|
Lehrveranstaltungen
|
Fortgeschrittene Verfahren der Mehrebenenanalyse und des internationalen Vergleichs: Multilevel analysis using comparative micro data -
- Dozentinnen/Dozenten:
- Michael Gebel, Chen-Hao Hsu
- Angaben:
- Seminar, 4 SWS
- Termine:
- Mo, 8:00 - 12:00, RZ/00.07
- Voraussetzungen / Organisatorisches:
- Voraussetzungen / Organisatorisches:
Students have to be familiar with the contents of the compulsory lecture “Research design” and multiple linear and binary logistic regression analysis. Moreover, students are required to be familiar with the statistics package Stata. These skills could either be acquired or refreshed in self-studies or by attending an online tutorial course. Link to Stata Tutorial: [https://vc.uni-bamberg.de/course/view.php?id=61372]
Registration:
Please register in the VC [https://vc.uni-bamberg.de/course/view.php?id=60856] of the event until 13.04.2023. Only the persons who are registered in the VC will receive all information about the course.
Type of instruction:
On-site teaching.
Language of instruction:
English.
Module-related examination:
Portfolio (time: 3 months) in English
- Inhalt:
- Learning targets:
The central aim of this course is to empower participants to critically discuss basic concepts and assumptions of multilevel analyses using comparative micro data, to conduct theory-driven empirical research, to choose and specify the appropriate models according to the ideas of modern causal analysis, to take the levels and scales of measurement of variables into account, to independently carry out multilevel analyses using the statistics package Stata, to correctly interpret and clearly present the results of multilevel analyses in tables and graphs.
Course contents:
Multilevel analysis has been especially applied to the case of many clusters and few observations within each cluster, e.g. in education research on pupils nested in classes and schools. However, multilevel analysis is more and more often used in international comparative social research. Following the latter comparative research tradition, this course is devoted to the specific problem of a small number of countries and large numbers of individuals within each country. This course quickly repeats the foundations of multiple regression analysis, introduces the basics of multilevel analysis and focuses on advanced topics of multilevel analysis. Models for continuous and binary dependent variables are presented. Moreover, it is discussed how many and which kind of macro level variables should be included to account for the macro context. A specific focus is on the specification and interpretation of cross-level interactions. Exemplary empirical studies and current debates in multilevel analysis of comparative micro-data are discussed. In lab sessions participants will learn how to practically implement multilevel analyses of comparative microdata using the statistics package Stata. The lab sessions and the seminar theses will draw on topical sociological questions and data of the European Social Survey (ESS).
|
|
Fortgeschrittene Verfahren der Querschnittsanalyse: Advanced Regression Analysis using Stata -
- Dozentinnen/Dozenten:
- Michael Gebel, Daniel Zeddel
- Angaben:
- Seminar, 4 SWS
- Termine:
- Mo, 14:00 - 18:00, RZ/00.07
- Voraussetzungen / Organisatorisches:
- Voraussetzungen / Organisatorisches:
Students have to be familiar with the contents of the compulsory lecture “Research design” and the basics of multiple linear and binary logistic regression analysis. Moreover, students are required to be familiar with the statistics package Stata. These skills could either be acquired or refreshed in self-studies or by attending an online tutorial course. Link to Stata Tutorial: [https://vc.uni-bamberg.de/course/view.php?id=61372]
Registration:
Please register in the VC [https://vc.uni-bamberg.de/course/view.php?id=60858] of the event until 13.04.2023. Only the persons who are registered in the VC will receive all information about the course.
Type of instruction:
On-site teaching.
Language of instruction:
English.
Module-related examination:
Portfolio (time: 3 months) in English or German
- Inhalt:
- Inhalt:
Learning targets:
The aim of this course is to empower participants:
-to critically discuss basic concepts and assumptions of multiple linear and logistic regression analyses,
-to conduct theory-driven empirical research
-to choose and specify the appropriate regression models according to the ideas of modern causal analysis
-to carry out regression analyses (multiple linear, binary logistic, ordinal logistic, and multinomial logistic) using the statistics package Stata,
-to interpret and present the results of regression analyses in tables and graphs.
Course contents:
We will shortly repeat the foundations of bivariate and multiple linear regression analysis and, then, focus on advanced topics of multiple regression analysis. The course is structured around four key topics of cross-sectional data analysis using parametric regression techniques: (1) multiple linear regression, (2) binary logistic regression, (3) ordinal logistic regression, and (4) multinomial logistic regression.
In lab sessions, participants will learn how to implement regression analyses using the statistics package Stata. The lab sessions and the seminar theses will draw on sociological questions and data of the German Social Survey (ALLBUS).
|
|
![](/img/anew/void.gif) |
![](/img/anew/void.gif) |
|
UnivIS ist ein Produkt der Config eG, Buckenhof |
|
|