Online classrooms are de facto rich data gathering platforms. Educators can collect this data and use it to improve student outcomes through predictive analytics.
1. BIG DATA IN THE
ONLINE CLASSROOM
JAMES P. HOWARD, II
FEBRUARY 13, 2016
2. OUTLINE
• Big data has changed and continues to change almost every business
• It has the potential to radically transform education and its processes to benefit students
• In this presentation, we examine four questions at the heart of big data and online education:
1. What are the data that you could or should collect when you are teaching, either in the learning management
system or performance measures?
2. How would you collect, manage, and utilize the data?
3. Are there any concerns or issues of these learning analytics?
4. Any other tips/recommendations when using the analytics?
3. PERFORMANCE MEASURES
WHAT ARE THE DATA THAT YOU COULD OR SHOULD COLLECT WHEN YOU ARE TEACHING,
EITHER IN THE LEARNING MANAGEMENT SYSTEM OR PERFORMANCE MEASURES?
4. STUDENT INTERACTION DATA
• How often the student logs in, even for passive
reading and review
• How often the student interacts with the course
materials
• What types of interactions the student has
• When the student interacts (day, time)
CollegeDegrees360 / Flickr / CC-BY Baepler & Murdoch (2010)
5. GRADE DATA
• Student grades by assignment delivery
mechanism
• Homework, timed quizzes, proctored exams
• Discussion and interactive
• Student grade trends, going up or down
• Student grades in other program courses
• Especially prerequisites
Mathieu Plourde / Flickr / CC-BY Arnold (2010)
6. TEXT ANALYTICS
• Text analysis can be used to study a student
writing
• Used in cheating detection
• Used to automatically grade the SAT’s essay
• Text analysis looks at the words used and
sentence structure to
• Classify text by topic area
• Identify the sentiment or “feeling” of the text
Nick Ares / Flickr / CC-BY Hara, Bonk, & Angeli (2000)
7. SOCIAL NETWORK ANALYSIS
• Social network analysis studies the interactions
among individuals in a group
• Can be applied to student interactions
• Can show how information flows around the
classroom
Christiane Birr / Flickr / CC-BY D’Andrea, Ferri, & Grifoni (2010)
9. BETTER GUIDANCE AND COUNSELING
• Student data from prior students and courses
can help with course selection
• Students can be placed into the right level of
English or mathematics depending on
proficiency
• Students will achieve more based on correct
placements
Lenarc / Wikimedia Commons / CC-BY Amey & Long 1998
10. EARLY INTERVENTION
• Early intervention can come from grade data
• Students who are underperforming early can be
treated quickly to improve retention and success
• Early treatment can be targeted what students
need most
Eastern Mennonite University / Flickr / CC-BY Amey & Long (1998)
11. AUTOMATED GRADING
• Advanced text analysis can be used for
automated grading
• Hand-scoring can be needed frequently
• Automated grading can increase the
student/faculty ratio
• Cuts costs to the institution
Mixabest / Wikimedia Commons / BY-CC Hara, Bonk, & Angeli (2000)
13. EQUITY
Michael Coghlan / Flickr / CC-BY
• Some students may receive intervention while
others do not
• Students who need intervention may not be
successfully identified
• Students who do not qualify for intervention
may receive insufficient support if instructors
are spending too much time with those who do
14. DEPERSONALIZATION
Keir Mucklestone-Barnett / Flickr / CC-BY
• Instruct runs risk of losing connection to
individual students
• Scorecard-approach reduces student learning to
box-checking
• Automated grading increases this disconnect
• Students run risk of losing connection to
instructor
15. PRIVACY
Mike Mozart / Flickr / CC-BY
• Student data, once collected, may be shared
internally
• Student data may be leaked or stolen
• Data security cannot be guaranteed
• Students may be put off by intervention
attempts
17. ADDITIONAL THOUGHTS AND SUMMARY
• Big data has a place in online education
• The place is still poorly defined by lack of access, training, and tools
• Better integration will lead the way
• There are risks from adopting too much data reliance for both students and institutions
18. REFERENCES
Amey, M. J., & Long, P. N. (1998).
Developmental course work and early
placement: Success strategies for
underprepared community college students.
Community College Journal of Research and
Practice, 22(1), 3-10.
Arnold, K. E. (2010). Signals: Applying Academic
Analytics. Educause Quarterly, 33(1), n1.
Arnold, K. E., & Pistilli, M. D. (2012, April).
Course signals at Purdue: using learning
analytics to increase student success. In
Proceedings of the 2nd international
conference on learning analytics and
knowledge (pp. 267-270). ACM.
Baepler, P., & Murdoch, C. J. (2010). Academic
analytics and data mining in higher education.
International Journal for the Scholarship of
Teaching and Learning, 4(2), 17.
D’Andrea, A., Ferri, F., & Grifoni, P. (2010). An
overview of methods for virtual social
networks analysis In Abraham, Ajith et al.
Computational Social Network Analysis: Trends,
Tools and Research Advances, (pp. 3-25).
Springer London.
Hara, N., Bonk, C. J., & Angeli, C. (2000).
Content analysis of online discussion in an
applied educational psychology course.
Instructional Science, 28(2), 115-152.
Timetrax23 / Flickr / BY-CC