380 likes | 559 Views
Learning Analytics. NEASC 126 th Annual Meeting and Conference December 2011. Malcolm Brown, EDUCAUSE Learning Initiative. Johann Ari Larusson Brandeis University. Dr. Becky Wai-Ling Packard Mt. Holyoke College. What’s ahead. Football What is Learning Analytics? Examples
E N D
Learning Analytics NEASC 126th Annual Meeting and Conference December 2011
Malcolm Brown, EDUCAUSE Learning Initiative Johann Ari Larusson Brandeis University Dr. Becky Wai-Ling Packard Mt. Holyoke College
What’s ahead • Football • What is Learning Analytics? • Examples • Possible benefits • Potential perils and pitfalls
Run a play Analyze results Run a play Analyze results Etc.
Run a play Analyze results Run a play Analyze results Etc.
Run a play Conduct a course Analyze results Analyze evaluations Run a play Conduct a course Analyze results Analyze evaluations Etc. Etc.
What is Learning Analytics? Compilation and analysis of student usage data… …to observe and understand learning behaviors… … to enable appropriate interventions.
SIS data Analysis Visualizations LMS data Reports Portfolio data Longitudinal data Alerts Comparative data
Audience Instructor facing Student facing
Before & after Before LA What happened?
Before & after Before LA After LA What is happening? What happened? What will happen?
Social Networks Adapting Pedagogical Practice (SNAPP) http://research.uow.edu.au/learningnetworks/
Your questions Is there a way to see when your students are “in” higher order learning modes? Is there a way to see what activities your students are typically doing when in higher order modes?
Learner dialog types • Disputational • Cumulative • Exploratory co-reasoning challenging evaluative Reasoned & equitable knowledge sharing
Learning analytics 1 Uses data (esp. learner-produced data) 2 Performs analysis of that data 3 Discovers information about the learners 4 Enables appropriate intervention
Possible benefits • Evidence-based decisions • Measures what students actually do • Identify who needs help • Identify course-wide patterns
Potential pitfalls • Privacy • Profiling • Information sharing • Data stewardship • Impoverished model of learning
Johann Ari Larusson Brandeis University
Point of Originality Example of an R&D Learning Analytics Project
The trend/problem • Migration towards larger “gateway” courses • Negative impact on the student’s learning process • Less useful for fostering higher order thinking • Evaluating/monitoring the quality of students’ work • In terms of the depth of the students’ learning • Extremely time consuming, even in smaller classes • Also, the instructor’s work is self-reflective • However • Technology, like blogs, extend the physical boundaries of the classroom, introduce and foster learning communities, even in larger classes • Automatically produce data (electronic form) that can be analyzed • Is in itself a black box but enables us to peek inside the black box
Point of Originality • Automated analysis tool: • Via lexical analysis, track students’ language migration from mere paraphrase to mastery • Isolating the moment in time when students demonstrate the ability to explain concepts in their own words, their “point of originality” in time • Recreates the same cognitive activity that educators might ordinarily undergo • Not an automated grading tool • Core components: • WordNet: arranges words by their conceptual-semantic and lexical relationships, notes similarity between two words that don’t have literally identical meanings. • Using input query terms (that relate to key course topics), algorithm calculates how far a student’s language has evolved to explain the course content.
Testing the hypothesis • Data: • Co-blogging in 25+ interdisciplinary course (heavy reading list) • Correlating originality scores on blogs and papers to blogging activity • Grades assigned two years earlier by person not involved in the project. • Some results: • More original co-blogging work leads to higher paper grades • Originality variance: • Students with papers above avg. grade have a negative variance. They were at the height of understanding during blogging. • Higher exposure • (to other students’ writing/dialogs) in the blogosphere leads to more original papers • Active participation: • Making more contributions to the blogosphere yields higher paper originality scores
Dr. Becky Wai-Ling Packard Mt. Holyoke College
Student Retention- Our Coordination Process • Identifying the Key People • Selecting Appropriate Tools • Access • Data Sources
Data Sources and Access • Grades and Course Schedule: Advisor • Learning Disability and Mental Health: Dean • Finaid; Transfer Status: Admission • Disciplinary; Dorm Change Requests: Res Life • Current Activities including Sports: Coach • Midsemester Report: Professor, Advisor, Dean • Goals, Accomplishments: Career Center
Midsemester Report • Is students’ attendance satisfactory? No • Written work required, turned in, quality acceptable? No • Quality of the student’s work: BORDERLINE Instructor Comments: Jenny, we talked about your absences and missed assignments. As this is a fast-paced course, your absences and missed assignments have added up quickly and have had a negative influence on your overall grade. Please come and see me so we can figure out what is still possible for you.I’d really like to see you succeed.
Visual: Key People Initial Faculty Advisor Major Advisor Class Dean Student Profs Career Center Student Advisor (dorm) Pre Prof’l Advisor Coach Student Dept Liaison Coach