What data from 3
million learners can
tell us about effective
course design
John Whitmer, Ed.D.
Director, Analytics & Research
john.whitmer@blackboard.com | @johncwhitmer
22
Meta- questions driving our Learning Analytics research @ Blackboard
1. How do students & teachers use our platforms? How is this use
related to student achievement? [or satisfaction, or risk, or …]
3. How can we integrate these findings into features/functionality
that apply to the broad spectrum of ways people use our
platforms?
2. Do these findings apply equally to students ‘at promise’ due to
their academic achievement or background characteristics? (e.g.
race, class, family education, geography)
33
Outline
1. Positioning Learning Analytics
2. Research Findings
3. Q & A
4
1. Positioning Learning Analytics
55
66
Economist. (2010, 11/4/2010). Augmented business: Smart systems will disrupt lots of industries, and perhaps the entire economy. The Economist.
200MBof data emissions annually
77
Logged into course within 24
hours
Failed first exam
Interacts frequently in
discussion boards
No declared majorHasn’t taken college-level math
8
“ ...measurement, collection, analysis and reporting of data about learners and
their contexts, for purposes of understanding and optimizing learning
and the environments in which it occurs.”
Learning and Knowledge Analytics Conference, 2011
What is learning analytics
9
Strong interest by faculty and students
1010
Blackboard’s Learning Data Footprint (2015 #’s)
1.6M Unique Courses
40M Course Content Items
4M Unique Students
Blackboard Learn =
¼ total data
775M LMS Sessions
1111
Commitment to Privacy & Openness
• Analyze data records
that are not only
removed of PII, but
de-personalized
(individual &
institutional levels)
• Share results and open
discussion procedures
for analysis to inform
broader educational
community
• Respect territorial
jurisdictions and safe
harbor provisions
12
2. Research Findings
13
Why this is hard?
• LMS data is messy
• Adoption is hugely varied
14
Same model doesn't work for all courses
• 1.2M students
• 34,519 courses
• 788 institutions
• Overall effect size < 1%
15
But strong effect in some courses
(n=7,648, 22%)
16
The Effectiveness of Blackboard Learn Tool
Usage
17
Purpose of Investigation
To determine the most important tools in Bb Learn, by observing:
– Tools that are used the most (in minutes, for instance)
– Tools that have strongest relationship with final grade
– Tools that are ‘underused’ the most (by learners & instructors); tools that have the greatest
potential to improve learning outcomes
Allows us to see which tools educate students, and are therefore useful
Reinforce the educational impact of the Blackboard Learn platform
18
Data Filtering
Filters decreased the number of students analyzed from 3.37 million users in 70,000 courses from 927 institutions
to 601,544 users (17% of total) in 18,810 courses (26.8% of total) from 663 institutions (71.5% of total)
Class Size
between 10 and 500 students
Activity Rates
over 1 hour online as a course average
Grade Distribution
average grade between 40% & 95%
19
Finding: Tool Use & Grade
Tool use and Final Grade do not have a linear relationship;
there is a diminishing marginal effect of tool use on Final Grade
Interpretations
• Students absent from course activity are at
greatest risk of low achievement.
• The first time you read/see a PowerPoint
presentation, you learn a lot, but the
second time you read/see it, you learn
less.
• Getting from a 90% to a 95% requires
more effort than getting from a 60% to a
65%.
20
Finding: Tool Use & Grade
Tool use and Final Grade do not have a linear relationship;
there is a diminishing marginal effect of tool use on Final Grade
Interpretations
• Students absent from course activity are at
greatest risk of low achievement.
• The first time you read/see a PowerPoint
presentation, you learn a lot, but the
second time you read/see it, you learn
less.
• Getting from a 90% to a 95% requires
more effort than getting from a 60% to a
65%.
Log transformation shows
stronger trend
21
Investigation Achievement by Specific Tools Used
Analysis Steps
• Identify most frequently used tools
• Separate tool use into no use + quartiles
• Divide students into 3 groups by course grade
• High (80+)
• Passing (60-79)
• Low/Failing (0-59)
22
Finding: MyGrades
At every level, probability of higher grade increases with increased use.
Causal? Probably not. Good indicator? Absolutely.
23
Finding: Course contents
More is not always better. Large jump none to some; then no relationship
24
Finding: Assessments/Assignments
Students above mean have lower likelihood of achieving a high grade than students below the mean
25
Next Steps & Discussion
26
Next Steps in Product & Research
• Refine Ultra “Learning Analytics” triggers for low/high achievement; focus on
getting started, not achieving “top of class” in activity.
• Explore data points beyond time on task; semantic analysis, writing level analysis,
other more rich data points
• Investigate course design structures and patterns in how teachers create course
experiences using Learn
• Collaborate with institutions on research to consider alternative measures of
success besides course final grade (course evaluations, grades in subsequent
courses)
Questions?
John Whitmer, Ed.D.
john.whitmer@blackboard.com
@johncwhitmer

What data from 3 million learners can tell us about effective course design

  • 1.
    What data from3 million learners can tell us about effective course design John Whitmer, Ed.D. Director, Analytics & Research john.whitmer@blackboard.com | @johncwhitmer
  • 2.
    22 Meta- questions drivingour Learning Analytics research @ Blackboard 1. How do students & teachers use our platforms? How is this use related to student achievement? [or satisfaction, or risk, or …] 3. How can we integrate these findings into features/functionality that apply to the broad spectrum of ways people use our platforms? 2. Do these findings apply equally to students ‘at promise’ due to their academic achievement or background characteristics? (e.g. race, class, family education, geography)
  • 3.
    33 Outline 1. Positioning LearningAnalytics 2. Research Findings 3. Q & A
  • 4.
  • 5.
  • 6.
    66 Economist. (2010, 11/4/2010).Augmented business: Smart systems will disrupt lots of industries, and perhaps the entire economy. The Economist. 200MBof data emissions annually
  • 7.
    77 Logged into coursewithin 24 hours Failed first exam Interacts frequently in discussion boards No declared majorHasn’t taken college-level math
  • 8.
    8 “ ...measurement, collection,analysis and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs.” Learning and Knowledge Analytics Conference, 2011 What is learning analytics
  • 9.
    9 Strong interest byfaculty and students
  • 10.
    1010 Blackboard’s Learning DataFootprint (2015 #’s) 1.6M Unique Courses 40M Course Content Items 4M Unique Students Blackboard Learn = ¼ total data 775M LMS Sessions
  • 11.
    1111 Commitment to Privacy& Openness • Analyze data records that are not only removed of PII, but de-personalized (individual & institutional levels) • Share results and open discussion procedures for analysis to inform broader educational community • Respect territorial jurisdictions and safe harbor provisions
  • 12.
  • 13.
    13 Why this ishard? • LMS data is messy • Adoption is hugely varied
  • 14.
    14 Same model doesn'twork for all courses • 1.2M students • 34,519 courses • 788 institutions • Overall effect size < 1%
  • 15.
    15 But strong effectin some courses (n=7,648, 22%)
  • 16.
    16 The Effectiveness ofBlackboard Learn Tool Usage
  • 17.
    17 Purpose of Investigation Todetermine the most important tools in Bb Learn, by observing: – Tools that are used the most (in minutes, for instance) – Tools that have strongest relationship with final grade – Tools that are ‘underused’ the most (by learners & instructors); tools that have the greatest potential to improve learning outcomes Allows us to see which tools educate students, and are therefore useful Reinforce the educational impact of the Blackboard Learn platform
  • 18.
    18 Data Filtering Filters decreasedthe number of students analyzed from 3.37 million users in 70,000 courses from 927 institutions to 601,544 users (17% of total) in 18,810 courses (26.8% of total) from 663 institutions (71.5% of total) Class Size between 10 and 500 students Activity Rates over 1 hour online as a course average Grade Distribution average grade between 40% & 95%
  • 19.
    19 Finding: Tool Use& Grade Tool use and Final Grade do not have a linear relationship; there is a diminishing marginal effect of tool use on Final Grade Interpretations • Students absent from course activity are at greatest risk of low achievement. • The first time you read/see a PowerPoint presentation, you learn a lot, but the second time you read/see it, you learn less. • Getting from a 90% to a 95% requires more effort than getting from a 60% to a 65%.
  • 20.
    20 Finding: Tool Use& Grade Tool use and Final Grade do not have a linear relationship; there is a diminishing marginal effect of tool use on Final Grade Interpretations • Students absent from course activity are at greatest risk of low achievement. • The first time you read/see a PowerPoint presentation, you learn a lot, but the second time you read/see it, you learn less. • Getting from a 90% to a 95% requires more effort than getting from a 60% to a 65%. Log transformation shows stronger trend
  • 21.
    21 Investigation Achievement bySpecific Tools Used Analysis Steps • Identify most frequently used tools • Separate tool use into no use + quartiles • Divide students into 3 groups by course grade • High (80+) • Passing (60-79) • Low/Failing (0-59)
  • 22.
    22 Finding: MyGrades At everylevel, probability of higher grade increases with increased use. Causal? Probably not. Good indicator? Absolutely.
  • 23.
    23 Finding: Course contents Moreis not always better. Large jump none to some; then no relationship
  • 24.
    24 Finding: Assessments/Assignments Students abovemean have lower likelihood of achieving a high grade than students below the mean
  • 25.
    25 Next Steps &Discussion
  • 26.
    26 Next Steps inProduct & Research • Refine Ultra “Learning Analytics” triggers for low/high achievement; focus on getting started, not achieving “top of class” in activity. • Explore data points beyond time on task; semantic analysis, writing level analysis, other more rich data points • Investigate course design structures and patterns in how teachers create course experiences using Learn • Collaborate with institutions on research to consider alternative measures of success besides course final grade (course evaluations, grades in subsequent courses)
  • 27.