learning analytics
DESCRIPTION
An overview of Learning Analytics in Higher EducationTRANSCRIPT
Learning Analytics: Student Learning and
RetentionDr Barbara Newland
Principal Lecturer Learning and Teaching (e-Learning)Centre for Learning and Teaching
What are Learning Analytics?
Current local and global developments
Improving retention
Enhancing student learning
Discussion
Summary
Overview
“refers to the interpretation of a wide range of data produced by and gathered on behalf of students in order to assess academic progress, predict future performance, and spot potential issues.” (Horizon, 2012)
“applies the model of analytics to the specific goal of improving learning outcomes.” (ELI, 2011)
Learning Analytics
Increasing global interest and the 2012 Horizon report predicts that there will be widespread adoption of learning analytics within 2 to 3 years
The output from learning analytics can be tailored for students, academics and institutions
Often this representation is visual ie graphs, diagrams etc.
Global developments
Example - bipartite sociogram
Adam Cooper’s work blogblogs.cetis.ac.uk
Society for Learning Analytics Research (SOLAR)
Reasons for learning analytics
Institutions can look for patterns across the institution and within Schools or degree programmes
Academics can look at the data to decide when to intervene to enable better outcomes both for retention and achievement.
Student use of learning analytics tools can enable them to view their levels of activity, attendance, progress and grades in comparison with other students
Studentcentral ◦ Performance dashboard so academics can see who has been
accessing the module and centrally to see if academics have been using their own areas
BSMS503 – run stats twice per year since 2007 (Tim Vincent) ◦ Graphs are typically the most helpful ◦ ‘Hits per day' graph shows clearly relative activity on the module:◦ Huge spikes of use prior to finals exams indicating that students
use the resource as a revision tool and repeatedly too.◦ Mostly in the evenings and how many students are working in the
early hours of the morning!◦ Saturdays were the most popular for using the quizzes
Current situation at Brighton
Methodology ◦ individual institutions and multi-institutional projects such as the
Predictive Analytics Framework (PAR)◦ analysis of big data sets of digital breadcrumbs looking for
patterns◦ “in general it includes information about the frequency with
which students access online materials or the results of assessments from student exercises and activities conducted online” (ELI, 2012)
Findings◦ it is easier to collect the data than to know how to use it to help
students◦ PAR project has found that similar models of learning analytics
can be used in different institutions.
Current research
In some universities it is being used to help to retain students through predicting those at risk of leaving
Retention
Jenzabar http://www.jenzabar.net/higher-ed-solutions/retention
Forsythe, R et al
Multi-institutional project - 16 institutions, over 1,000,000 student and 6,000,000 course level records
Similar models were used in each institution
System is predictive ie it sends an alert to an academic counselor that a student might not attend the following week so the counselor can contact the student.
Predictive Analytics Framework (PAR)
Unsurprisingly, students who do not engage achieve lower grades
5 years of data clearly correlating student VLE activity and grades found: ◦ Students earning a D or F at UMBC tend to use
Blackboard on average 39 percent less than students earning higher grades.
◦ http://www.umbc.edu/blogs/oit-news/reports/
Student learning
Grand Canyon University VLE is designed to enable data capture on student interaction with the system
Students, academics and administrators all receive different
perspectives on the data according to their needs
Students can see their performance relative to other students and compare their time on different learning activities (Kutty and Mueller, 2012)
At Purdue and Rio Salado College – LA used to make predictions and anticipate problems
Based on personalized data and predictive algorithms, system alerts trigger individualized interventions that can help students, advisors, and/or faculty tap resources to avert failure (Oblinger, 2013)
Improving Student Outcomes using Predictive Analytics
Purdue University – course signals
http://www.itap.purdue.edu/learning/tools/signals/
Accesses by Grade (SP2012)
Blackboard Analytics for Learn: Student View
McGraw-Hill LearnSmart
http://learnsmart.prod.customer.mcgraw-hill.com/about/take-a-tour/
LearnSmart video
“data can point learners to personalized learning pathways tailored to their needs, aspirations, abilities, and timelines.”
“data is actually most useful to inform thinking, questioning, planning, and next steps.”
(Oblinger, D. 2013)
Personalised learning pathways
How can analytics be used to identify and promote effective learning behaviors?
What types of alerts and dashboards for insights into analytics data are most useful, and who should be using them?
What are the issues?
Discussion
“Analytics requires a culture of inquiry, and inquiry creates an analytics culture.”
“Ask good questions; use good data.“
“Analytics is an investment”
“Technology makes education more personal, not less. Systems don't replace people; they empower people—both advisors and students—to make better decisions.”
(Oblinger, D, 2012)
Summary
“Data, by itself, does not improve student success. Although learning analytics offer great promise for transforming the accountability, personalization, and relevance that promise will not be fully realized until we put the power of better-informed decision making into the hands of front-line educators.”
(Wagner and Rice, 2012)
Summary
“analytics should be a torch and not a hammer“
Clay Shirky
ELI, 2011, 7 Things You Should Know about First Generation Learning Analytics, 2011, Educause http://www.educause.edu/library/resources/7-things-you-should-know-about-first-generation-learning-analytics
Forsythe, R., Chacon, F. J., Spicer, D. Z., Valbuena, A, 2012, Two Case Studies of Learner Analytics in the University System of http://www.educause.edu/ero/article/two-case-studies-learner-analytics-university-system-maryland
Horizon Report, 2012, Educause Kutty, M and Mueller, B, 2012, “Grand Canyon University: How We Are Improving Student
Outcomes using Predictive Analytics.”, Educause conference MacNeill, S. Analytics; What is Changing and Why Does it Matter? A Briefing Paper CETIS
Analytics Series Vol.1, No.1 Oblinger, D, 2012, Analytics: What We're Hearing http://
www.educause.edu/ero/article/analytics-what-were-hearing Oblinger, D. (2013)Analytics: Changing the Conversation, EDUCAUSE Review, vol. 48, no. 1
(January/February 2013 )Jan 28, 13 http://www.educause.edu/ero/article/analytics-changing-conversation Predictive Analytics Framework (PAR) http://wcet.wiche.edu/advance/par-framework SOLAR – Society for Learning Analytics Research http://www.solaresearch.org/ Wagner, E and Ice, P, 2012, Data Changes Everything: Delivering on the Promise of Learning
Analytics in Higher Educationhttp://www.educause.edu/ero/article/data-changes-everything-delivering-promise-learning-analytics-higher-education
References