Retaining Online Students


First:  A word about Power Pointless!  My colleagues and I at the University of Illinois Springfield Center for Online Learning, Research and Service present often at academic conferences.  We encourage attendees to freely share presentation materials online using web-native tools that encourage collaboration and updating. You may follow the session using your personal mobile media to dig deeper into the topics we discuss and share with others following the conference:



Retention 

Overall persistence of students in higher education is dropping - now at 68.8%, down 1.2% since 2009.  (persistence refers to students who remain enrolled at any college)

Overall retention remains at 58.2%. (retention refers to students who remain at the same college) 

Data from the National Student Clearinghouse Research Center

Clearinghouse Study




What is at stake?
  • Student hopes, dreams, plans, finances!
  • Careers left unfilled, families left without resources impacting generations
  • Countless reports of students paying loans for degrees never completed
  • In many cases, where we do not have classes filled to capacity, there is no substantial incremental cost per student retained - only marginal costs
  • 45 credit hours left in the degree @ $225/credit hour = $10,125
  • 45 credit hours left in the degree @ $300/credit hour = $13,500
  • 45 credit hours left in the degree @ $350/credit hour = $15,750
Lose 100 students = a million dollars or more in tuition revenue let alone human cost

Dedicating 10 staff members and associated software/hardware to save 100 students a year is cost effective.  


Retention 

Online retention rates at many colleges and universities hover in the 50-60% range

Lack of preparation, maturity, self-discipline are often blamed, but blame does not solve the problem....

How do we address the problem?  
  • Create a committee
  • Attend a workshop
  • "Let's try... " gut reaction
  • Focus group
There are cases where retention of online students approaches 90%

Successful programs vigilantly collect data and promptly act on findings
    • Examine student characteristics of those who do and do not persist - GPA, test scores, courses taken, transferring schools
    • Carefully examine admission requirements, online orientation and advising
    • Determine early interventions 
    • Design new approaches
    • Follow-up monitoring
    • Re-examine admission requirements, online orientation and advising
    • Design new approaches and interventions
    • The cycle continues
Examples: 

UIS one+two+one community college transfer program >80% persistence
Text messaging reminders prompt millennial student action





Descriptive and Diagnostic Analytics to Determine Blockages

This is the step too many colleges/universities skip - use your enterprise system (Banner, People Soft, etc.) and your learning management system to collect the data you need to make informed interventions
  • At what point are students dropping out / stopping out?
    • In the first two weeks of class
    • Before or after midterm
    • After their first class
    • After a particular class / instructor
  • Can we drill down to find clues in the performance of students in individual classes?
    • Does participation drop off?
    • Are grades of non-completers consistently lower in certain classes?
      • What strategies may be employed to improve performance there?
        • More low-stakes self quizzes
        • Peer mentoring
        • Pro-active tutoring
  • Are students changing majors?
  • Do students truly drop out - or do they transfer?
    • If they transfer elsewhere, how is that institution different?
MOOC dropouts - What we learn from those who leave


Why Students Drop / Stop

Plethora of reasons that students drop or stop out of online classes, here are some of those most often cited, how do we quantify/collect data on these:

  1. Student expectations that online classes are easier than on campus classes
  2. Students not adequately prepared for class (lacking pre-requisite knowledge or skills)
  3. Class availability issues - students cannot make steady, efficient progress toward degree/certificate completion
  4. Lack of discipline / self-motivation of students to attend to the class (association with maturity of students?)
  5. Poorly designed classes that fail to engage or adequately support students (questions of relevancy to interests/careers; unclear expectations-rubrics)
  6. Social environment in online class that tolerates bullying, prejudice or other non-supportive culture
  7. Faculty failure to engage students in meaningful/supportive dialog
  8. Faculty failure to identify and address problems as they occur (not using dashboards, formative assessments, or other tools to trigger interventions)
  9. Lack of student support and encouragement outside of the online classroom (advising, tutoring, peer-tutoring)
  10. Courses, programs lack clear career path linkage
  11. Financial issues - inability to pay tuition/fees
  12. Life intervenes (adult students are particularly vulnerable to such factors as child/parent health issues, job loss, other factors related to family/household) 




Monitoring Progress / Dashboards

Performance Dashboards enable better monitoring of student progress

Blackboard Performance Dashboard

Starfish dashboard and "kudos"


Purdue Course Signals program allows students to gauge performance against the rest of the class


Predictive Analytics

In higher education predictive analytics enables administrators to predict success or challenges based on the history of students (or applicants) with the same or similar characteristics such as:
  • prior education success
    • performance at prior institutions
    • performance in prior classes
  • age
  • zip code of residence/school/etc.
  • gender
  • ethnicity
  • test scores

Predictive Analytics Reporting Project


Prescriptive Analytics

Paired with predictive analytics are prescriptive interventions.  Given specific analytics for a student the previously-tested (and prescribed) interventions may include such activities as:
  • Intensive advising
  • Developmental module or class
  • Tutoring
  • Peer tutoring
  • Supplemental classes
  • Additional assignments
  • Course materials appealing to different learning style (visual, verbal, auditory, problem-based)
  • Adaptive learning delivery to assure successful scaffolding 

Implementing 12 Essentials of Prescriptive Analytics for Student Success


Putting Analytics to Work

Babson College research suggests that organizations go through stages in implementing data-driving decision making, these include
  • Analytically impaired (lack data, tools, expertise, commitment)
  • Localized analytics (some departments begin to apply analytics, while others lag)
  • Analytical aspirations (successes within or among competitors drive commitment)
  • Analytic implementation but still not founded in strategic focus and advantage
  • Enterprise-wide implementation




How can I keep up on the latest news and research?

Professional Continuing & Online Ed Blog


Ray Schroeder's Daily Blogs





Contact information:

Ray Schroeder 
@rayschroeder
Associate Vice Chancellor for Online Learning
University of Illinois Springfield
~
Director of the UPCEA Center for Online Leadership and Strategy
 ~
217-206-7531