|
Lehrende: M.Sc. David Broska
Veranstaltungsart:
Seminar / Übung
Orga-Einheit: Politics, Administration & International Relations
Anzeige im Stundenplan:
Computational Politi
Semesterwochenstunden:
3
Credits:
6,0
Standort:
Campus der Zeppelin Universität
Unterrichtssprache:
Englisch
Min. | Max. Teilnehmerzahl:
4 | 35
Prioritätsschema: Standard-Priorisierung
Inhalte:
Digital technologies and the relentless shift towards digital modes of communication bring fundamental changes to societies. This transformation challenges the integrity of privacy and information on the web, stipulates regulatory oversight, and necessitates decision-makers to find ways in which technological change can broadly benefit society. For social scientists, these and related topics create fascinating lines of inquiry.
However, the digital transformation does not only create new research topics. The increase in computing power, online communication, and the abundance of data also expands the ways in which we study social phenomena. This course provides an overview of useful quantitative methods to collect and analyze data from online resources.
For example, participants gain hands-on experience with computer-assisted data collection (web scraping) and the use of databases with SQL. In line with the burgeoning interest of political scientists in natural language processing, we also cover methods to analyze large text corpora. These techniques include, but are not limited to, extracting features of texts such as content categories, word counts, dictionary counts, or parts of speech. Statistical methods are used to draw inferences about the texts or their authors. Political scientists have used those methods to track public opinion on social media, reveal censorship through government agencies, or infer political ideology from speeches and party manifestos.
This course does not solely focus on the application of methods. We will also discuss the advantages and caveats of the above techniques to build an intuition for assessing their respective strengths, weaknesses, and trade-offs. This will be done by studying applications of the above-mentioned methods in published academic work exploring various sociopolitical phenomena.
There are no strict prerequisites for this course but basic R programming skills - or the eagerness to learn - are expected. For applications in quantitative text analysis, we will use the quanteda package for R.
Lernziele:
The overall goal of this course is to let participants enrich their methodological repertoire and strengthen their skills in conducting computational social science research. Participants will therefore be responsible for carrying out their own small-scale research project on a topic of their interest using some of the methods presented in this course. Three coding assignments are intended to lay the foundation for the final project: participants practice the creation and analysis of textual data using content analytic and statistical software.
Weitere Informationen zu den Prüfungsleistungen:
Project (70%)
Assignments (30%)
Literatur:
Alvarez, R. M. (2016). Computational Social Science: Discovery and Prediction. Cambridge University Press.
Blätte, A., Behnke, J., Schnapp, K.-U., & Wagemann, C. (2018). Computational Social Science: Die Analyse von Big Data. Nomos Verlag.
Grimmer, J., & Stewart, B. M. (2013). Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Texts. Political Analysis, 21(3), 267–297.
Wenn Sie E-Learning Funktionalitäten nutzen möchten, tragen Sie bitte "Ja" ein.:
Ja
|
