(organisiert vom Projekt GENIAL-forschen +. GEschlechterpoteNzIALe nutzen - Gesellschaft verändern)

Inhalt:

Das Actor-Partner Interdependence Model (APIM) ist ein Verfahren der Strukturgleichungsmodellierung, das speziell für dyadische Daten entwickelt wurde. Es berücksichtigt, dass zwei Personen innerhalb einer Dyade nicht unabhängig voneinander handeln, sondern sich wechselseitig beeinflussen. Damit eröffnet APIM Forschenden die Möglichkeit, Interdependenzen in Paaren, Freundschaften, Eltern bzw. Erziehungsperson-Kind- oder Kolleg:innen-Beziehungen systematisch zu analysieren.

Während klassische statistische Modelle in der Regel von Unabhängigkeit zwischen Untersuchungseinheiten ausgehen, trägt APIM der Tatsache Rechnung, dass Interaktionen und Ergebnisse in dyadischen Beziehungen miteinander verknüpft sind. So lassen sich sowohl Akteureffekte (Einfluss einer Person auf ihre eigenen Ergebnisse) als auch Partnereffekte (Einfluss einer Person auf die Ergebnisse der anderen Person) simultan schätzen.

Gerade in der geschlechtersensiblen Forschung ist der Einsatz von APIM sinnvoll. Geschlecht prägt dyadische Beziehungen nicht nur individuell, sondern vor allem in der Dynamik zwischen den Beteiligten. APIM stellt so ein besonders geeignetes Instrument dar, um die komplexen Wechselwirkungen von Geschlecht und sozialer Interaktion adäquat zu erforschen. 

Über den Referenten:
Dr. Kay Brauer forscht an der MLU Halle-Wittenberg u.a. zu Playfulness, dem Imposter Syndrom und romantischen Beziehungen. Er hat APIM u. a. in Studien zu Playfulness in romantischen Beziehungen angewandt und in internationalen Fachzeitschriften veröffentlicht.

Die praktische Anwendung erfolgt mit der Statistik Software MPLUS. Zur Teilnahme ist die kostenlose MPLUS-Demoversion ausreichend (Download hier)

Zeit: 15. Oktober 2025, 9:00–16:00 Uhr (inkl. Mittagspause)

Ort: MG2/00.09 (PC-Pool - Verwendung eines eigenen Laptops möglich!) 

Arbeitseinheiten: 8

 Anmeldung hier bis zum 08.10.2025 oder bis zur Vergabe aller Plätze möglich

(organized by BAGSS and LIfBi)

Short outline

This course will introduce you to the concepts and methods of causal inference and causal modeling in the social sciences. It will highlight the relevance of research design, analytical methods, and their systematic combination to optimize the validity of causal inferences drawn from empirical studies, with a special focus on longitudinal analysis.
Building on existing knowledge concerning linear regression modelling and research design, the course will then cover key methods to estimate causal effects, including fixed effects estimations with various addons, event study analyses, matching, difference-in-differences, regression discontinuity, and instrumental variables. Throughout the course, you will apply these concepts and methods in hands-on sessions to real-world examples in the social sciences. The application
will be conducted with the statistical software package Stata. A solid background in Stata is expected.
Before the course starts, participants will complete a survey in which they will be asked to state their preferences regarding what they would like to be covered in the course from a set of potential topics.

Topics (potentially) covered include:
- The counterfactual framework and treatment effects: What are ATE, ATT, and LATE?
- Defining estimands and how they relate to estimation
- Fixed Effects Regressions (with a focus on time-distributed fixed effects/event studies/impact dummies and individual-slope regressions)
- Regression Discontinuity Designs
- Difference-in-Differences estimations
- Instrumental Variable Regressions

The course is split between sessions in which the concepts and methods are introduced theoretically and via existing research, and between hands-on sessions in which we will apply the methods in practice using Stata.
 

Prerequisites

- Knowledge of basic statistical concepts and their formal background, including the principles of linear regression
- Solid background in Stata
- Basic understanding of designing quantitative studies
 

Date & Time

Monday, November 3, 2025, 09.00 a.m. to 04.00 p.m. and Friday, November 7, 2025, 09.00 a.m. to 03.00 p.m.
 

Place

BAGSS, Feldkirchenstraße 21, 96050 Bamberg, Room FG1/00.06
 

About the Trainer

Matthias Collischon is a researcher (tenure-track) at the Institute for Employment Research in Nürnberg. He is an IZA-research affiliate and a LASER research fellow. His research interests include gender differences in the labor market, the effects of unemployment on various wellbeing measures and long-term effects of the COVID-19 pandemic on the labor market. His work has been published in high-ranking journals such as Social Forces, Sociological Science, the Journal of Human Resources, European Sociological Review and Socio-Economic Review, among others.

Registration

To register, please send an email to courses.bagss(at)uni-bamberg.de by Monday, October 13, 2025. Registration is mandatory. The number of participants is limited to 16.

Short Outline:
Qualitative Content Analysis is an established method for systematic analysis of text-based data. Simultaneously artificial intelligence, especially large language modeling and integrated AI-tools in QDA-software, provides new opportunities – and challenges – for the analysis pro-cess. How can we use these tools efficiently without putting the methodological stringency and the interpretive sovereignty of scientists at risk?
This workshop establishes a connection between classic qualitative content analysis (esp. see Kuckartz/Rädiker) and current developments in AI. The aim is to provide strategies for reflecting on AI tools from a methodological perspective and for efficiently implementing them in Quali-tative Content Analysis.
Topics covered:

- Fundamental principles of Qualitative Content Analysis
- Application scenarios: LLM-Browsers vs. AI-Tools (e.g. Tailwind and QInsights)
- Hands-on integration in the process steps of QCA
- AI to support the exploratory phase and initial material inspection
- AI as a “sparrings partner” in inductive category building and deductive application form categorical systems. AI for summaries von material within categories. Strategies for val-idation and critical assessment of AI-generated results
- Methodological reflection and pitfalls: Discussion about bias and data protection. Secur-ing the traceability and agency of scientists.

After the workshop, participants will be able to assess where the use of AI tools can be useful in QIA. They will be familiar with specific application examples for LLM's and/or CAQDAS AI functions and will be able to test and use these tools critically and reflectively in their own re-search practice, taking methodological and ethical aspects into account.

Instructor:
Thorsten Pehl, audiotranskription

Date & Time:
Monday, December 01, 2025, 09.00 a.m. to 04.30 p.m. (s.t.)

Place:
BAGSS, Feldkirchenstraße 21, 96050 Bamberg, Room FG1/00.06

Registration:
To register, please send an email to courses.bagss(at)uni-bamberg.de by Monday, November 10, 2025. Registration is mandatory. The number of participants is limited to 20.
 

Short Outline

This workshop explores how artificial intelligence can support and enhance data visualization practices in the social sciences. Through a combination of conceptual input and applied exam-ples, participants will examine how AI-assisted tools can help create clear, engaging, and ana-lytically meaningful visual representations of complex datasets. Special attention will be given to the visualization of longitudinal data, the design of dynamic graphics, and the integration of visualization into the broader research process. The workshop is intended to be accessible and relevant to researchers from various social science backgrounds.
During the sessions, there will be some reference to R for practical examples and demonstra-tions.

Date & Time:
Tuesday, October 07, 2025, 09.00 a.m. to 04.00 p.m. (s.t.)
Wednesday, October 08. 2025, 09.00 a.m. to 04.00 p.m. (s.t.)

Place:
BAGSS, Feldkirchenstraße 21, 96050 Bamberg, Room FG1/00.06 

Registration:
To register, please send an email to courses.bagss(at)uni-bamberg.de by Monday, September 22, 2024. Registration is mandatory. The number of participants is limited to 16

About the Trainer:
Giovannipaolo Ferrari is Lecturer in the Department of Human, Philosophic and Education Sciences (DISUFF) at the University of Salerno in Italy where he teaches Sociology, Sociol-ogy of Sport and Well-being, Sociology of Health, Research Methods (Qualitative and Quantitative), Data Analysis and Digital Methods. His research interests focus on sports, environmental, social, and labour policy analysis, digital methods for social research, and public sociology.