Learning objectives

On successful completion of this module, participants will achieve:

- Confidence with ggplot2 syntax: Gain proficiency in utilizing the ggplot2 syntax, mastering the creation of a wide array of visualizations for effective data representation.
- Tailoring plots to data types: Develop the skill to select and customize appropriate plot types based on the nature of the data, ensuring clarity and relevance in visual communication.
- Best practices in data visualization: Acquire a deep understanding of best practices in data visualization, encompassing principles such as color theory, label clarity, and the avoidance of chartjunk, to produce compelling and accurate visualizations.
- Understanding data curation: Explore the critical concept of data curation, learning how to assess, clean, and manage datasets for optimal analysis and visualization outcomes.
- Data preparation for tasks: Learn the essential techniques for preparing data in the correct format for specific tasks, ensuring that datasets are appropriately structured and refined to meet the requirements of analysis and visualization.
- Extraction of insights from large data: Develop the ability to navigate and extract meaningful information, including means, tables, and other relevant insights, from large datasets, enabling informed decision-making and interpretation of complex data structures.
- Other, TBD

Program

- ggplot2: Understanding the fundamentals of the ggplot2 package. Exploring the grammar of graphics for creating versatile and customized visualizations. Diving into intricate features and advanced customization options within ggplot2. Hands-on exercises to solidify proficiency in constructing complex plots.
- dplyr: Comprehensive coverage of the dplyr package for efficient data manipulation. Mastering essential verbs such as filter, mutate, select, arrange, and summarise. Understanding the power of the pipe operator (%>%) for creating readable and concise data manipulation workflows. Applying pipelining to enhance code readability and reproducibility.
- ggeffects: Exploring the capabilities of the ggeffects package for visualizing marginal effects from statistical models. Understanding how ggeffects enhances the interpretability of complex model results. Utilizing ggeffects for visualizing interaction effects and other nuanced model outcomes.
- Other, TBD

Readings

Some previous knowledge of quantitative methods is required. Because of the limited scope of the course, it’s impossible to tackle more introductory topics from the very start. Participants are expected to know basic concepts in linear regression (e.g. mixed vs non-mixed models, dependent vs independent variables, interactions, intercepts and slopes etc.). The book “Statistics for linguists” from Bodo Winter, 2019, is recommended to brush up on these topics ahead of time.

Didactic methods

Univr, aula e orario TBD
Referente: Chiara Melloni (chiara.melloni@univr.it)

Scheduled Lessons