0.10 wombat-2025 2025-09-29 2025-09-30 2 00:05 https://conf.nectric.com.au Australia/Melbourne 8.42 Workshop 2025-09-29T09:00:00+10:00 09:00 03:30 Organisations of all types collect vast amounts of time series data, and there is a growing need for time series analytics to understand how things change in our fast-moving world. This tutorial provides a practical introduction to time series analytics and forecasting using R, utilising the tidyverse and tidy time series tools to enable analysis across many time series. Attendees will learn about commonly seen time series patterns, and how to find them with specialised time series graphics created with ggplot2. Then we will use fable to capture these patterns with statistical time series models, and produce probabilistic forecasts. Finally, participants will gain insights into evaluating model performance, ensuring the accuracy and reliability of their forecasts. Through a combination of foundational concepts and practical demonstrations, this tutorial equips participants with the skills to extract meaningful insights from time series data for informed decision-making in various domains. Learning Outcomes: * How to use the tidyverse to wrangle and manipulate time series data. * Visualise data and identify common time series patterns. * Produce forecasts from a statistical model that captures dynamic time series patterns. * Evaluate the model’s forecasting performance to select the best model. wombat-2025-27-tidy-time-series-analysis-and-forecasting Mitchell O'Hara-Wild en false https://conf.nectric.com.au/wombat-2025/talk/VDYBYE/ 8.42 Workshop 2025-09-29T13:30:00+10:00 13:30 03:30 This hands-on workshop introduces researchers, students, and academics to **Quarto**, a powerful, open-source framework for creating reproducible academic documents and presentations. You'll learn how to write a report, a journal article using a ready-to-use template, and build polished, code-integrated slide decks for presentations. The workshop will also cover bibliography management, embedding code and figures, formatting tables and equations, and controlling output for various formats, such as HTML and PDF. By the end of the session, participants will be able to confidently use Quarto to produce reproducible manuscripts and dynamic, professional presentations. wombat-2025-25-reproducible-reporting-and-research-with-quarto Jayani LakshikaKrisanat Anukarnsakulchularp en false https://conf.nectric.com.au/wombat-2025/talk/Q3BPJZ/ 8.43 Workshop 2025-09-29T09:00:00+10:00 09:00 03:30 From exploring variability in a dataset to communicating the distribution of estimates, uncertainty visualisation plays a role at every stage of data analysis. This tutorial provides an overview of uncertainty, examines how it is represented in R, and introduces a range of techniques for visualising it. The first session will introduce the concept of uncertainty and cover general visualisation approaches. The second session will focus on spatial data and the creative methods that emerge when uncertainty must be expressed using only a limited set of visual aesthetics. Wherever there are statistics and data, there is uncertainty, making the ability to visualise it effectively an essential skill for any statistician. wombat-2025-26-visualising-uncertainty Dianne CookHarriet Mason en false https://conf.nectric.com.au/wombat-2025/talk/LDTUU8/ 8.43 Workshop 2025-09-29T13:30:00+10:00 13:30 03:30 Clear, well-designed figures are essential for effectively communicating scientific ideas, but good design is often overlooked in academic training. This practical workshop, led by Dr Jess Hopf from Knowlegible Designs, introduces core graphic design principles and demonstrates how they can be applied to scientific figures to improve clarity, impact, and accessibility. Through a mix of theory and hands-on application, we’ll cover cognitive load, visual hierarchies, common communication pitfalls (like the curse of knowledge), and a step-by-step process for building and refining figures. Attendees are encouraged to bring a visual they’re currently working on, as there will be opportunities to apply workshop concepts and receive feedback. You don't need any prior design experience or specialist software. This workshop is suitable for researchers at any career stage who want to elevate the quality and communicative power of their visual materials. wombat-2025-23-building-better-figures-a-scientific-graphic-design-workshop Jess Hopf en false https://conf.nectric.com.au/wombat-2025/talk/XY8EW8/ 8.44 Workshop 2025-09-29T09:00:00+10:00 09:00 03:30 Do you have R code you want to share? You can transform your R code into a shareable tool by developing it into an R package. This course bridges the gap between writing isolated R functions and creating documented packages that can be easily distributed, and is designed for those who might be curious about R package development but haven't had the time or guidance to get started. You'll learn the essential tools of the trade: `usethis` and `devtools` for package structure, `roxygen2` for documentation, `testthat` for testing, and Git/GitHub for sharing your work. Whether you're looking to organise your personal code collection or contribute to the broader R ecosystem, this course provides the foundation you need. **Prerequisites**: - Comfortable with some R fundamentals (data types, functions, reading data). - Experience writing basic R scripts. - No prior experience with package development required. **Learning Outcomes**: By the end of this course, you will be able to: - Create the basic structure of an R package. - Manage dependencies with `usethis` and `devtools`. - Create documentation with `roxygen2`. - Write and run unit tests with `testthat` to verify package functionality. - Use Git and GitHub to put your R package online. - Understand next steps for advanced package development, including: - Automatically run tests with continuous integration via GitHub Actions. - Make your R package easily installable with the [R Universe](https://r-universe.dev/search). - Create professional package websites using `pkgdown`. wombat-2025-22-introduction-to-r-packages Nicholas Tierney en false https://conf.nectric.com.au/wombat-2025/talk/PLVPUV/ 8.44 Workshop 2025-09-29T13:30:00+10:00 13:30 03:30 Code written in C++ runs significantly faster than code written in R, which provides the basis for many leading R packages. Such a design enables users to leverage the best of both worlds: the computational speed delivered by algorithms written in C++ and the convenience of data analysis using R. Fortunately, recent developments have simplified programming in C++ for R applications by automating many processes. In this session, Tomasz will guide attendees through the basics of this approach, working in RStudio, ensuring object compatibility, using basic algorithmic structures and functional programming, and extending R packages with C++ code. A sequence of hands-on exercises with applications to every statistical modeller toolset, including linear algebra, maximum likelihood estimation, and Gibbs sampling, supports all this. wombat-2025-24-getting-started-with-c-for-faster-statistical-modelling-in-r Tomasz Woźniak en false https://conf.nectric.com.au/wombat-2025/talk/KCZEMB/ 8.53 Workshop 2025-09-29T09:00:00+10:00 09:00 03:30 Regression analysis is one of the most widely used and powerful quantitative analysis methods used across almost every sector. This workshop provides a comprehensive introduction to commonly used regression models, including the theory behind these models, their application in R, validation techniques, and the interpretation of results. The course begins with an introduction to linear regression models, before advancing to the more flexible family of generalised linear models. Topics covered as part of this course include: * Linear regression: concepts, assumptions, application, and interpretations * Diagnostics and validation of linear regression models * Generalised linear models: beyond continuous outcomes * Poisson regression: how to model counts and rates, and how this differs from linear regression * Best practices in communicating results of regression analysis Participants will leave with a solid grounding (or refresher) in how to correctly perform hands-on regression modelling tasks in R for a variety of problems. wombat-2025-29-introduction-to-regression-analysis-in-r Dean Marchiori en false https://conf.nectric.com.au/wombat-2025/talk/9X7BMV/ 8.03 Talk 2025-09-30T08:45:00+10:00 08:45 00:15 Welcome to WOMBAT! An overview of the day, safety information, and housekeeping. wombat-2025-36-welcome Mitchell O'Hara-Wild en false https://conf.nectric.com.au/wombat-2025/talk/JDA7Q9/ 8.03 Keynote 2025-09-30T09:00:00+10:00 09:00 01:00 Data visualisation can be a very efficient method of identifying patterns in data and communicating findings to broad audiences. Good data visualisation requires appreciation and careful consideration of the technical aspects of data presentation. But it also involves a creative element. Authorial choices are made about the “story” we want to tell, and design decisions are driven by the need to convey that story most effectively to our audience. Software systems use default settings for most graphical elements. However, each visualisation has its own story to tell, and so we must actively consider and choose settings for the visualisation we are building. In this talk, Nicola will showcase why you should visualise data, present some guidelines for making more effective charts, before discussing examples of good and not so good charts. wombat-2025-30-designing-for-decision-making-how-to-build-effective-data-visualisations Nicola Rennie en false https://conf.nectric.com.au/wombat-2025/talk/MBAGKH/ 8.03 Talk 2025-09-30T10:30:00+10:00 10:30 00:30 Scrollytelling is an innovative way to tell data-driven stories, but it often demands sophisticated data visualisation engineering skills and budgets beyond individual researchers and small newsrooms. [Closeread](https://closeread.dev/) makes scrollytelling accessible and easy for researchers and analysts using their existing Quarto authoring skills. ​In this talk, James discusses some of the factors that motivated Closeread's design, the first scrollytelling contest with Posit, and new features on the horizon, including scrolling video support and better integration with R and Python tools. wombat-2025-31-closing-the-scrollytelling-gap-with-closeread James Goldie en false https://conf.nectric.com.au/wombat-2025/talk/AN9GFB/ 8.03 Talk 2025-09-30T11:00:00+10:00 11:00 00:30 Generative AI tools, like ChatGPT, haven’t necessarily created new challenges in data science education. Rather, the availability and rapidly growing capabilities of these tools have exposed weaknesses in learning design and assessment that educators have been ignoring for a while. In this talk, I will draw on my experience in teaching psychology students about computational reproducibility and the value of R-based data workflows. I will argue for the need to rethink what learning to learn looks like and talk about how we might reframe assessment as the process of gathering evidence that learning has occurred. There is no avoiding the fact that students now need to learn how to learn alongside Generative AI. Data science educators can help them by designing learning experiences that build self-regulated learning skills and normalise what real learning feels like. wombat-2025-38-rethinking-data-science-education-in-the-age-of-genai Jenny Richmond en false https://conf.nectric.com.au/wombat-2025/talk/79GEP8/ 8.03 Talk 2025-09-30T11:30:00+10:00 11:30 00:30 Nothing gives me greater joy than building little tools to make my life easier — even if they take ten times longer to make than they’ll ever save. In this talk I’ll share a few of these creations and the mindset behind them: the `teleport` GitHub Action, _Comment Viewer_ Positron extension, _Unilur_ Quarto filter for generating assignment solutions, and a sneak peek at a gamified teaching tool for ggplot2. Along the way, I’ll argue that sometimes it’s not the smartest hire you need, but the laziest one — a sentiment probably attributed to Bill Gates (though I’ve never bothered to check). wombat-2025-42-when-laziness-leads-to-innovation-making-things-you-didn-t-know-you-needed Michael Lydeamore en false https://conf.nectric.com.au/wombat-2025/talk/L8QMTF/ 8.03 Talk 2025-09-30T13:00:00+10:00 13:00 00:20 Preferential datasets usually represent rankings for an individual voter across multiple columns or rows in a rectangular format. This can make working with these datasets cumbersome and unintuitive. This talk will introduce `prefio`, an R package that lets you work with preferences in a tidy way without spanning multiple rows or columns, while providing convenient operations for manipulating preferential data. There will be an interactive component to the talk, so get ready to cast some votes! wombat-2025-40-tidy-analysis-of-preferential-votes-with-prefio Floyd Everest en false https://conf.nectric.com.au/wombat-2025/talk/SAPKKB/ 8.03 Talk 2025-09-30T13:20:00+10:00 13:20 00:20 Adding uncertainty representation in a data visualisation can help in decision-making. There is an existing wealth of software designed to visualise uncertainty as a distribution or probability. These visualisations are excellent for helping understand the uncertainty in our data, but they may not be effective at incorporating uncertainty to prevent false conclusions. Successfully preventing false conclusions requires us to communicate the estimate and its error as a single “validity of signal” variable, and doing so proves to be difficult with current methods. In this talk, we introduce ggdibbler, a ggplot extension that makes it easier to visualise uncertainty in plots for the purposes of preventing these “false signals”. We illustrate how ggdibbler can be seamlessly integrated into existing visualisation workflows and highlight the effect of these changes by showing the alternative visualisations ggdibbler produces for a choropleth map. wombat-2025-39-visualising-uncertainty-with-ggdibbler Harriet Mason en false https://conf.nectric.com.au/wombat-2025/talk/YXKDVV/ 8.03 Talk 2025-09-30T13:40:00+10:00 13:40 00:50 Research software engineering, open source software, reproducibility research -- what do these terms really mean, and what's special about creating 'statistical software' and R packages? In this conversation with Professor Rob Hyndman, and Dr. Nick Tierney, we attempt to demystify and highlight many ways that software design and open source packages can contribute to the development, dissemination and accessibility of data-driven analysis. wombat-2025-32-in-conversation-rob-hyndman-nick-tierney-on-research-software-hosted-by-cynthia-huang Nicholas TierneyCynthia HuangRob Hyndman en false https://conf.nectric.com.au/wombat-2025/talk/G9CLV3/ 8.03 Talk 2025-09-30T15:00:00+10:00 15:00 00:30 Code for research is more flexible than point-and-click statistical softwares, but can be more error-prone. These errors may be conceptual (e.g., implementing the wrong function for a given task), programmatic (e.g., indexing the wrong column of a data frame), or syntactic (e.g., the incorrect spelling of a statement or function). Although peer review is part of the scientific process, it rarely (though increasingly) involves review of research code. Part of the difficulty with implementing formal peer review of code is the time, expertise, and supportive environment required to successfully execute it. Particularly for more involved analyses, review of code requires significant time to understand the context, questions, data, methods, and aims. It can also be very difficult to identify people with the appropriate skills in both the given code language and the methods to effectively review the code. Lastly, despite peer review itself being a common and integral part of the research process, people are still less prepared to open up their code itself for review and so a constructive, supportive, peer space is necessary. We present a pilot program for peer code review conducted within a research consortium setting, which may represent a useful model to overcoming these challenges. The Australia-Aotearoa Consortium for Epidemic Forecasting & Analytics (ACEFA) aims to support timely and effective responses to epidemic diseases in Australia and New Zealand through real-time data analytics, modelling, and forecasting. One of our main activities this winter is reporting short-term forecasts for daily case counts of several respiratory pathogens, for each Australian state and territory and for New Zealand, to government health committees and stakeholders. We have multiple forecasts from models developed and maintained by one or more research academics in our consortium. In parallel with a review period for the methods in these models, we are also planning to conduct peer code review of models with the following aims in mind: 1. To confirm that each model implements the documented methods; 2. To identify potential improvements in each model's implementation; 3. To share expertise about good coding practices; and 4. To increase exposure to, and familiarity with, code review as a means to support computational research. We will conduct pre- and post- evaluation surveys to evaluate how well we address each aim, and to provide evidence for iterative improvements for future rounds of code review. We will present this initiative as a component of our community of practice, and hope to initiate a discussion with participants about its merits and dissemination of the model. wombat-2025-33-piloting-peer-code-review-in-a-research-consortium-community-of-practice Saras Windecker en false https://conf.nectric.com.au/wombat-2025/talk/YTKMW3/ 8.03 Talk 2025-09-30T15:30:00+10:00 15:30 00:30 Every new data team faces the same dilemma: stakeholders expect immediate results while the team knows that achieving long-term impact needs investment in its capabilities. Should they build that reporting dashboard everyone's asking for, or spend three months building a new dataset that could inform future priorities? If they choose the quick win, the team risks being pigeonholed into a reporting team, unable to focus on longer-term projects, with more and more regular products requiring maintenance. Focus on building capacity, and the team might lose organisational support before they can prove their value, limiting their impact before they could even get going. Drawing on experience from working in the Central Analytics Hub at the Department of Prime Minister and Cabinet during the pandemic and working to help address capability gaps to better inform migration policy reform, this talk will explore the practical realities of navigating these tensions. I'll share stories from both experiences—from delivering urgent daily briefings while building the necessary capability to pivot to changing priorities, to establishing credibility while building the infrastructure needed to inform pressing policy priorities. This talk will offer honest reflections on the messy reality of building data teams in government, providing practical insights for anyone wrestling with similar challenges. wombat-2025-34-building-a-new-data-team-balancing-quick-wins-and-long-term-investments Tyler Reysenbach en false https://conf.nectric.com.au/wombat-2025/talk/ZHFDRZ/ 8.03 Talk 2025-09-30T16:00:00+10:00 16:00 00:30 The Mayi Kuwayu Study is a longitudinal survey and the largest national study of Aboriginal and Torres Strait Islander culture, health and wellbeing. I joined the study team last year and have been redeveloping the data pipeline to be more transparent, reproducible, and easier to maintain. My original plan was to incorporate best practices, such as using `targets` and `renv` for ensuring reproducibility and `pointblank` and `testthat` for validation and testing, as well as writing nice code. It was a good plan, but it didn't fully consider the people in the study team. Being able to run code from start to finish without error might be reproducible in a technical sense, but is someone actually reproducing it if they don't understand what the code is doing? Code is only reproducible so long as it's maintainable. Since I'm the only experienced R-user (for now), each additional package I use is a package someone else has to understand and runs the risk of making the code less reproducible. In this session I'll be talking about: - the Mayi Kuwayu Study and team context - development of the data pipeline - the role people played in design decisions - using the development as an opportunity to upskill my team wombat-2025-41-designing-data-infrastructure-where-people-come-first Ben Harrap en false https://conf.nectric.com.au/wombat-2025/talk/W3K3VM/ 8.03 Talk 2025-09-30T16:30:00+10:00 16:30 00:30 A summary of the day, with some information on other events and organisations in the local data analysis space. wombat-2025-35-closing-discussion Cynthia Huang en false https://conf.nectric.com.au/wombat-2025/talk/B8MJBL/