Data analysis and statistics

A.Y. 2019/2020
Overall hours
Learning objectives
The increasing availability of data allows to analyse several issues related to social, labour and organizational dynamics. The course aims to introduce the logic of theory-driven statistical analysis in social research, providing useful tools for data manipulation, exploratory data analysis, statistical test, categorical data analysis, and regression.
The course consists of both theoretical lessons, where the main statistical techniques are introduced, and empirical lessons, where techniques are empirically implemented (learning by doing). Lessons are coupled with 20 hours of lab, aiming to consolidate the topics of the classes and to work on the students' final research.
Expected learning outcomes
At the end of the course, students will be able to conduct multivariate analyses and to apply the basic principles of statistical inference. Moreover, students will be able to use the statistical software STATA and to conduct autonomously a research project, with the aim of informing conclusion and supporting decision-making. Finally, students will be able to interpret scientific contributions based on statistical multivariate analysis, considering both the potentials and the limitations of data analysis.
In detail, students will:
- collect data and enter their own datasets for analysis;
- identify appropriate statistical methods;
- conduct their own analysis using the software program STATA;
- interpret the results of the data analysis;
- check the assumptions on which each analysis depends and make appropriate adjustments or select alternative methods of analysis.
Course syllabus and organization

Single session

Lesson period
First trimester
Course syllabus
Data sources for social research
Data matrix, cases and variables.
Distributions and numerical summaries.
Data manipulation.
Scaling and standardizing.
Measuring inequality.
Smoothing time series.
Sampling and statistical inference.
Contingency tables.
Chi-square statistic.
Scatterplots and linear relationships.
Linear regression model.
Causal explanation and multivariate analysis.
Multiple regression and logistic regression models.
Longitudinal data and approaches to event history modelling.
Prerequisites for admission
Students must be numerically and computer literate. Students should be familiar with basic statistical concepts.
Teaching methods
Frontal lessons and labs
Teaching Resources
- Finlay, B., & Agresti, A.. Statistical methods for the social sciences.
- Lecture notes (available on ARIEL website during the course).
Assessment methods and Criteria
Written test at the end of the course and optional research report. The report can be developed by small-groups (2/3 students).
SECS-S/04 - DEMOGRAPHY - University credits: 6
Lessons: 40 hours
Educational website(s)
Monday, 14.30-17.30
Room 1