Suchen nach Veranstaltungen - Einzelansicht

The seminar is about methods to analyse open-ended questions in the field of Political Science, especially election studies and research on immigrant-related attitudes. This way of data collection and analysis is an excellent possibility to conduct your own analysis for your empirical theses! The seminar consists of two major parts. First, we read and discuss central works on the topic together in synchronous sessions. We identify common problems, see which research questions can be answered with this sort of data collection and gather information about possible methods. Second, we use the methods ourselves to code open-ended questions. This section will be the larger part and consist of synchronous and asynchronous units. We learn about coding schemes and apply manual coding, use these codes to analyse specific empirical questions, may apply Latent Class Analysis to cluster people according to the codings, and look at possible ways to use automatic classification. We will discuss the specific contents in the first session and take into account your specific skills.

The seminar will be held in English (we can switch to German if everybody speaks German) and the topics can easily accommodate the participants' (reasonable...) wishes. I have several data sets we can analyse, but we can also, if we agree, conduct our own surveys.
You need to be interested in methods, and have already worked with Stata and/or R. Some sort of statistical knowledge is highly advisable, you should know about bivariate correlations, mean difference tests as well as OLS regressions.

You will be expected to read the literature, taken an active part in the seminar, do your own codings, and write an empirical term paper (8-10 pages) that uses open-ended questions that you coded on your own.