Universität Wien

400023 SE Content Analysis in the Social Sciences: Human and Automated Approaches (2016W)

SE Methods for Doctoral Candidates

Continuous assessment of course work

Registration/Deregistration

Note: The time of your registration within the registration period has no effect on the allocation of places (no first come, first served).

Details

max. 21 participants
Language: English

Lecturers

Classes

MO 12.12.2016 09.00-11.00 Seminarraum 12, Währinger Straße 29 2.OG
DI 13.12.2016 9:45-13:00 Seminarraum, Rathausstraße 19/9, 1010 Wien
DI 13.12.2016 14.00-16.30 PC-Unterrichtsraum 1, Währinger Straße 29 1.UG
MI 14.12.2016 09.45-13.00 Seminarraum 8, Währinger Straße 29 1.OG
MI 14.12.2016 14.00-16.30 PC-Unterrichtsraum 6, Währinger Straße 29 2.OG
DO 15.12.2016 09.00-12.30 Seminarraum 9, Währinger Straße 29 2.OG
DO 15.12.2016 14.00-16.30 PC-Unterrichtsraum 1, Währinger Straße 29 1.UG


Information

Aims, contents and method of the course

Instructor: Prof. Stuart Soroka, Michael W. Traugott Collegiate Professor of Communication Studies and Political Science, University of Michigan, ssoroka@umich.edu.

Outline: The increasing ready availability of digital texts has led to an explosion of interest in content analytic methods. This has been true for those interested in the content of news programming, in traditional newspapers, television transcripts, and online news. It is true for entertainment media researchers, now able to analyze years
of scripts from movies and film, and from video games. Social media analysts have ready access to miles of data, particularly from Twitter. And that same content has been of interest to scholars interested in predicting election outcomes, macro-economic trends, and stock prices.

Content analysis has been around for a long time, however; and many of the core ideas developed 40 years ago are equally important today. The objective of this short course is to offer a theoretically-informed introduction to large-scale content-analytic methods for the social sciences (albeit with an emphasis on political communication). We begin by reviewing past work on human-coded methods, and then proceed to current automated approaches, both dictionary based and machine learning.

Classes combine lectures, discussion, and computer-based lab work. It is expected that students will have completed readings before class, and will have both R, RStudio, and Lexicoder installed on their computers. Please bring your own computers with you to the course. This software is free, and multi-platform; we will
be using it for content analysis of texts. If students are interested in working with particular corpuses, they are welcome to bring those to class; otherwise, example corpuses will be provided. By the end of the class, students will be familiar with a range of content analytic approaches, and will be able to conduct automated analyses of
large-scale corpuses in R.

Assessment and permitted materials

Minimum requirements and assessment criteria

Examination topics

Reading list


Association in the course directory

Last modified: Mo 07.09.2020 15:47