As Associate Director of Skylight, The Science Centre for Learning and Teaching , Warren Code, has has been involved with a range of teaching and learning projects across the Faculty of Science, with a particular focus on the Carl Wieman Science Education Initiative and other faculty/campus-wide initiatives. A substantial part of his work is advising and professional development for their department-based Science Education Specialists, evaluating impact on students and faculty, and connecting with people from UBC and other institutions interested in the accomplishments in teaching and learning within the Faculty of Science. He has also been connected to some analytics projects, including analysis of student response/log data in WeBWorK. In this session, Warren will give an update on the current state of Learning Analytics in the Faculty of Science:
“I’ll be presenting an update on learning analytics in the Faculty of Science: some notable recent projects, and a variety of ideas people at Science would like to explore with the right kinds approaches, partnerships, and support. I’ll be interested in hearing your experiences in any similar types of projects and any recommendations you may have as to how we might proceed. You will not need to prepare anything in advance.”
Currently, “New Analytics” can be enabled on a per-course basis as a beta feature on UBC’s installation of Canvas. Instructure intends to disable “Old Analytics” and cutover to “New Analytics” in their March 2020 release.
LA Project team has started a mission to review the features of New Analytics, in comparison to Old Analytics. In this session, Dennis Foung, Support Analyst from the project, will walk us through an overview of the identified differences.
This will be followed by a group discussion around some of the outstanding issues with New Analytics. In preparation for the transition, UBC is working on a change management plan. Any feedback from the community around issues that require prompt attention or escalation to Instructure would help institution in this process.
Exploring the data from CLAS/WeVu: A video annotation platform for learning (Session 1 of 2)
Leah Macfadyen is a tenure track Instructor in the Department of Language and Literacy Education, UBC Faculty of Education and Associate Director of and Instructor in the UBC Master of Educational Technology Program. Leah will be leading one of two LAVA sessions exploring CLAS/WeVu and the data generated from it.
“In this introductory session, I will demonstrate and discuss how I have used a collaborative video annotation tool with students in an online course. CLAS (the Collaborative Lecture Annotation System) – now available as WeVu to users outside UBC, was developed in the UBC Faculty of Arts, and is actively in use in various UBC faculties and elsewhere to support learner engagement with video (as well as audio and now image) resources in a variety of ways: whole-class annotation of lecture videos, one to one instructor feedback on learner ‘performance’ of different kinds, asynchronous group ‘video seminars’ and more. CLAS/WeVu offers learners additional tools to support self-regulated learning, offers educators tools to gather and give richer feedback, and – I would argue – can support more effective social presence in virtual learning environments.
CLAS/WeVu collects some data that might be usefully explored to reveal learner usage patterns. I will review what data is currently available, and share some sample data (and a data dictionary) with you. So far, little use has been been made of this data to give instructors or learners richer feedback.
I invite you to join in a small collective exploration of the data (either independently or working with others), experiment with analyses or visualizations that might be useful, and to return to LAVA on Tuesday November 19th to demo what you have achieved. Recommendations for future improvements in data collection and data structure will also be welcome.”
A few published studies investigating CLAS data:
Mirriahi, N., Jovanovic, J., Dawson, S., Gašević, D., & Pardo, A. (2018). Identifying engagement patterns with video annotation activities: A case study in professional development. Australasian Journal of Educational Technology, 34(1). https://doi.org/10.14742/ajet.3207
Mirriahi, N., Liaqat, D., Dawson, S. et al. (2016). Uncovering student learning profiles with a video annotation tool: reflective learning with and without instructional norms. Education Tech Research Dev 64: 1083. https://doi.org/10.1007/s11423-016-9449-2
Pardo, A, Mirriahi, N, Dawson, S, Zhao, Y, Zhao, A & Gasevic, D (2015). Identifying Learning Strategies Associated with Active use of Video Annotation Software. in Proceedings of the 5th International Conference on Learning Analytics & Knowledge (LAK 2015). ACM Press, New York, NY, USA, pp. 255-259. https://doi.org/10.1145/2723576.2723611
Mirriah, N. & Dawson, S. (2013). The pairing of lecture recording data with assessment scores: a method of discovering pedagogical impact. Proceedings of the Third International Conference on Learning Analytics and Knowledge.doi>10.1145/2460296.2460331
In this session, we will be watching “Practical Analytics in Curriculum Complexity – A Summer Analytics Academy Webinar”, which was part of the Webinar series offered by the Centre for Student Analytics (Utah State University) over the summer.
In this webinar, Dr. Mitchell Colver discusses the curriculum tool https://curricula.academicdashboards.org/ in the framework of sustainable analytics. One of the big questions discussed in the webinar is “How do you socialize analytics in a practical way?”.
There a lot of points that are valuable to consider and discuss in our contexts at UBC. We don’t expect to get through the entire video, and our hope is for it to generate some discussion amongst ourselves as we go.
Shenia has been working as a work learn student at UBC Faculty of Arts over the summer, designing and developing student-facing dashboards for edX MOOC courses. In this session, Shenia walked us through her project:
“As a summer Work Learn student, I developed a student-facing dashboard for MOOC (Massive Open Online Course) students on edX. The dashboard design is similar to widget developed by Dr. Dan Davis, available here https://dan7davis.github.io/papers/LAL_Davis.pdf, surrounding the central idea that MOOC learners can benefit from understanding the learning habits of successful MOOC students in a past iteration of the course they are taking. This project spanned from May 2019 to August 2019 under the supervision of Ms. Sanam Shirazi at UBC Arts ISIT.”
Marko is a Student Learning Analytics Developer at Sauder. In this session, he will be telling us about the Canvas project that he has been working on this last term. From Marko:
“As a newcomer to Learning Analytics, I’ve spent the last several months exploring techniques for data collection, cleaning, and visualization in an effort to pull meaningful insights from Canvas data. During this LAVA meeting, I will discuss some of the interesting discoveries and talking points I’ve come across throughout this process — with an emphasis on the Tableau Web Data Connector (WDC) technology. I’ll be going through the process of building and using a Tableau WDC at a high level, as well as discuss some potential technical challenges and shortcomings with this tool. I’ll root my discussion in an example WDC and visualization that I’ve built, and pose some higher-level LA and Data Vis questions to the group (I’d love to learn more!). Note, the talk will be somewhat technical in spirit but doesn’t require any background knowledge on programming concepts or Tableau.”
We have three of the LAVA regulars presenting; Craig Thompson (Research Analyst, Learning Analytics Project), Alison Myers (Research Analyst, Sauder School of Business) and Sanam Shirazi (Senior Research Analyst, Faculty of Arts). We will also be joined by Igor Kuznetsov (Research Analyst, Planning and Institutional Research ).
In this session, will be talking about and demonstrating some projects related to “enrollment pathways & course sequencing”. Each of the presenters will talk about examples of questions and related output regarding course pathways/student program progression etc. The goals are to share some of the work we have each done in the area and generate discussion: about the questions being asked, the methods and tools used to answer the question, or other points of interest.
*Note – this meeting was rescheduled from February 11th
Firas Moosvi is a Learning Scientist working on the UBC Learning Analytics Project. In this session, he will discuss the potentials of Learning Analytics to surface classroom inequities.
“Though the field of learning analytics is new to me, I continue to be fascinated by its potential to transform teaching practice. For the first part of this LAVA meeting, I will present examples of how Learning Analytics has been used to surface trends of classroom inequities to instructors and institutions. Inequities include everything from gender, ethnicity, prior knowledge, and even personalities. I have a couple of papers already, but if you know of any examples that you thought are particularly striking, feel free to send them my way! In the second part, I hope to generate some discussion with the following prompt: “It is our responsibility as members of a university community to help highlight and address classroom inequities (where possible). No prior readings, come by for a fun and (maybe) spirited discussion! As always, standard disclaimer: views and opinions are my own.”
Stoo is a PhD Candidate in Educational Psychology at the University of Wollongong Australia, exploring the role that gestures play in cognition and learning. In this session, he will tap into multimodal learning analytics as a means to capture such physical interactions:
“To explore the role of gestures, a novel instrument for data collection was developed, which seeks to capture the physical interactions learners have with multimedia learning materials focusing on learning geometry. By leveraging the touch-based technologies in iPads, physical interactions with learning materials as well as quantitative data related to these interactions can be captured, and parsed. This novel instrument has implications for the field of Multimodal Learning Analytics (MMLA), potentially opening the door to a new field of Embodied Learning Analytics (ELA) and presents an interesting path forward for the measurement of embodied data, and how this may inform teaching and learning. This presentation will provide a brief theoretical overview, as well as a demo of the developed apps.”