Zurück
| Semester | Herbstsemester 2016 |
| Angebotsmuster | unregelmässig |
| Dozierende | Joachim Scholderer (joachim.scholderer@unibas.ch, BeurteilerIn) |
| Inhalt | The course will provide participants with a state-of-the-art introduction to the most important family of statistical techniques in the social sciences: regression analysis and its extensions. The first part of the course will focus on linear regression models: - From conditional means to simple linear regression - Least-squares estimation and statistical inference for coefficients - Multiple linear regression - Analysis of variance - Assessing the performance of regression models The second part of the course will focus on situations where the assumptions of linear regression models are violated. Multi-level modelling is an extension of regression analysis that is particularly helpful when the observations in a study are not independently sampled and when the relationships between predictors and dependent variables are heterogenous across sub-populations: - Research designs resulting in clustered samples and longitudinal data - The mixed linear model - Correlated errors, random intercepts and random coefficients - Assessing the performance of multi-level models The third part of the course will focus on situations where the dependent variables are binary, nominal or counts of events: - Regression with transformed variables - Logistic regression - Generalised linear models and generalised linear mixed models The application of the techniques will be demonstrated in the context of contemporary social science data (incl. European Social Survey), using open-source software (R) as well as commercial statistical software packages (JMP, SAS, SPSS, Stata). |
| Lernziele | The Participants are able to - understand and critically discuss the results of regression and multi-level analyses reported in empirical research publications, - specify regression and multi-level models in such a way that they can provide answers to particular research questions and hypotheses, - estimate regression and multi-level models, using a statistical software package of their own choice, - document the results of regression and multi-level analyses, using generally accepted publication standards. |
| Literatur | Gelman, A. & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical models. Cambridge: Cambridge University Press. |
| Unterrichtssprache | Englisch |
| Einsatz digitaler Medien | kein spezifischer Einsatz |
| Intervall | Wochentag | Zeit | Raum |
|---|
Keine Einzeltermine verfügbar, bitte informieren Sie sich direkt bei den Dozierenden.
| Module |
Modul Methoden der Soziologie: quantitativ (Master Studienfach: Soziologie) Modul Soziologische Theorie (Master Studienfach: Soziologie (Studienbeginn vor 01.08.2013)) |
| Leistungsüberprüfung | Lehrveranst.-begleitend |
| Hinweise zur Leistungsüberprüfung | Participants will work on a small research project of their own (in teams of two to three) in which they apply the techniques discussed in this course. Each team will formulate a research question, operationalise it by selecting variables from the European Social Survey (ESS), specify an appropriate model and estimate it, document the results, discuss potential problems with the model and the data, and draw conclusions regarding the research question. The project should be documented in a paper that conforms to the standard structure of an empirical journal article (abstract, introduction, method, results, discussion) but may be considerably shorter. |
| An-/Abmeldung zur Leistungsüberprüfung | Anmelden: Belegen; Abmelden: nicht erforderlich |
| Wiederholungsprüfung | keine Wiederholungsprüfung |
| Skala | Pass / Fail |
| Wiederholtes Belegen | beliebig wiederholbar |
| Zuständige Fakultät | Philosophisch-Historische Fakultät, studadmin-philhist@unibas.ch |
| Anbietende Organisationseinheit | Fachbereich Soziologie |