84613 - Social Statistics

Learning outcomes

At the end of the course the student has skills on complex sampling plans. In particular, the student is able to: - design complex sample surveys - analyze and synthesize the information obtained in an advanced statistical perspective - identify and use the appropriate estimators for the inferential problem to be faced - evaluate and communicate the degree of uncertainty of the estimates obtained .

Course contents

• Inference in sampling from infinite and finite populations.
• Inference based on the sampling design. Selection and inclusion probabilities, the Horvitz–Thompson estimator.
• Simple random sampling with and without replacement.
• The design effect.
  
• Stratified sampling.
• Determination of sample size.
• Cluster sampling.
• The use of auxiliary variables: ratio estimators, regression estimators, calibration estimators.
• Two-stage sampling.
• Inference from non-probability samples (overview).
• The survey instrument: the questionnaire. Structure, types, and wording of questions. Assessment of questionnaire validity: tools for reliability analysis.
• Models for the analysis of questionnaire data.
• The theoretical introduction to the fundamental concepts will be complemented by practical sessions devoted to exercises and the analysis of case studies using the R software.

Readings/Bibliography

Additional bibliographic references will be provided throughout the course.

Teaching methods

Lectures covering the theoretical foundations, complemented by laboratory sessions using the R software environment, devoted to practical exercises and the analysis of case studies.

Assessment methods

The examination is designed to assess the achievement of the following learning outcomes:

• a thorough understanding of the theoretical concepts covered during the lectures;
• the ability to design and plan sample surveys;
• the ability to apply the above concepts using the R statistical software.

Assessment consists of a compulsory written examination and an optional oral examination. The written examination lasts two hours and comprises both theoretical and applied exercises to be completed using the R statistical software.

During the written examination, students may consult a set of formula sheets prepared by themselves. These formula sheets must not exceed four A4 pages. No other materials may be consulted. The formula sheets must be submitted together with the examination paper at the end of the test.

The final grade will be awarded according to the following scale:

• Below 18: Fail
• 18–23: Satisfactory
• 24–26: Fair
• 27–28: Good
• 29–30: Very Good
• 30 cum laude: Excellent

With regard to assessment, the use of artificial intelligence (AI) is strictly prohibited. Any use of AI will be considered a breach of academic integrity.

Teaching tools

Slides, data sets and reports on statistical surveys

Office hours

See the website of Carlo Trivisano