B8267 - INTERPRETARE E COMUNICARE I DATI

Learning outcomes

At the end of the course, the student - is able to effectively present the results of statistical analyses using graphical and tabular visualisation tools and data storytelling; - has acquired the ability to move with awareness among multiple data-based information presented by the media; - possesses the skills necessary to recognise critical issues in the way such information is sometimes reported.

Course contents

The course builds on the statistical knowledge and skills already acquired by students and applies them to the communication of results. The activities combine short methodological introductions, the analysis of real-world cases, and the preparation of communication materials.

Part I – Critically interpreting data and quantitative information (4 hours)

From the research question to the message: distinguishing between data, results, interpretations, and claims.

Quality and suitability of sources: primary and secondary data, institutional sources, metadata, operational definitions, and comparability.

Critical interpretation of indicators and comparisons: absolute and relative values, denominators, percentage changes, averages, and distributions.

Uncertainty, variability, and the limits of inference: correlation and causation, data selection, missing data, and unreported results.

Analysis of articles, press releases, reports, and social media content in order to trace the original source and assess the validity of the conclusions.

Part II – Visualising data and statistical results (6 hours)

Principles of data visualisation: selecting the most appropriate representation according to the type of variable, the research question, and the communication objective.

Errors and misleading practices in data visualisation; critical analysis and redesign of graphs and tables drawn from real-world cases.

Part III – Developing and presenting a data story (5 hours)

Defining the target audience, the communication objective, and the key message.

From exploratory analysis to communication: selecting the most relevant evidence without compromising rigour or context.

Narrative structure, logical sequence of arguments, titles, explanatory text, and conclusions.

Differences between scientific, institutional, and public-facing communication; adapting the register and level of detail to the intended audience.

Readings/Bibliography

Beyond the teaching material provided by the lecturer (and available on Virtuale) selected references for further reading are:

• Alberto Cairo, Come i grafici mentono. Capire meglio le informazioni visive, Raffaello Cortina Editore, 2020
• Donata Columbro, Ti spiego il dato, Quinto Quarto Edizioni, 2021.
• Donata Columbro, Quando i dati discriminano. Bias e pregiudizi in grafici, statistiche e algoritmi, Trento, Il Margine, 2024.
• Andrea De Mauro, Data Analytics per tutti. Imparare ad analizzare, visualizzare e raccontare i dati. Milano: Apogeo, 2022.
• David J. Hand, Il tradimento dei numeri, I Dark Data e l’arte di nascondere la verità, Bur, Rizzoli, 2020
• Tim Harford, Dare I numeri, Breschi, 2021
• Hans Rosling, Ola Rosling, Anna Rosling Rönnlung Factfulness. Dieci ragioni per cui non capiamo il mondo. E perché le cose vanno meglio di come pensiamo, Rizzoli, 2018
• David Spiegelhalter, L’arte della statistica. Cosa ci insegnano i dati, Piccola Biblioteca, Einaudi, 2020

Teaching methods

The course is delivered in person and adopts a principally active-learning approach. Short introductory lectures are combined with guided discussion of case studies, individual and group exercises, the critical evaluation and redesign of data visualisations, the development of short data-driven narratives, and peer review.The case studies will be drawn from institutional reports, news articles, websites, and social media. Activities may be carried out using R/RStudio and other data analysis and visualisation tools with which students are already familiar.

Assessment methods

The work must include:

• a brief description of the research question, the target audience, and the communication objective;
• an indication of the data source, the main characteristics of the data, and any relevant limitations;
• a reasoned selection of graphs, tables, or indicators;
• a coherent data story;
• a discussion of the communication choices made and of any potential interpretative issues.

The assignment, which should normally consist of no more than 8 slides, must be submitted at least three days before the examination date and will be presented and discussed during the scheduled session.

Assessment will take into account the accuracy of the statistical interpretation, the appropriateness of the visualisations, the coherence of the narrative, transparency regarding data sources and limitations, and the overall effectiveness of the presentation.

The final result will be expressed as pass/fail.

As for learning assessment, limited, declared, and non-substantial use of AI is allowed for support activities (summaries, rewording). Substantial use for completing parts of the test is not allowed.

Teaching tools

Slides, datasets and statistical reports, institutional data websites, articles and other informational content, R/RStudio and packages for data visualisation and reporting, spreadsheets, and other open-source or freely available tools.

Office hours

See the website of Stefania Mignani