Understanding the role of learning analytics

The 2023 EDUCAUSE Horizon Action Report: Holistic Student Experience Edition emphasizes the importance of "unified data models for learning analytics" as a transformative technology expected to significantly impact higher education. 

Learning analytics starts with the systematic collection of data from various educational platforms, particularly Learning Management Systems (LMS) and other digital tools that track student activity. This data can include log-ins, time spent on tasks, engagement in discussion boards, and results from assessments such as quizzes and assignments. 

Siemens (2013) highlights that this data is diverse and comprehensive, capturing both academic performance and behavioral aspects of learning, such as student interactions with content and peers. These datasets form the foundation for deeper analysis, which can offer educators insights into how students are engaging with their courses. By systematically analyzing this data, several key benefits emerge:

Identifying patterns and trends

Learning analytics enables educators to translate raw data into actionable insights, revealing relationships between student engagement behaviors—such as participation in discussion forums—and their performance in assessments (Gašević, Dawson, & Siemens (2015). 

Aligning with competency-based education (CBE) 

Learning analytics tracks students' progress toward well-defined learning outcomes, allowing for personalized learning paths and instructional adjustments to meet individual needs.

Informing educational strategies and improving learning outcomes

Ultimately, the true value of learning analytics lies in its ability to enhance educational strategies and outcomes, enabling institutions to personalize learning and reinforce competency mastery, essential in CBE frameworks.

Transforming data into strategic insights

Learning analytics provides institutions with the tools to not only monitor and assess student progress but also make informed decisions that directly support student success and institutional goals. Below are actionable steps that demonstrate how institutions can use learning analytics to achieve these goals, supported by concrete examples.

Identifying trends in student behavior and performance: Institutions can use learning analytics to monitor student engagement, participation, and performance over time. By tracking these metrics, educators can identify trends that indicate how students are progressing through courses and programs. These trends can reveal which areas of the curriculum are effective and which may need improvement.

Actionable example:

• Using a tool like FeedbackFruits student engagement dashboard, faculty can track participation levels in key activities such as discussion posts, assignments, and quizzes. If data shows a drop in engagement during a particular module, educators can intervene by adjusting the content or delivery method to re-engage students.
• At the program level, administrators can use Cohort Exports to review data across multiple courses and semesters, identifying areas where students consistently underperform. If students in a certain major struggle with specific competencies, additional resources, such as tutoring or redesigned assignments, can be allocated to that area.

Supporting early interventions for at-risk students: Learning analytics allows institutions to detect early warning signs for students who may be at risk of failing or dropping out. By analyzing metrics such as low engagement, poor performance, or failure to meet deadlines, educators can identify these students and offer timely support, preventing long-term academic struggles.

Actionable example:

• Instructors can use the FeedbackFruits Student engagement overview table to monitor activity completion rates, contribution levels, and overall engagement. If a student has not participated in discussions or submitted assignments for several weeks, the system can flag them as at-risk. Faculty can then reach out to the student, offer additional support, and adjust learning plans if necessary.
• Predictive analytics from FeedbackFruits cohort exports can identify students who are at risk based on historical data and trends, allowing advisors to intervene early. For example, if students with low engagement in the first few weeks of a course have historically failed, faculty can reach out to those students early on and offer targeted support or advising to get them back on track.

Making informed curriculum decisions based on data: Data-driven insights from learning analytics can help institutions make informed decisions about their curriculum and instructional methods. By understanding how students perform across different competencies and areas of study, faculty can refine course content to better align with student needs and institutional goals.

Actionable example:

• FeedbackFruits student success dashboard highlights trends in student competency development across a course or program. If analytics show that students consistently struggle with a particular learning objective, educators can redesign that part of the curriculum to include more interactive materials, additional resources, or different assessment methods.
• At the program level, administrators can use cohort data to adjust entire programs. For example, if a significant percentage of students in a program are struggling with advanced writing skills, institutions can implement a supplemental writing lab or adjust the sequence of courses to better prepare students for these challenges.

Optimizing resource allocation: Learning analytics enables institutions to allocate resources more effectively. By identifying areas where students need more support or areas where faculty could benefit from additional professional development, institutions can focus their efforts where they are most needed.

Actionable example:

• An institution might notice through FeedbackFruits student success overview that a large number of students are failing a particular course or competency. In response, additional teaching assistants can be allocated to help students in that course, or the institution might invest in additional professional development for faculty to enhance teaching methods in that area.
• Similarly, if data shows that students in certain courses are highly engaged but still underperform on assessments, it may indicate the need for more hands-on or practical applications of the material. Resources could be shifted to include more lab time, experiential learning opportunities, or one-on-one tutoring.

Continuous improvement through feedback loops: Institutions can use learning analytics to establish continuous feedback loops, where data is constantly reviewed and adjustments are made to improve the learning environment. Over time, this process leads to a more agile and effective educational strategy.

Actionable Example:

• Faculty can use FeedbackFruits course exports to review student performance data after each term, identifying patterns in student learning and areas for improvement. For example, if students consistently struggle with group assignments, instructors can adjust the structure of these tasks or provide more detailed guidance on how to collaborate effectively.
• By continuously analyzing learning data, institutions can set goals for improvement each semester. For example, if retention rates are low in the first-year experience program, institutions can design early intervention strategies such as mentoring programs or specialized workshops based on insights from learning analytics.

References

1. Siemens, G., & Long, P. (2011). Penetrating the fog: Analytics in learning and education. EDUCAUSE Review, 46(5), 30–40.
2. Alzahrani, A. S., Tsai, Y. S., Aljohani, N., Whitelock-Wainwright, E., & Gašević, D. (2023). Do teaching staff trust stakeholders and tools in learning analytics? A mixed methods study. Educational Technology Research and Development, 71(4), 1471–1501. https://doi.org/10.1007/s11423-023-10229-w
3. Education Analytics. (2024). Reporting & dashboards: Data-driven insights for educational improvement. https://www.edanalytics.org/what-we-do/reporting-dashboards