The Moneyball Effect: How Data Will Transform Student Success in 10 Years

Released: 01/01/1970

Guest Post on The Evolllution

By 

In his 2003 book “Moneyball,” bestselling author Michael Lewis chronicled the 2002 Oakland Athletics baseball team’s unprecedented run to win their division championship through a specialized analysis of baseball data called “Sabermetrics” (also referred to as “Moneyball”).

By analyzing objective, evidence-based data on historical player performance, the cash-strapped Oakland team built a repeatable, winning strategy that challenged conventional baseball wisdom. More than a decade later, the vast majority of professional sports teams — including the Dallas Mavericks, Los Angeles Clippers and Tampa Bay Rays — now employ statisticians and data analysts to turn player data into actionable coaching insights.

Just as player analytics have transformed decision making in professional sports teams, higher education and learning organizations will use student analytics to transform teaching models, better meet students’ needs and improve learning outcomes over the next decade.

Let’s explore how three types of learning analysis — predictive, adaptive and personalized — will harness the power of student metrics to impact performance.

1. Predictive learning analysis

Whether online or in the classroom, students today interact with various data systems. Learning management and student information systems are chronicles of students’ past and current learning activities. By analyzing this data using evidence-based research, organizations can identify trends, behaviors and patterns to better predict student outcomes at the individual and group level.

For example, learning organizations will analyze:

  • Time spent in class and online, relative to course completion rates;
  • Engagement with faculty, relative to outcomes;
  • Interim assessment scores, relative to final grades; and
  • Prerequisite knowledge and coursework, relative to success rates.

The most powerful predictive data is historical data, so it will be the accumulation of data longitudinally that will become particularly predictive and interesting. With predictive analysis, organizations will gain objective insights into student behaviors and the ability to adapt and personalize future learning to improve student retention and success.

2. Adaptive learning analysis

Combined with predictive analytics, organizations will use adaptive learning analysis to identify individual student needs and quickly intervene to improve the odds of success. Student metrics — such as learning time, response latency, engagement levels and assessment results — form the basis for these analyses.

Armed with these insights, instructors will:

  • Redefine student inputs based on successful outcomes;
  • Adapt assessments, relative to past performance;
  • Adjust student interaction dynamically;
  • Provide on-demand tutoring based on responses; and
  • Offer real-time prompts, clues and grading.

Beyond simply making adjustments to the course materials and delivery pace to improve student retention and outcomes is the ability to constantly and iteratively test those changes. As suggested in a recent article by Chris Proulx, “Three Archetypes of the Future Post-Secondary Instructor,” he suggests the emergence of the “Course Hacker” role in higher education, where:

“…the Course Hacker would be a faculty member with strong technical and statistical skills who would study data about which course assets were being used and by whom, which students worked more quickly or slowly, which questions caused the most problems on a quiz, who were the most socially active students in the course, who were the lurkers but getting high marks, etc.  Armed with those deep insights, they would be continually adapting course content, providing support and remedial help to targeted students, creating incentives to motivate people past critical blocks in the course, etc.”

For example:

“The data tells us that this student is having a hard time getting through the materials in the allotted time. Let’s make Tweak A and see if that improves course completion. No? Let’s try Tweak B and compare the data.”

And as large-scale participation courses, such as MOOCs (massive open online courses), become more prevalent over the next decade, adaptive learning analysis will allow learning organizations to provide scalable, agile interaction to meet the unique needs of individual students.

3. Personalized learning

As organizations use analytics to better understand students’ distinct learning behavior profiles, it will open the door to personalized learning. Where adaptive learning is used to quickly intervene when students are struggling, personalized learning focuses on providing the student with choices to determine when, what and how they learn.

For example, organizations will personalize learning by:

  • Adjusting the pace of learning;
  • Creating personalized learning paths based on interest and existing knowledge;
  • Localizing the curriculum to regional or cultural needs;
  • Developing communities of similar students;
  • Offering alternate learning times to accommodate personal schedules; and
  • Using algorithms to dynamically create peer-to-peer relationships.

Winning: Data analytics to drive student performance

Already, new technology tools are making it easier for learning organizations to access and analyze student metrics. But data alone is not actionable information. Instead, learning organizations need objective analytics that help them predict, adapt and personalize their pedagogy to maximize learning outcomes online or in the classroom. Over the next decade, organizations that can do this, and do it well, will prove to be the winners.