An Instructor’s Guide to Learning Data, Analytics + Privacy

UBC’s Learning Analytics Project is focused on supporting teaching, learning, and the student experience. Part of the Project’s work is to ensure that learning analytics efforts are carried out in a manner that’s ethical, in agreement UBC policy and BC legislation, and sensitive to students’ and faculty members’ privacy concerns. This work is led by the Learning Data Committee and complemented by the work of UBC’s Data Governance Program.

Learn more about why UBC launched the Learning Analytics Project and about the learning analytics governance structures and principles that guide UBC’s approach to this emerging field.

What data is retained about faculty and students?

Canvas, as well as many other online platforms, retain logs of user activity. The data that’s collected through UBC’s online learning tools can be broken down into two main types of information, both of which can be referred to as learning data:

1. Data that users create when interacting with a learning tool (such as Canvas). This includes discussion posts that users create and the answers users provide when taking quizzes.

2. Data that a learning tool logs data about user activity. This includes when users log in, what files they download, which resources they use, and how often or how long they’re online.

Canvas collects both types of learning data about all users as they interact with the system, regardless of their role at UBC. Instructors have access to learning data from their own courses, which they can choose to use for quality assurance and course enhancement purposes. Learning data may also be accessed by UBC staff to fulfill requests for support from the course instructor. UBC Faculties and Departments do not access or use learning data for the measurement or comparative analysis of faculty teaching approaches or performance.

Learning data is collected and accessed according to best practices in privacy and security, UBC policy, and BC legislation, including BC’s Freedom of Information and Protection of Privacy Act (FIPPA).

Who has access to learning data? How is that data used?

UBC policy requires that personal information about instructors and students only be accessed as part of an employee’s official UBC duties. This is the overarching principle that determines who has access to learning data.

Learning data is made available primarily to faculty members and their teaching teams, as well as to senior administrators, support staff, software vendors, quality assurance investigators, researchers, and students.

Detailed examples outlining who can access learning data and in which circumstances are provided below.

Teaching staff

Faculty members and their teaching teams (e.g., TAs and any other staff that faculty members appoint to their teams) can access learning data from their own courses without requesting additional approval in order to improve their courses or their teaching, reflect on their prior courses offering, and generally speaking, for quality assurance purposes. These efforts are supported by UBC’s Learning Analytics Pilots and conducted independently by UBC faculty.

If a faculty member’s quality assurance activities will require access to additional data that’s not available within their own course, then a formal application needs to be submitted to the Learning Data Committee. (Please see the sections below on Quality Assurance and Researchers for more information.)

Faculty can also access some learning data from their own courses directly via Canvas Analytics or the Canvas API (application programming interface). Additionally, learning data that’s stored in Canvas Data (a reporting and analytical data warehouse) can be requested by contacting the Learning Technology Hub.

Learn more about the data and analytics tools that are available to faculty through Canvas and about other learning analytics tools that are currently being piloted at UBC.

Students

The Canvas API (application program interface) allows all users, including students, to download their own data and write programs that interact with Canvas. User permissions for the Canvas API are set in the same way as the web application — individuals can only use the API to access courses they’re enrolled in, and instructors have a greater degree of access than students.

Students who are interested in learning more about accessing and working with Canvas data are encouraged to sign up for UBC’s next Learning Analytics Hackathon, taking place in the fall of 2019.

Support staff

Some UBC support staff are authorized to access courses and learning data as part of their duties at UBC. This access is generally granted for the explicit purpose of supporting instructors with teaching tasks, troubleshooting technical issues, and answering students’ and instructors’ questions. Staff also may be appointed to assist with research and quality assurance projects that involve the use of learning data, as mentioned in the sections on Researchers and Quality Assurance below.

To be granted access to learning data, support staff need to sign an agreement that outlines their terms of access and the steps that they will need to take to maintain security and confidentiality. Alternatively, staff can be granted access to courses and learning data if they receive permission from the instructor.

In addition to the data that’s generated in courses, some technical support staff have access to the following information:

•    Users’ internet protocol (IP) addresses.

•    Device event information, such as crashes, system activity, hardware settings, browser type, browser language, the date and time of request, and referral URL.

•    On-page clicks and activity, such as logins, views, tool launches, file uploads, posts, and quiz submissions.

This information would typically only be accessed by support staff for technical support purposes or for system maintenance.

Researchers

Researchers or faculty members can request access to learning data for research projects being conducted at UBC or in collaboration with other institutions. To get permission to use learning data for research purposes, researchers or faculty members need to apply to the appropriate UBC Research Ethics Board.

If approval is granted by the Research Ethics Board, the request for access to data will then be reviewed by the person or group responsible for that data, called a data steward. The data steward for requests to access learning data is the Learning Data Committee.

If the data steward grants the researcher access to the requested learning data, this access will be granted with conditions of confidentiality and security. If the data steward decides that granting access to the requested data is not in the best interest of students or that granting access violates UBC policy, the steward can decide not to release the data — even if approval was received from the Research Ethics Board.

Faculty and staff involved in quality assurance activities

As mentioned above, faculty members have access to the learning data that’s generated in their own courses, which they can use for quality assurance purposes without needing to request additional permissions. The Office of Research Ethics has produced a checklist that can help you determine whether your efforts should be considered quality assurance activities or whether they constitute research and require ethical review.

In cases where ethical review is not needed, faculty, researchers, and UBC support staff can apply to the Learning Data Committee to request access to additional data they don’t already have access to (such as data from other courses) to support their quality assurance efforts. If access is granted, this decision will be reviewed annually.

Academic Administrators

Senior academic administrators, such as deans or associate deans, may request access to learning data in a few special cases. For example, if a student appeals a grade or requests a change of standing in a program, their learning data may be used to inform a decision in response their request. For some accredited programs, accreditation bodies require proof that specific learning outcomes are being met by the program as part of the accreditation review process. If learning data were provided for accreditation purposes, all identifying information (name and student number) would be removed.

Learning analytics for academic administrators is focused on supporting student learning and achievement, and does not include any comparative analysis or assessment of faculty teaching.

Software Vendors

Some UBC platforms and tools — Canvas for example — are operated by companies that are external to UBC. All vendors that work with UBC and that have access to personal information sign contracts requiring them to keep learning data secure and to protect students’ personal information as required by FIPPA and UBC policies.

UBC’s contracts with these vendors also set out the conditions for how vendors can request access to learning data. Generally speaking, access is permitted to allow vendors to update, test, and configure software and to improve system performance. When vendors access learning data for these purposes, your data would be combined with other students’ data and identifying details (such as your name and student number) would be removed.

Vendors may also be given access to your data temporarily to resolve technical issues. If needed, your data could be accessed by vendors outside of Canada. In that case, it would be handled as required by FIPPA, UBC policy, and the vendor’s contract with UBC.

Do you have comments or questions? Get in touch

If you have questions or comments about privacy, data access, or learning analytics at UBC, please get in touch using our online form.