 2 –
Analyse for
Churn or
Risk Patterns
Step 3 – Build
a Repeatable
Model
Step 4 –
Apply the
Model
Step 1 –
Obtain Good
Customer
Data
SAS
Enterprise
Guide®
Star Schemas
in SAS BI
Suite
SAS BI Suite
9. Step One: Obtain Good Customer Data /
Build a Good Data Warehouse
10. Step Two: Analyse for Churn or Risk
Patterns
Convention in the sector
— Low Social Economic Standing
— LowATAR (AustralianTertiaryAdmissions Rank)
— Students with a lowerWAM (WeightedAverage Mark)
Are much more likely to drop out
11. Step Two: Analyse for Churn or Risk
Patterns
— What is Risk?
— LowWAM
— Churn (Dropping out of UNSW)
— A number of variables were investigated for Churn and
WAM using SAS Enterprise Guide
12. Step Two: Analyse for Churn or Risk
Patterns
— Variables investigated for Churn andWAM using SAS
Enterprise Guide
— AdmittanceType - Cross Institutional, Exchange Student, Foundation Studies UNSW, FirstYear Student, Internal Program
Transfers, Readmit to Program etc.
— Application Method - Direct or UniversityAdmissions Centre
— Social Economic Standing by Postcode – Based onABS data
— Gender – Retention andWAM comparisons
— Language Spoken at Home
— High School Math - Subject and Grades
— Parental Education Level
— English Language Proficiency for international students
— Residency group – Local or international
— Students in a program that was not their first choice
— Blackboard and Moodle Usage – Learning Management System
— Moodle grades
— Age as the start of program
— Subjects Failed (tested against churn only)
— WAM falling (tested against churn only)
13. Step Two: Analyse for Churn or Risk
Patterns
Highest Parental
Education Level vs.
Retention
15. Step Three: Build a Repeatable Model
— Decided on three groups of attributes:
— Current LearningActivities – Given the most weight
— LMS Exam Result Rate
— LMSAccess Rate
— University Study History – Given the second most weight
— FailedThis Course Before
— Course Fails
— WAM Drop Level
— WAM Level
— University Entry Ranks – Given the least weight
— ATAR Score
— High School Math Proficiency
— Ranked Entry Score
— Written English Proficiency
— Total English Proficiency
16. Step Three: Build a Repeatable Model
— We built a linear model fairly
simple, able to be explained (one of
the goals)
— Ultimate would be to have multiple
models and evolve them over time
and potentially select students who
show up in the models
— We still don’t KNOW what is
happening in the student’s life
PredictiveAnalytics:The Power to PredictWhoWill
Click, Buy, Lie, or Die by Eric Siege
17. Step Four: Apply the Model
— Pilot – picked four subjects and ran a pilot program
doing intervention
— Showed that the model was helping us find students we need to
talk to
— Allowed focus on building methods for intervening
— Output of model fed into CRM from Semester 2 2013
— 2014 – Beginning to focus on risk for specific courses
such as Math intensive course, possible expansion to
include Physics
18. Recap
Step 2 –
Analyse for
Churn or
Risk Patterns
Step 3 – Build
a Repeatable
Model
Step 4 –
Apply the
Model
Step 1 –
Obtain Good
Customer
Data
SAS
Enterprise
Guide®
SAS Enterprise BI Server
SAS® Data Integration Studio
SAS®Web Report Studio
SAS Enterprise BI
Server
SAS Data Integration
Studio
SASWeb Report
Studio