Can you use Data to Optimise Student Learning?
Teaching & Learning in today’s landscape is becoming more digitalised as new technologies are introduced. Data, is a game-changing by-product with digitalised teaching activities. But harnessing that data for key insights can be challenging.
What data sources to gather and how to go about it?
How to prepare data for analysis and run predictive models?
How to extract insights from existing dashboards?
Elevate Teaching & Learning with Basic Analytical Techniques
Join us for this practical and hands-on 2-day workshop to learn more about tools you can adopt to extract valuable insights on students. Gain an overview of a multi-stage approach to Learning Analytics (LA) projects and be exposed to the practical applications of LA. Discover how to tap on built-in reports on LMS such as Moodle and perform advanced data collection. Conduct data exploration and modelling with Python. Tell compelling and informative stories with effective data visualisation. Uncover good questionnaire designs and practice processing qualitative data.
- Led by Dr Patrick Tran, Educational Developer at the UNSW Canberra
- Explore Moodle, a popular LMS and unlock built-in reports and other LA plugins available
- Experience advanced data extraction, exploration and modelling on Python with the assistance of Juypter Notebook
- Deep dive into qualitative data and semi-structured text data analysis
- Get a tour of and practice with tools such as:
- Microsoft Excel
- Microsoft Power BI Dashboards
Benefits of Attending
- Understand key features in defining and implementing various LA projects
- Extract learning data from built-in reports in Moodle and tap on ad-hoc reports and plugins
- Explore Python programming language and how to scrap key course data from LMS webpages
- Gain basic data wrangling techniques on Python such as loading, processing and visualising data
- Apply advanced predictive modelling and other popular algorithms on raw data
- Experience a visual, hands-on drag-and-drop approach to designing graphs and data stories
- Uncover best practices when designing learner questionnaires and conducing factor analysis
- Perform analysis on student surveys and highlight common themes sourced from data
- Distribute effectively visualised information and data on interactive mobile dashboards
- Hear guiding principles, themes and processes when building and managing a LA project
Dr. Patrick Tran
Educational Designer & Developer, Learning & Teaching Group,
University of New South Wales (UNSW) Canberra
Patrick is a data scientist by training, instructional designer by choice and educator at heart! His main interests lie at the intersection of innovation, technology, and leadership. Patrick received a PhD in computer science for a thesis on improving performance of network Intrusion Detection Systems using Machine Learning. He is also a certified PRINCE2 project manager with an MBA in Accounting and Finance.
Through applied research and teaching, Patrick seeks to leverage data analytics to drive pedagogical decisions, revamp learner experience and revolutionize existing approaches to learning and teaching.
He is currently working at the confluence of educational technology, learning analytics and educational research at University of New South Wales (UNSW) Canberra. His expertise lies in the domain of learning analytics, but Patrick has had a vast experience with all corners of the academic world, including research, teaching, academic program management and instructional design.
In the education space, Patrick has a wide range of research interests that are centred around learners such as:
• Learning Analytics and Educational Data Mining
• Educational Technologies
• Learning Theories
• Inventive problem solving
• Foresight – futures studies
Prior to UNSW, Patrick has held various positions in both teaching and management capacities at University of Technology Sydney, Victoria University and Australian National University. Over the years, he undertook several technology-enabled learning initiatives that involved analysing learners’ interaction data with intelligent tutoring systems and developing early intervention strategies for their success.
Who Should Attend
Lecturers, Teachers, Professors, Academics looking to implement learning analytics in the classroom. No prior experience in data analytics is required.
Registration: 8.30am • Workshop: 9.00am – 5.00pm
Morning, afternoon refreshments & lunch will be served at appropriate intervals.
Session 1: Understanding Learning Analytics (LA)
This session explains what LA is and why it is widely used by institutions today. We will dive into what is involved in defining and implementing LA projects.
- An overview of LA: the what and the why
- Education data: everything about learners and courses
- Recent developments and future outlook of LA
- LA projects: problem identification and goal settings
- A typical LA framework: how to approach your LA projects
Session 2: Learning Management System (LMS) tools for LA
This session provides a quick overview of LMS in general, and Moodle as the most popular LMS in particular. We will explore the power of built-in reports that allow extraction of learning data as well as supported customisable communication with learners. Advanced ad-hoc reports and tools are demonstrated as an intervention tool as well.
- LMS from the viewpoint of learner, instructor and administrator
- A quick tour of Moodle as the most popular LMS
- Use built-in reports for basic data extraction and communication
- Use ad-hoc reports and other LA plugins for advanced data collection and intervention
Session 3: Advanced data extraction with Python
This session introduces powerful Python packages for web harvesting, a technique to extract data from webpages. You will code your first web crawler program scraping course data from the LMS and other online systems.
- A quick tour of Python programming language
- The basics of the Web and HTML
- Build your first web crawler in Python: system login and data extraction
- Extract LMS data on course contents and activities
Session 4: Data exploration and modelling with Python
This session introduces basic data structures and techniques used to manipulate data sets. We will learn how Python can help with loading, processing and visualizing data which makes up the first step of data analysis: data exploration and preparation. We then explore several modelling techniques, starting from classical models to more advanced machine learning algorithms.
- Introduction to data structures and basic data operations
- Basic data wrangling and programmable data visualisation with Python
- Classical analysis: hypothesis testing and statistical modelling
- Advanced predictive modelling: popular algorithms
Session 5: Visual analytics: tools and techniques
This session introduces intuitive yet very powerful specialised system for visualising your data. You will experience a visual, hands-on drag-and-drop approach to designing graphs and data stories in Tableau. This is a great way to present your analytics findings to non-technical audiences. You will then see how easily, and effectively visualised information is distributed as interactive dashboards via a mobile app.
- Introduction to visual analytics
- Design data visualizations with Tableau: interactive graphs and simulation
- What story is your data telling?
- Online dashboard with MS Power BI: LA mobile apps
Session 6: Analysing qualitative data
This session covers a very important source of learners’ data: self-reported data or feedback, containing both quantitative data and semi-structured text data. You will learn factor analysis techniques and how to design a good questionnaire ready for your research. We then discuss thematic analysis techniques and the powerful NVIVO software that can be used to extract common themes emerging from the data.
- Factor analysis and questionnaire design
- Design and run your first learners’ experience survey
- Thematic analysis with NVIVO
Session 7: LA in action
This session discusses the non-technical key issues of LA and their impacts on individuals and the society as a whole. This covers potential pitfalls and caveats in using LA, best practices in managing LA projects. You will work as a team on a final project to solve a LA problem of choice using the skills and tools covered in this workshop.
- Key issues in the use of LA: ethics, data privacy and interpretation
- Managing LA projects: guiding principles, themes and processes
- Final LA project