Introduction One of the many definitions of Predictive Analytics says that it “describes a range of analytical and statistical techniques used for developing models that may be used to predict future events or behaviors 1 ” For some time now, retailers like Amazon and Tesco have leveraged predictive analytics to acquire an amazing depth of customer understanding and deliver personalized propositions. Automotive giant BMW is using it to identify and fix vulnerabilities before putting new models into production. Financial institutions are leveraging this technology for purposes as diverse as improving cross-selling effectiveness and mitigating risk. For the education industry, a comparative latecomer to this field, it is in student retention that predictive analytics will find its biggest role. The challenge of churn Minimizing dropout ranks among the top priorities of education providers today. Globally, the problem of youth unemployment is being exacerbated by the absence of critical job skills. The only way to address this talent gap is by improving the delivery of higher and tertiary education. However, to achieve great learning outcomes, educational institutions must not only provide course content and teaching of high quality, but also ensure that every student completes the program. When students drop out of their courses, it not only impacts the employable talent pool but also wastes the considerable resources that were spent to enroll them. Just to see things in perspective, consider that in 2012-13, private institutions in the United States offering four-year programs reported a median spending of US$ 2,433 per student recruitment 2 ; several for-profit universities ended up spending more than twice that amount 3 . A post graduate from IIM, Lucknow with about 20 years’ experience Arvind, has spent the last 10 years in leadership roles advocating technology in Learning, Training and Assessment with customers across the globe. About the authors Predictive Analytics for Student Retention Arvind Thothadri Vice President – New Initiatives Co-founder of MeritTrac Services, India’s largest assessment company that develops and delivers over 2 million online and paper assessments each year, Mohan drives development and innovation on EduNxt Learning Ecosystem. Mohan Kannegal Head - Learning Solutions Group
Our thoughts on how a comprehensive Learning ecosystem can help in predicting outputs. Analytics is good but if it is possible to predict the likely dropouts in a college, that would be great. That's what we have attempted
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Introduction One of the many definitions of Predictive Analytics says that it “describes a
range of analytical and statistical techniques used for developing models
that may be used to predict future events or behaviors1” For some time
now, retailers like Amazon and Tesco have leveraged predictive analytics to
acquire an amazing depth of customer understanding and deliver
personalized propositions. Automotive giant BMW is using it to identify and
fix vulnerabilities before putting new models into production. Financial
institutions are leveraging this technology for purposes as diverse as
improving cross-selling effectiveness and mitigating risk. For the education
industry, a comparative latecomer to this field, it is in student retention that
predictive analytics will find its biggest role.
The challenge of churnMinimizing dropout ranks among the top priorities of education providers today. Globally, the problem of youth unemployment is being exacerbated by the
absence of critical job skills. The only way to address this talent gap is by
improving the delivery of higher and tertiary education. However, to achieve
great learning outcomes, educational institutions must not only provide
course content and teaching of high quality, but also ensure that every
student completes the program. When students drop out of their courses, it
not only impacts the employable talent pool but also wastes the
considerable resources that were spent to enroll them. Just to see things in
perspective, consider that in 2012-13, private institutions in the United
States offering four-year programs reported a median spending of US$
2,433 per student recruitment2; several for-profit universities ended up
spending more than twice that amount3.
A post graduate from IIM, Lucknow
with about 20 years’ experience
Arvind, has spent the last 10 years
in leadership roles advocating
technology in Learning, Training
and Assessment with customers
across the globe.
About the authors
Predictive Analytics for Student Retention
Arvind ThothadriVice President – New Initiatives
Co-founder of MeritTrac Services,
India’s largest assessment
company that develops and
delivers over 2 million online and
paper assessments each year,
Mohan drives development and
innovation on EduNxt Learning
Ecosystem.
Mohan Kannegal Head - Learning Solutions Group
Small wonder then that minimizing dropout ranks among the top priorities of
education providers today. As schools, universities, private educational
organizations and corporate entities take a number of steps – ranging from
a more selective enrolment process to student counseling – to improve
retention, they are finding a valuable source of support in predictive
analytics.
Institutions can employ analytics at various stages of the student lifecycle to
manage dropout. Information that is collected at the admissions stage, to
determine student eligibility such as financial background and academic
proficiency, can be later revisited to identify students that need special
attention and support. However, education providers will maximize
outcomes only if they deploy predictive analytics throughout the course of
engagement.
The role of learning management platforms The choice of learning management platform is critical because of its role in
collecting student data & predicting graduation rates. While most platforms
manage the student lifecycle from end-to-end (enrolment through
certification) quite efficiently, they don’t necessarily have the same analytical
prowess. EduNxt is a platform with comprehensive, proven analytics
capability. EduNxt Analytics evaluates and monitors, Student Performance,
Content Quality, Course Health and Student Engagement to provide user
institutions and enterprises with key insights, which are extremely relevant
to student retention. A student’s performance in terms of the number of
hours spent on a course versus the class average and minimum required to
pass, is one of the factors that determines test scores and likelihood of
course completion. EduNxt Analytics draws attention to other such red
flags, like a higher than average bounce rate for a particular topic, poor
participation or performance in assessments and interaction sessions, and
most telling of all, flagging engagement. Some of the factors which seem to
greatly influence a student’s decision to continue with the program include
the time spent on the platform and the “Recency Effect” produced by the
results of the last semester exam.
Choosing a platform with
comprehensive, proven
analytics capability.
The Right Analytical ModelEduNxt Analytics is driven by a robust modeling methodology and engine,
which studies data gathered from various systems in the ecosystem to
identify influential factors in student retention, such as course delivery,
content, student engagement and student performance. This resolves a
major challenge faced by institutions, whose data usually rests in multiple
departmental and transactional silos – Student Information, Learning
Management, Examination, Results Consolidation, and Helpdesk systems
etc. – making it difficult to see a holistic view of the problem. EduNxt
Analytics overcomes this by providing a single view of all student information
along with access to historical data.
Further, the platform provides clear insights and actionable points to
empower educational institutions and their faculty to make timely
interventions.
ConclusionOne of the biggest goals of education providers is to maximize learning
outcomes for their students. To achieve this, institutions must ensure that
students go the distance and complete the program. Program dropout not
only diminishes
learning outcomes but also wastes the resources invested in student
enrolment. Predictive analytics plays a crucial role here by identifying
students at risk early and suggesting remedial courses of action at every
stage in a program.
The EduNxt Learning Ecosystem has worked with educational institutions around the world to transform the way they manage student retention and deliver value to each student.
EduNxt is the Unified Learning Solution from Manipal Global Education
Service Pvt Ltd. EduNxt is the backbone through which online learning is
driven across the various Educational institutes in the Manipal ecosystem.
EduNxt plays a key role in learning and assessments during the entire life
cycle of a student from enrolment to graduation. A very robust ( over
250,000 users ) and easy-to-use platform that is mobile enabled, it comes
With the help of EduNxt
Analytics, Sikkim Manipal
University, the largest
privately owned distance
education provider in India
was able to identify
students at maximum risk
of dropping out with 96%
accuracy.
with top end analytics that facilitate key stake holders from students,
teachers and administrators to leverage and take pro-active remedial