98804 - Analysis of Public Opinion Laboratory (LM)

Learning outcomes

The laboratory introduces to the method of empirical research in the field of public opinion analysis, basic statistical techniques, and the use of survey data for the analysis of opinions on issues of public interest, political attitudes, and voting behavior. At the end of the laboratory, the student knows: - how to design and conduct opinion surveys; - how the results of such surveys can be interpreted through statistical analysis to study the factors underlying political orientations and voting choices.

Course contents

The aim of the laboratory is to introduce participants to the use of software and basic statistical techniques for conducting original analyses of survey data and textual data taken from social media in order to interpret "public opinion orientations" and test hypotheses regarding the factors influencing underlying individual attitudes.

To this end, students will be introduced to some advanced functions of Microsoft Excel and to the use of the statistical software STATA (freely available for Unibo students). The exercises will cover the implementation of monovariate (simple tables), bivariate (tables of two cross variables), and multivariate (linear and logistic regression models) analyses, text classification (Linear Discriminant Analysis) and interpretation of results.

The objective will be to carry out empirical analyses 'on their own', by directly examining the data, regarding the topics covered in the first module: the way in which individuals position themselves along the fractures and divisive issues salient today (statism/liberalism, globalism/nationalism, climate change, immigration, civil rights); the factors influencing the formation of these opinions and voting choices (ascribed characteristics, position in the social structure, long-term political predispositions, evaluation of incumbent governments and leaders).

During the final laboratory lectures, participants are expected to present their own elaborations on one of these topics.

Readings/Bibliography

The course is based on the lecturer's lectures and tutorials. Attendance is compulsory. Targeted reading recommendations will be given depending on the prior skills of each student.

Teaching methods

The course consists of fifteen working sessions in which students will be invited to actively participate. The first three sessions will consist of classic lectures recapitulating the essential contents of research methodology and public opinion analysis. The bulk of the sessions will consist of exercises in which participants will be introduced to the use of the main statistical techniques, using STATA software, on comparative survey data (European Social Survey, World Value Survey). Participants will also be required to take turns in presenting their work in preparation for the final paper. All lectures will be held in presence. It is necessary for each participant to have their own personal computer connected to the Internet on which an up-to-date version of Stata must be installed (instructions on how to install Stata will be given during the lessons).

Attendance is compulsory. No more than three sessions may be missed. To facilitate individual study and practice with statistical techniques, video recordings of some lectures will be made available to participants.

Assessment methods

Each student will be required to prepare and present in class an analysis based on original processing of data from one of the surveys provided by the lecturer and presented in class, or a statistical analysis of textual data taken from social networks. The final mark will be based on the assessment of the presentation as well as an oral interview covering the course content as usual.

Teaching tools

Projector, PC, Unibo licence for STATA statistical software, classroom equipped with power sockets and wireless internet connection, cloud sharing of course materials (presentations, datasets, syntax files).

Office hours

See the website of Salvatore Vassallo