1 David Stevens, Manager of Web Services, Lehman College, City University of New York Joe Medved, Manager of Database & Applications, Lehman College, City University of New York Aarti Deshmukh, Senior Applications System Developer, Lehman College, City University of New York Beyond Recruitment & Retention: Success Via a Data-Centric Ecosystem
17
Embed
Institutional Success Via a Data-Centric Technology Ecosystem
The content and data landscape is vast and decentralized, and aggregating assets via a unified technology strategy poses significant challenges. This presentation describes how to design a federated technology ecosystem to inform recruitment, marketing, and outreach efforts, and increase student retention through the innovative use of data modeling and predictive analytics. Collecting data for actionable insight, leveraging CRM, web, and mobile channels, and tracking retention and graduation rates will be highlighted.
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
1
David Stevens, Manager of Web Services,
Lehman College, City University of New York
Joe Medved, Manager of Database &
Applications, Lehman College, City University
of New York
Aarti Deshmukh, Senior Applications System
Developer, Lehman College, City University of
New York
Beyond Recruitment & Retention:
Success Via a Data-Centric Ecosystem
2
What?: The suite of tools & applications that comprise
Lehman’s enterprise publishing & communication systems.
Why?:
To align the College’s messaging, brand identity, and delivery of
personalized content.
To expose critical information and create calls-to-action re
How?: By integrating silos and shadow systems, streamlining internal processes & enhancing user experience through a
federated strategic technology architecture.
Ecosystem Defined
3
Ecosystem At a Glance
4
Ecosystem in Practice
Event Management
Event category syndication maximizes exposure to key events contextually by publishing to multiple locations (e.g. college homepage, affiliate websites, & CUNY calendar).
Events may be promoted to Social Media channels & calls-to-action facilitate user engagement. Google analytics tracks traffic spikes and conversion points.
5
Ecosystem in Practice
Event Management
Event category syndication maximizes exposure to key events contextually by publishing to multiple locations (e.g. college homepage, affiliate websites, & CUNY calendar).
Events may be promoted to Social Media channels & calls-to-action facilitate user engagement. Google analytics tracks traffic spikes and conversion points.
6
Ecosystem in Practice
WordPress Newsletter:
Intuitive tagging automates the publishing of college news, blogs, and announcements to department and affiliate websites.
Calls-to-action facilitate
user engagement, and Google analytics track traffic spikes and conversion points.
7
Ecosystem in Practice
Digital Connect: Media Asset Repository
College videos may be published to YouTube, Vimeo, or iTunes U and published to Digital Connect, Lehman’s one-stop-shop for rich media content.
Through an intuitive categorization system, videos are published to multiple web properties from a single content source.
8
Ecosystem in Practice
Personalized Delivery of Content
Via Active Directory login, student course schedules, grades, alerts, and notifications can be delivered to our secure intranet site, Lehman Connect.
Personalized content will soon be sent to students via college’s mobile app. Critical alerts will appear as notifications.
From Data to Knowledge
Big Data: Data from traditional & digital sources for
discovery and analysis. Characterized by 3 V’s.
Business Intelligence: Tools and practices used to
analyze & optimize decisions and performance.
Analytics: Statistical discovery of meaningful
patterns for predictive scenarios
9
Different Questions & Tools
10
Reporting/BI > Analytics > Prescriptive
Davenport,
12/13 HBR
Lehman Examples
What is happening: LCD/BI reporting on
enrollment and retention
What is likely to happen: Rapid Insight
Analytics/predictive modeling
11
Lehman College: Predictive Analytics
Lehman is pursuing the power of regression and
predictive analytics.
Regression analysis: study of statistical relationships
among dependent and independent variables.
Based on the variables, we may be able to impact
student attrition, enrollment, graduation rates, etc.
End result of the analysis: a predictive model,
suggesting intervention strategies and possible
outcomes in future semesters.
12
Student Attrition Model
We studied attrition in a cohort of 454 FT/FT freshman students
that started in Fall 2011& followed their attrition rates through Spring 2014.
Relationships among attrition and 100+ parameters were
examined, including probation status, SAT scores, credits
attempted/earned, cumulative GPA, etc., for each semester.
The model showed a 26% attrition rate though Spring 2014:
336 students were retained and 118 attrited.
The model predicted that 109 students would attrit at the end of