LEARNING ANALYTICS IN HIGHER EDUCATIONPROMISING PRACTICES AND LESSONS LEARNED
Bodong Chen, University of Minnesota
October 27, 2016, Manila, Philippines
@bod0ng
BODONG CHENAssistant Professor in Learning Technologies
University of Minnesota-Twin Cities Research interests: online learning, learning analytics, CSCL,
knowledge building
... my main concern is the well being of the
plant materials ... And because of the diversity
of plants that we grow, we have to have a
wide range of niches to put those plants into.
Some need it to be a little cooler. Some want it
a little warmer. Some want to be drier. Some
want to be wetter. Our job here is to work with
Mother Nature and to try to provide the
conditions optimal for growth.
Source
PRECISION AGRICULTURE
Credit: Airborne
, U.S. Department of Homeland SecurityVisual Analytics Law Enforcement Toolkit (VALET)
LEARNING ANALYTICS IS“The measurement, collection, analysis, and
reporting of data about learners and their
contexts”
WHAT
Long, P., & Siemens, G. (2011). Penetrating the Fog: Analytics in Learning and Education. EducauseReview, 46(5), 30–32.
LEARNING ANALYTICS IS“The measurement, collection, analysis, and
reporting of data about learners and their
contexts
for understanding and optimising learning
and the environments in which learning
occurs”
WHY(Long & Siemens, 2011)
Buckingham Shum, S. (2012). . UNESCO Institute forInformation Technologies in Education.
UNESCO Policy Brief: Learning Analytics
PART 1: A SNAPSHOT OF AUSSIE UNIVERSITIES
Colvin, C., Rogers, T., Wade, A., Dawson, S., Gasevic, D., Buckingham Shum, S., … Fisher, J. (2015).
. Australian Department of Education.Student retention and learning analytics: A snapshot of Australian practices and a framework for
advancement
AN INTERVIEW STUDYRESEARCH QUESTION
How senior institutional leaders perceived learning analyticsincluding the drivers, affordances and constraints?
PARTICIPANTSSenior institutional leaders
(Deputy Vice Chancellors)
ANALYSESQualitative Coding + Cluster Analysis
CLUSTERS NOT DEFINED BY 'YEARS' ...Cluster 1 (7) Cluster 2 (26)
Purpose Measure Understand
Driver type Ef�ciency Learning andstudent success
Retention Independent Inter-dependent
Learningdimensionality
Unidimensional Multidimensional
Analytics Predictive Learning itself
... ... ...
(Colvin et al., 2015)
GETTING PEOPLE INVOLVED!InstructorsStudentsAdvisorsAdministratorsResearch Faculty
Credit: UMN LA Team
UMN LEARNING ANALYTICSDATA
Student Information SystemLearning Management SystemsStudent Advising Systems
A COMMON DATA LAYER
ANALYTICSDashboardsPredictive engines
BROADER DISCUSSIONSEthics: ethical use of dataStudents: interacting with studentsQuality: data qualityLeadership: the U's leadership structureResearch: esp. related to Unizin's deidenti�ed data
Credit: UMN LA Team
MY PEDAGOGICAL GOALSPromote forum participation from students?Help students become more aware and re�ective of their participation
1. LEARNING ANALYTICS NOT NEUTRALData are not neutralOur analytics are our pedagogy (Knight et al., 2014)
Interventionist by nature educational visions and values Replicate - Amplify - Transform (Hughes, Thomas, & Scharber, 2006)
2. CONVERSATIONSAmong datasetsAmong peopleBetween data and peopleAcross levels
Credits: , , 1 2 3
OPPORTUNITIES FOR OPEN UNIVERSITIESOpennessComprehensiveness of dataUnique local contextsCross-institution collaboration. . .
THANK YOU!
bodong.ch@bod0ng
ACKNOWLEDGEMENT