Preliminary Program
We are currently working on a learning system that will accompany our program and provide centralised access to all relevant training materials, details, and updates. We are also in the process of inviting relevant speakers and mentors to support our participants; all confirmed speakers will be listed in the “People” section of this site.
Please note: The program outlined below is preliminary. Session descriptions may change at the discretion of our speakers and will be adjusted accordingly. The final program and materials will be shared with participants throughout May and June.
Week 1: Foundations, Methods & Theory (22 - 26 June 2026)
Day 1: Introduction to Computational Social Science (CSS)
Summary: All participants and speakers meet and get to know each other. The program begins with foundational discussions on CSS and its value for interdisciplinary research. The aim is to provide a clear understanding of the field, its opportunities, its limitations, and the work that needs to be done.
Session 1 (90 min) — Keynote Dialogue: What is Computational Social Science and Why It Matters in Australia?
Speakers: To be announced
We will discuss the foundational principles of CSS and its development in the Australian context. We will also talk about its unique value for research and its role in advancing what we know about our societies, providing an opportunity for participants to engage with the speakers, introduce themselves, and ask questions.
Learning Outcomes:
- Explain what CSS is.
- Identify its value in HASS&I (Humanities, Arts, Social Sciences, and Indigenous) and interdisciplinary research.
- Discuss opportunities for research in CSS across disciplines.
Session 2 (90 min) — Keynote: Bias in Computational Social Science
Speakers: To be announced
We will learn about common sources of bias in digital media data, particularly those that stem from data collection strategies, computational methods, and research designs. This keynote contributes to one of our main discussions: what are the limitations and opportunities of CSS and digital methods?
Learning Outcomes:
- Identify methodological bias.
- Recognise inherent limitations of CSS methods.
- Evaluate possible risks and opportunities of a chosen method.
Session 3 (90 min) — Panel: Meaningful Impact Using CSS Methods and Research
Speakers: To be announced (featuring senior computational scientists and practitioners)
We will discuss research that has been produced in the CSS area and made a tangible impact on policy or social issues. Together with our panel members, we will explore how CSS helps us solve issues from both academic and industry perspectives.
Learning Outcomes:
- Recognise practical applications of CSS methods.
- Identify areas of CSS application.
- Connect research and practical outcomes in the context of CSS.
Day 2: Working with Data: Ethics and Practices
Summary: The focus of today’s sessions is ethical foundations and responsible data practices. Participants will gain practical frameworks for ethical data workflows to help them understand the steps required before choosing a data collection methodology.
Session 1 (90 min) — Panel: Ethics in Computational Social Science
Speakers: To be announced
The panel will facilitate a discussion on one of the most critical issues when working with digital data: ethics. Topics will include complying with legislation and terms of service, issues of consent when working with participant data, and satisfying human research ethics committee (HREC) requirements.
Learning Outcomes:
- Make sense of HREC requirements.
- Develop an ethical data collection strategy.
- Understand the necessary steps to ensure compliance with FAIR and CARE principles.
Session 2 (90 min) — Workshop: Data Donations and Participant-Centric Research
Speakers: To be announced
We will discuss how to conduct studies that require participants to share their own social media data, alongside the ethical considerations that surround it. We will focus on participant enrolment, guiding participants through the process, decreasing attrition, keeping them informed, and managing their data securely.
Learning Outcomes:
- Design data donation studies.
- Collect digital trace data ethically.
- Manage participant attrition effectively.
Session 3 (90 min) — Keynote: Indigenous Data Sovereignty
Speakers: Representatives from ARDC and/or UniMelb Indigenous Data Network
We will introduce and discuss the principles of Indigenous Data Sovereignty in the context of CSS. We will identify ways to design and implement ethical data workflows with a strong focus on community rights and governance.
Learning Outcomes:
- Define Indigenous Data Sovereignty.
- Construct ethical data workflows.
- Recognise possible limitations and data governance considerations.
Day 3: Data Collection and Working Across Disciplines
Summary: Participants will explore different existing and emerging data collection strategies while learning how to work effectively in an interdisciplinary team. By the end of the day, attendees will know how to source and validate digital data responsibly.
Session 1 (45 min) — Panel: Cross-Disciplinary Collaboration
Speakers: To be announced
We will talk about what it means to be part of an interdisciplinary team and how to make sure collaboration works well. Drawing from individual experiences, panel members will discuss the challenges and rewards of collaborating across backgrounds and what considerations should be taken when managing a large project.
Learning Outcomes:
- Manage relationships with colleagues from different academic backgrounds.
- Identify personal goals when working in a team with competing priorities.
- Plan and structure interdisciplinary research projects.
Session 2 (60 min) — Workshop: APIs and Web Scraping: When They Work and When They Don’t
Speakers: To be announced
We will discuss the current state of data collection using APIs and web scraping, exploring possible use cases and the considerations required before choosing either method.
Learning Outcomes:
- Choose a data collection method based on research design.
- Plan a robust data collection strategy.
- Assess the methodological limitations of APIs and web scraping.
Session 3 (90 min) — Workshop: Collecting and Analysing Data Download Packages
Speakers: To be announced
We will introduce data download packages as an ethical and participant-centred approach to accessing digital trace data. Participants will learn how to set up secure collection pipelines, manage privacy requirements, and analyse donated datasets using computational tools developed or supported by the AIO.
Learning Outcomes:
- Design participant recruitment protocols for data donation.
- Analyse donated datasets while maintaining strict privacy standards.
- Evaluate the advantages and best use cases of data donation in CSS.
Session 4 (90 min) — Workshop: Working with Text Using Computational Techniques
Speakers: To be announced
We will demonstrate computational techniques for treating text as data in social science research. We will discuss resources and approaches that enable text processing, word frequency analysis, topic modelling, and sentiment analysis.
Learning Outcomes:
- Preprocess and transform data into analysis-ready formats.
- Implement fundamental text analysis techniques.
- Assess the suitability and limitations of different text analysis methods.
Day 4: Tools and Approaches to Data Analysis
Summary: We move towards methods that bridge social science and computation. Workshops cover qualitative and quantitative approaches, moving from annotation and qualitative foundations to the integration of LLMs and Retrieval-Augmented Generation (RAG).
Session 1 (60 min) — Workshop: Qualitative Text Analysis for Gold-Standard Datasets
Speakers: To be announced
We focus on creating high-quality, annotated “gold standards” for training, validating, and evaluating machine learning models and LLMs in CSS. Participants will learn systematic qualitative coding techniques, codebook development, and strategies for ensuring annotation consistency.
Learning Outcomes:
- Develop detailed codebooks tailored to social science research questions.
- Apply rigorous qualitative coding to build high-quality datasets.
- Understand how qualitative foundations enhance the validity of computational analyses.
Session 2 (60 min) — Workshop: Integrating LLMs in Research Workflows
Speakers: To be announced
We will learn how to use Large Language Models (LLMs) in research beyond simple chat windows. We will discuss how complex models can be integrated into research pipelines—facilitating work or hindering it—and explore ways to prevent unintended outcomes.
Learning Outcomes:
- Integrate LLMs into comprehensive research workflows.
- Deploy LLMs for advanced data analysis.
- Critically analyse LLM outputs and their suitability for academic research.
Session 3 (60 min) — Workshop: RAG 101
Speakers: To be announced
We will introduce Retrieval-Augmented Generation (RAG) and its use in research. We will focus on how RAG can facilitate more grounded research and aid in the data triangulation necessary for reliable results.
Learning Outcomes:
- Define RAG and understand its mechanics.
- Identify applications of RAG in research contexts.
- Use RAG for data retrieval and triangulation.
Session 4 (90 min) — Workshop: Screen Capture for Data Collection
Speakers: To be announced
Using the Mobile Ad Observatory as an example, we will discuss when and how to collect and analyse video recordings from users’ screens. We will cover the technical and ethical requirements for this form of data collection and how to approach the subsequent analysis.
Learning Outcomes:
- Design data donation studies that involve screen capture.
- Use screen capture safely and effectively as a source of participant data.
- Understand the analytical steps involved in processing screen capture data.
Day 5: Theory, Careers, and Industry
Summary: The final day of Week 1 discusses real-world applications of CSS and career pathways. Participants will then shift their focus to teamwork, preparing the research questions they will explore collaboratively in Week 2.
Session 1 (60 min) — Panel: Does Computational Social Science Lack Theory?
Speakers: To be announced
This panel addresses the debate surrounding the theoretical foundations of CSS. We will reiterate the importance of the theoretical component and discuss how choosing a specific computational tool can drastically alter research outcomes.
Learning Outcomes:
- Justify methodological choices using a theoretical framework.
- Employ theory-based approaches to computational data.
- Analyse the broader implications of computational methods.
Session 2 (90 min) — Panel: Working With and In the Industry
Speakers: To be announced
Bringing together computational social scientists working in (or closely with) industry and government organisations, this panel will share insights on career transitions, highly-valued skills, and real-world applications of CSS beyond academia.
Learning Outcomes:
- Understand career pathways outside of traditional academia.
- Identify key CSS competencies sought by employers.
- Learn practical strategies for translating academic research into industry impact.
Session 3 (90 min) — Workshop: Success in PhD Candidacy
Speakers: To be announced
This workshop equips HDR candidates and ECRs with strategies for project planning, timeline management, supervisor communication, milestone navigation, and building a strong research profile in Australia.
Session 4 (60 min) — Team Formation and Project Ideation
Speakers: AIO and ADM+S mentors
Guided by AIO and ADM+S mentors, participants will be presented with datasets and open research questions across multiple domains. Participants will form teams around topics of shared interest in preparation for Week 2.
Week 2: Collaborative Research Projects (29 June - 3 July 2026)
Days 6 - 9 (Monday - Thursday)
Summary: The second week transitions from formal instruction to hands-on, participant-led group research. Mornings will feature advanced methodological workshops, while the rest of the day is dedicated to collaborative teamwork supported by drop-in experts.
Morning Sessions (90 min) — Advanced Workshops
Throughout the week, the following morning workshops will be delivered:
1. Workshop: Validation in Computational Social Science
Speakers: To be announced
Focusing on rigorous validation techniques essential for trustworthy CSS outputs, participants will explore methods for human evaluation and strategies for addressing validity threats.
- Learning Outcomes: Mitigate common validity threats, assess the robustness of research designs, and produce transparent, reproducible research.
2. Workshop: Network Analysis in CSS
Speakers: To be announced
An introduction to network analysis as a method for understanding relationships, influence, and community structures within digital data.
- Learning Outcomes: Apply network metrics (centrality, clustering, community detection), understand methodological limitations, and interpret results in social/cultural contexts.
3. Workshop: Simulation in Social Science
Speakers: To be announced
Exploring agent-based modelling and simulation techniques for studying complex social phenomena that are difficult to investigate through traditional methods.
- Learning Outcomes: Design basic agent-based models, use simulations to test theoretical hypotheses, and evaluate the strengths and limitations of simulation methods.
Mid-Morning & Afternoon Sessions (90 min blocks) — Teamwork & Project Development
Speakers: Drop-in experts, organisers, and mentors
Participants work exclusively in their formed groups on their collaborative research projects. Drop-in experts, organisers, and mentors will be available to provide technical assistance, theoretical guidance, and feedback.
Day 10 (Friday)
Project Presentations & Closing
Participants will present the preliminary findings, methodologies, and proposed solutions from their collaborative week-long projects. The program will conclude with final feedback, networking, and closing remarks.
The Australian Internet Observatory received co-investment from the Australian Research Data Commons (ARDC) through the HASS and Indigenous Research Data Commons. The ARDC is enabled by the National Collaborative Research Infrastructure Strategy (NCRIS). Please share your feedback by emailing us at aio@rmit.edu.au.