BSTA 17A00 - QUANTITATIVE METHODS FOR HUMANITIES AND SOCIAL SCIENCES

The first-year course in quantitative methods for the humanities and social sciences serves as a foundational introduction to their uses. As a crucial intersection of interdisciplinary study within the social sciences, quantitative methods are utilized across most core disciplines taught at Sciences Po in the first year, including sociology, economics, and political science. While their application may be less prevalent in historical and legal scholarship, statistics nonetheless contribute to the evolution of certain research areas within those fields. The primary objective of this course is to familiarize students with the interpretation of numerical results and graphical representations, data analysis, and the practical application of statistical methods in social sciences. It is designed to cultivate students' competencies in accurately reading tables and graphs and in interpreting them by posing pertinent questions regarding their construction. In addition to their significance in scientific research, statistics function as an essential tool for practical application: they facilitate the management and governance of institutions and enterprises. Through a critical lens, we aim to equip students with a robust statistical literacy that is accessible to all. Statistics are not an exclusive domain; one does not require an extensive mathematical background to compare means or percentages, draw valid conclusions, and leverage this information for argumentation or decision-making. EDUCATIONAL OBJECTIVES: - Acquire knowledge of quantitative methods in their interdisciplinary dimension - Discover the epistemological foundations and construction methods to build critical knowledge - Read statistics and manipulate them in their concrete applications (using Google Sheets)

Chung Shue CHEN,Yuma ANDO,Pap Souleymane N'DOYE,Jules-Rémy SARANT,Myriam MAUMY,Aïssa ALLAOUI,Abderrahman EL SAMALOUTY,Clara LE GALLIC-ACH,Camille VOISIN

Séminaire

English

Spring 2024-2025

ORGANIZATION OF SESSIONS: Session 1: Quantitative Methods as a Tool for Argumentation Quantitative methods extend beyond mathematics; they engage with social, political, and other significant issues. In this context, statistics serve an argumentative purpose by providing one of several means of evidence. This session explores the implications of quantification—“putting numbers” to data as a result of a process of data construction. Concepts covered: quantification, epistemology and history of statistics, argumentation Session 2: Describing a Variable This session introduces students to the spreadsheet tool utilized in the course (Google Sheets) and provides an overview of key univariate statistics: measures of central tendency (which summarize a variable) and measures of dispersion (which indicate the variable's distribution). It also discusses the implications of representing a variable's distribution: what is the difference between presenting it as a table versus a graph? Concepts covered: mean, median, standard deviation, variance, quantiles Sessions 3 & 4: Relating Variables Establishing a relationship between two variables varies according to their nature. The appropriate methods and tools for studying two qualitative variables (such as gender or age category) differ from those for two quantitative variables (such as height or weight), or for one qualitative and one quantitative variable. These two sessions also introduce methods for examining cause-and-effect relationships between two variables by comparing a test group with a control group. Concepts covered: cross tables, comparison of means, covariance, correlation. Session 5: Where Do Statistical Data Come From? This session aims to provide an overview of the types of data used in statistics, along with their respective advantages and disadvantages: survey data, administrative data, and data constructed from traces. The goal is to equip students with the tools necessary to identify the source of a table or graph. It also revisits questions related to sampling or the population from which statistics are derived. Concepts covered: types of data, population, (random or non-random) sampling, margin of error Sessions 6 & 7: Constructing and Categorizing Data During these two sessions, students will explore fundamental principles of data construction. The sessions will first raise awareness about the challenges involved in designing a questionnaire: the distinction between open and closed questions, the impact of question order and wording, and the role of GDPR in information collection. They will then introduce the challenges associated with recoding data, which, far from being a mere technical operation, involves scientific and often political choices—especially when constructing a database from unstructured written or digital traces. Concepts covered: questionnaire, categorization, statistical class, recoding Sessions 8 & 9: Testing Significance To ensure that the relationship between two variables is not merely a chance occurrence, it must be subjected to a significance test by comparing it against the appropriate statistical distribution. These two sessions introduce students to this approach and its implementation. One session will focus on Student's t-test (for differences in means), while the other will address the Chi-squared test (for differences in proportions). Concepts covered: statistical test, standard error, confidence interval Session 10: Introduction to Linear Regression The aim of this session is to familiarize students with regression techniques and related concepts, enabling them to interpret regression results (simple, multiple, or logistic). A simple linear regression will be conducted in class. Concepts covered: scatter plot, regression line, R², p coefficients (p-value) Session 11: Introduction to Graphic Semiotics The final session addresses data visualization challenges, particularly distinguishing between the treatment of spatial data (which pertains to cartography) and non-spatial data: timelines, categories, diagrams, matrices, and results from statistical treatments. Concepts covered: cartography, graphics, data visualization (dataviz), scales Session 12: Final Exam