There are multiple definitions of learning analytics such as follows:
Learning analytics refers to the collection and analysis of data about learners and their environments for the purpose of understanding and improving learning outcomes. Learning analytics can be used to measure key indicators of student performance, support student development, understand and improve the effectiveness of teaching practices, and inform institutional decisions and strategy. Read more
Learning analytics is the measurement, collection, analysis, and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs. Read more
Learning Analytics can be performed on learning management system access logs, marks records and reflections records. Learning Analytics is provided as a plugin in popular learning management system (LMS) such as Moodle. It has been useful for online learning providers such as CASSMiLe in monitoring the learners activity and could be used as insights to improve the learning design. Example interface on learning analytics in CASSMiLe is provided here.
Tools such as Tableau, PowerBI, RapidMiner and Python. Visualizations can be performed through exploratory data analysis and while prediction can be conducted through predictive analytics. Below are some example case studies and presentation from this project.
Grades Chart: The grades distribution in a box graph to identify the differences among evaluations and students with problems.
Content Accesses Chart: Which users accessed many different resources.
Number of Active Users Chart: How many users are active in a certain time of day.
Assignment Submissions Chart: Which users have submitted assignments on time or late (tasks, quizzes and hot potatoes).
Hits distribution Chart: How each user is accessing the course and its resources in each course week.
Engagement analytics: provides information about student progress against a range of indicators. Engagement refers to activities which have been identified by current research to have an impact on student success in an online course.
Rubrics-enriched grading: Criteria based assessment that are associated to data from the analysis of learners’ interaction and learning behavior in a Moodle course, such as number of post messages, times of accessing learning material, assignments grades and so on.
Behavior analytics: links between nodes of the student whom accesses those activities.
Intelligent content recommendations (adaptive content): focused on maximizing students’ performance and retention.
Advanced predictive learning analytics: for a proactive action on ongoing learning processes.
On the right are the slides from the presentation from this project at ICADEIS2020 which provides the proposed model for the learning analytics that can support personalised learning by instructors.
Publications:
Sharef, N.M., Murad, M. A. A., Mansor, E. I., Nasharuddin, N. A., Omar, M. K., Samian, N., Arshad, N. I., Ismail, W., Shahbodin, F., (2020), "Learning-Analytics based Intelligent Simulator for Personalised Learning", International Conference on Advancement in Data Science, E-learning and Information Systems (ICADEIS2020)
The example below shows how learning analytics can be used to analyze the engagement of students in the learning management system (LMS) and their performance in the course. PowerBI has been used for this purpose. The dataset is also provided in the link.
The learning analytics dashboard could answer the following questions:
1) What is difference cross access frequency to the LMS throughout the semester?
2) What activities have resulted to increase or decrease of students access to the LMS?
3) What is the distribution of the learning events and components to the LMS throughout the semester?
4) How is the performance of the students in each assessment?
5) What is the relationship between the learning engagements (indicated by access frequency) with the students achievement (indicated by their marks)?
These questions have led to the following insights:
1) The frequency of access spikes in the weeks when test is conducted online.
2) When assessment and forum are conducted in the LMS, the number of access increased.
3) The variation of the students performance is relatively small.
4) Generally, students whom have higher access frequency obtain better performance in the course.
This example is built using Tableau to demonstrate how learning analytics can be used to analyze achievement of students within a cohort, cross semester and cross groups.
Records from two groups are analyzed, where focus is given to difference between achievements of students from different demographics and instructor.
More examples on learning analytics below:
https://sites.google.com/student.upm.edu.my/dataminingssk4602-assignment1/home
1.Does the student’s performance maintain from beginning to the end or is there any improvement?
2.What are the average of the students performance?
3.Is there any pattern on how much the students spend their time according to the assignment given?
4.Which activity is the most accessed in each of the component?
5.What are the student’s behaviour when accessing the learning material?
https://sites.google.com/student.upm.edu.my/lms-data-analysis/results-and-analysis?authuser=1
1) What is the popularity of each context by all students?
2) What is the popularity of each context by each student?
3) What is the popularity of each context by week?
5) What is the achievement by all students?
6) What is the achievement for each individual student?
7) What is the relationship between the engagement and achievement in the online learning environment?
8) What is the trend of context shifting by time?
9) Which period in terms of semester week that students engaged in the course the most?
10) How does the student’s performance change throughout the semester?
11) What is the changes in terms of quantity of context throughout the semester?
https://sites.google.com/student.upm.edu.my/ssk4604-data-mining/home?authuser=1
1. What is the trend of access to each learning event?
- Which URL has the highest access?
- Which Folder has the highest and lowest access?
- Which event has the highest popularity among student?
2. Identify student access above average and lower average.
3. What is the trend of access on the whole semester?
4. What is the carry mark of students throughout the semester?
5. What is the distribution of grade?
6. What is the performance for every student?
https://sites.google.com/view/ssk4604-portfolio/introduction?authuser=0
Does a higher access to activities result into a higher grade?
Which Event has been visited the most overall by all students
Does getting a lower Test 1 marks increased a student’s access to the site?
Does access increase or decrease throughout the semester? Is accesses higher at the beginning or end ?
Is there any big difference between the access rate of the student with the highest grade vs the student with the lowest grade?
https://sites.google.com/view/dataminingtrex/home?authuser=0
1. Which assessment has the highest marks ?
2. What is the relationship between the frequency of student access and student's performance
3. How many students are more likely to get a good result for their final ?
4. Which student needs the most attention ? (based on carry marks)
5. What is the activities that has the highest frequency accessed by the learners ?
6. Which week has the most learning materials posted ?
7. Which week have the highest frequency of access?
https://sites.google.com/view/dm-proj-ssk4604-s2-2021-jsdp/homepage?authuser=1
1. To identify the popularity of each event context by the whole students by each student by each week
2. To identify the level of students capability when learning with exercises by each student
3. To identify which students perform the best throughout the course of learning
4. To identify the which day used mostly by student to keep on with the activity
5. To identify the student performance at the end of semester
IADLEarning https://www.iadlearning.com/learning-analytics-dna/
LearnerScript https://learnerscript.com/
13 startups leading learning analytics worldwide https://tedsf.org/12-startups-leading-learning-analytics-worldwide/ (accessed 19th Jul 2021)
Global Learning Analytics Market Size USD 7956.6 Million by 2026 at CAGR 20.8% | Valuates Reports https://www.prnewswire.com/in/news-releases/global-learning-analytics-market-size-usd-7956-6-million-by-2026-at-cagr-20-8-valuates-reports-899364130.html (accessed 19th Jul 2021)
Alta-Adaptive Learning Technology https://www.knewton.com/the-power-of-altas-adaptive-technology/ (accessed 19th Jul 2021)
How higher-education institutions can transform themselves using advanced analytics https://www.mckinsey.com/industries/public-and-social-sector/our-insights/how-higher-education-institutions-can-transform-themselves-using-advanced-analytics (accessed 19th Jul 2021)
Society for Learning Analytics Research (SOLAR) https://www.solaresearch.org
Course Insights https://elearning.uq.edu.au/guides/course-insights
RiPPLE: An active, social and personalised learning platform https://itali.uq.edu.au/resources/learning-analytics/ripple
789 Educators Agree: This Is The Mapping Of Urgent Shifts We Need In 2021 https://www.lmspulse.com/2021/5-urgent-shifts-we-need-in-education-in-2021/ (accessed 22nd Jul 2021)
Moodle Learning Analytics with Power BI https://blogs.kcl.ac.uk/digitaleducation/moodle-learning-analytics-with-powebi/