This course explores the consequences of the digital traceability of collective phenomena with a critical and empirical approach. Considering a variety of computational methods, it offers first-hand experience of digital quantification, examining its potential, but also its shortcomings and biases. Students will learn techniques of data collection, corpus cleaning, exploratory analysis, network analysis, natural language processing, artificial intelligence, and information visualization. Working in groups, they will apply these techniques to actual societal situations. Through this experience, they will be led to consider reflexively the insights of the transdisciplinary field of critical data studies.
Course Structure\ Pedagogical Approach\ Learning Outcomes\ Class Programme\ Evaluation
The course is composed of three main parts:
- Critical theories of quantification, digital and online media.
- Digital, computational and qualitative-quantitative methods.
- Case study group work.
The version of the course presented in this page grants equal space to the three parts, but each can be expanded according to the curriculum in which the course is integrated.
-
The theoretical part can be developed considering the impacts of media and digital infrastructures to encourage critical reflection on subjects related to Gender Studies and Political and Cultural Geography, discussing the social and cultural consequences of digital media, particularly with regard to constructions of gender, class, ethnicity, etc.; Political Economy, Security Studies and Political Sciences, by extending the reflections on the effects of quantification in the government of modern societies at the national and international levels.
-
The methodological part can be developed for curricula in Digital Humanities, Journalism, Media Studies and Social Research and Intervention, by introducing more advanced numerical and computational methods and discussing the advantages and disadvantages compared to traditional qualitative and quantitative methods, as well as the possible integration with them.
-
The practical part can be developed and focussed on case studies around subjects directly relevant for programs in Public Management, Sustainable Development, Science and Technology Studies, Innovation Management, etc.
While data literacy is increasingly considered as an essential skill in a variety of curricula, its teaching is often unfit to prepare students for their future occupations. Trained on predefined exercises and artificial datasets, students are rarely exposed to the messiness of real data practice. They assimilate techniques of digital analysis and visualization, but fail to learn the subtler craft of thinking about and with data. Focussing too much on data crunching and not enough on the conditions of the production of data and on the consequences of their use, this type of training discourages reflexivity and encourages a naively technical approach to data literacy. Students are rarely brought to reflect on the work necessary to distill datasets from social and natural phenomena and even less on the (side)effects of managing those phenomena through their data-doubles. To promote a richer and more realistic approach to "data literacy", this course draws on the literature from disciplines such as science and technology studies, media studies, political sciences, security studies etc. Data Inquiries, however, does not deliver these insights through a conceptual instruction, but rather through group-work on real projects developed by civil organizations working with data.
- Explore a variety of different digital techniques for collecting, cleaning, analyzing and visualizing data. Be able to compare the strengths and weaknesses of the different conceptual, mathematical and technical tools and to choose the ones that suit best the resources and objectives of one's project.
- Develop the ability to adapt to the constant evolution of computational technologies. Know how to update one's repertoire of technical and conceptual tools and learn how to find new software, scripts, libraries and how to adapt them to one's needs.
- Become familiar with all the stages of a data inquiry and learn to manage such a project from start to finish, by effectively linking different tools and dealing with the accumulation of transformations and possible distortions that this implies.
- Develop a critical reflection on the impacts of the increasing quantification of social life promoted by digital technologies. Learn to appreciate its potential for scientific research, public debate and the management of communities and businesses, but also to recognize its possible biases and its economic and political consequences.
- Know how to work in groups and in collaboration with societal actors. Know how to organize a data project as a collective and participatory effort. Work in cooperation with other students, but also with stakeholders inside and outside the academy.
- In preparation for the class, read this syllabus carefully and prepare at least one question about the course.
- During the class, we will present the objectives of the course, its pedagogical approach and its expectations in terms of effort during and between the classes. We will also introduce the potentials, but also the risks of the social use of data and computational methods.
-
In preparation for the class, read and prepare to discuss one of the following texts:
- Desrosières A (1990) How to Make Things Which Hold Together : Social Science, Statistics and the State in Discourses On Society. Dordrecht: Kluwer Academic Publishers.
- Bowker GC and Star SL (1999). “The Case of Race Classification and Reclassification under Apartheid” in Sorting Things Out: Classification and Its Consequences. Cambridge Mass: MIT Press.
- Porter TM (1995). “How social numbers are made valid” in Trust in Numbers. The Pursuit of Objectivity in Science and Public Life. Princeton: University Press.
-
During the class, through a discussion of the above texts, we will discuss the effects of categorization and quantification by showing that, far from being neutral and descriptive, these operations transform and perforate their objects.
-
In preparation for the class, read and prepare to discuss one of the following texts:
- D’Ignazio C and Klein LF (2020) “What Gets Counted Counts” in Data Feminism. Cambridge Mass .: MIT Press.
- Boullier D (2017) Big data challenges for the social sciences: from society and opinion to replications. ISA eSymposium for Sociology July.
-
During the class, discussing the above texts, we will examine the different ways in which the “datafication” produced by digital technologies creates and structures society and maintains certain asymmetries and power relations.
-
In preparation for the class, read and prepare to discuss one of the following texts:
- Gerlitz C and Helmond A (2013) The like economy: Social buttons and the data-intensive web. New Media & Society 15 (8).
- Venturini T (2019) “From Fake to Junk News, the Data Politics of Online Virality”. In: Bigo D, Isin E, and Ruppert E (eds) Data Politics: Worlds, Subjects, Rights. London: Routledge, pp. 123–144.
- Bach D, Tsapatsaris MR, Szpirt M, et al. (2020) The Baker’s Guild: The Secret Order Countering 4chan’s Affordances. oilab.eu/the-bakers-guild-the-secret-order-countering-4chans-affordances/.
-
During the class, discussing the above texts, we will consider how public debate is transformed by the mediation of social media platforms.
-
After the class, divide into groups of 4 students each.
-
In preparation for the class, read and prepare to discuss one of the following texts:
- Gitelman L (2013). “Dataveillance and Countervailance” in “Raw Data” Is an Oxymoron. Cambridge Mass .: MIT Press.
- Gabrys J, Pritchard H and Barratt B (2016). Just good enough data: Figuring data citizenships through air pollution sensing and data stories. Big Data & Society 3 (2).
- Gray J, Lämmerhirt D and Bounegru L (2016). Changing what counts. DataShift: 52.
-
During the class, discussing the above texts, we will consider the forms of resistance emerging in our societies against the dominant quantification mechanisms and we will reflect on how to draw inspiration from these efforts in our data inquiries.
-
After the class, choose an inquiry idea among those offered in the Data Inquiries clearing/green-house.
-
In preparation for the class, read:
- Rogers R (2017). “Foundations of Digital Methods: Query Design” in Schaefer M and van Es K (eds) The Datafied Society. Amsterdam: University Press, pp. 75–94.
-
During the class, we will explore different techniques for collecting data on the Web. We will consider the importance of avoiding always turning to the same sources (e.g. Twitter or Wikipedia) and of seeking data specifically relevant to one's investigation. We will learn the subtle art of querying a search engine, as well as the basics of scraping and crawling.
-
After the class, find a source of data suitable for your investigation and extract a set of records.
-
In preparation for the class, read:
- Dumit J and Nafus D (2018). “The other ninety per cent: Thinking with data science, creating data studies” in Ethnography for a Data-Saturated World, pp. 252–274.
-
During the class, we will go over the somewhat boring, but crucial operation necessary to transform a set of digital records into a corpus of exploitable data. We will consider how the ability to separate information from "noise" is crucial to the success of any data investigation, but also how this separation is never obvious, natural or simple. We will practice tools and techniques for data wrangling and descriptive statistics.
-
After the class, clean up all the records extracted following the previous class and transform them into a corpus ready for analysis.
-
In preparation for the class, read:
- Behrens JT and Chong-Ho Y (2003) Exploratory Data Analysis. In: Weiner IB (ed.) Handbook of Psychology. London: Wiley.
-
During the class, we will discuss the differences between confirmatory analysis and exploratory analysis and we will introduce the main mathematical and visual techniques for the latter.
-
After the class, use exploratory data analysis to improve the cleaning of your dataset and to produce a first report on the distribution of its values.
-
In preparation for the class, read
- Tufte ER (1983) “Data-Ink Maximization and Graphical Design” in The Visual Display of Quantitative Information. Cheshire: Graphics Press.
- Drucker J (2011) Humanities Approaches to Graphical Display. DHQ: Digital Humanities Quarterly, 5 (1): 1–20.
-
During the class, drawing on the visual semiotics of Jacques Bertin, we will discuss the characteristics of different visual variables and how to use them to visualize different types of numerical variables. We'll also discuss the importance of finding ways to preserve data ambiguity.
-
After the class, use one of the software or websites presented in the class to produce at least three visualizations of your dataset. For each, write a detailed caption explaining how to read/interpret it.
-
In preparation for the class, read:
- Venturini T, Jacomy M and Jensen P (2020) What do we See when We Look at Networks. An Introduction to Visual Network Analysis and Force-Directed Layouts. Big Data & Society (forthcoming).
-
During the class, we will learn to analyze a relational dataset. We will introduce the basics of social network analysis and complex network analysis and the metrics of graph theory. We will also experiment with Table2net (medialab.github.io/table2net) and Gephi (gephi.org)
-
After the class, use Table2net to extract one or more networks from your dataset. Use Gephi to generate one or more images of your network. For each, write a detailed caption explaining how to read/interpret it.
-
In preparation for the class, read:
- Costa C (2020) “A Tour of Machine Learning Algorithms Read this introductory list of machine learning algorithms” in Toward Data Science towardsdatascience.com/a-tour-of-machine-learning-algorithms-466b8bf75c0a
- Cielen D, Meysman ADB and Ali M (2016) “Machine Learning” in Introducing Data Science: pp. 57–84.
-
During the class, we will introduce the basics of machine learning and we will test some of the most popular algorithms. We will also consider the advantages and disadvantages of this approach and we will ask ourselves in which technical and social contexts it can be useful or problematic.
-
After the class, prepare a detailed report on the progress of your group project. Identify the blockages and define the backcasting timeline for the finalization of your inquiry. [OPTIONAL: apply one machine learning technique to the data of your project]
- During the class, the groups encountering technical or conceptual difficulties in their data inquiry can ask for help to overcome the obstacles that hinder their progress.
-
In preparation for the class:
- Read: Venturini T, Bounegru L, Gray J, et al. (2018) A reality check (list) for digital methods. New Media & Society, 20 (11).
-
During the class, we will discuss together the results obtained by each group and test their soundness and clarity.
-
After the class, write a text (between 2,500 and 5,000 words) presenting the results of your investigation.
-
In preparation for the class, read
- Segel E and Heer J (2010) Narrative visualization: telling stories with data. IEEE Transactions on Visualization and Computer Graphics, 16 (6).
-
During the class, we will work on the metrics, figures and texts produced so far and we will try to connect them into a clear and convincing story.
-
After the class, publish a “one-page website” on medium.com with the results of your inquiry.
-
In preparation for the class, read:
- Bruno I, Didier E and Vitale T (2014) Statactivism: forms of action between disclosure and affirmation. The Open Journal of Sociopolitical Studies, 2 (7): 198–220.
-
During the class, we will discuss how the results of the inquiries of different groups can be turned into public interventions, in collaboration with the social actors they concern.
-
After the class, finalize your inquiry and identify the means to make it public and actionable.
####Presentation of the results of group projects In this class, the students will present the results of their investigation and their strategy to make them public and actionable in society. Civil society actors concerned by the projects will be invited to participate in discussion.
The course grade is the average of the three grades received by students in each part of the course:
- The evaluation for the first part of the course is an individual grade. Students are evaluated on their understanding of the texts (60%) and their engagement in the discussion (40%).
- The evaluation for the second part is a collective grade on the various statistical and visual analyses produced by each group. Groups are evaluated for their punctuality in handing in the various exercises (40%) and by the quality and richness of their results (60%).
- The evaluation for the third part of the course is a collective grade on the data inquiry carried out by each group. Groups are evaluated on the overall quality of their results (33%), the ability to present them in a coherent narrative (33%) and on their ability to make them public and socially impactful (33%).