GESIS Training News

The GESIS Summer School 2024 is scheduled from 24 July to 16 August 2024. With just two weeks until the start, now is your last chance to register! Some courses will be conducted onsite in Cologne, while others will be held online. Join lecturers and attendees from diverse backgrounds and locations worldwide, both in-person and virtually, to participate in Europe's leading summer school in survey methodology, research design, and data collection.

This year's courses:

Week 0 (24–26 July) – Short Courses

Introduction to R for Data Analysis [24–25 July | waiting list]

Jan Schwalbach (GESIS), Dennis Abel (GESIS)   

Developing, Translating, and Pretesting Questionnaires for Cross-cultural Surveys

Dorothée Behr (GESIS), Cornelia Neuert (GESIS), Lydia Repke (GESIS)   

Introduction to Stata for Data Management & Analysis

Mazlum Karatas (GESIS), Lynn-Malou Lutz (GESIS)   

Week 1 (29 July–02 August)

Factorial Survey Design

Katrin Auspurg (LMU Munich), Carsten Sauer (Bielefeld University), Alisia Bauer (LMU Munich)   

Introduction to Questionnaire Design

Marek Fuchs (Darmstadt University of Technology)   

Designing and Implementing Web Surveys

Melanie Revilla (IBEI)    

Causal Inference with Directed Acyclic Graphs (DAGs) [31 July–02 August]

Paul Hünermund (Copenhagen Business School)   

Week 2 (05–09 August)

Causal Inference Using Survey Data

Heinz Leitgöb (Leipzig University), Tobias Wolbring (FAU Erlangen-Nürnberg)

Survey Sampling and Weighting

Simon Kühne (Bielefeld University)   

Mixed-Mode Surveys [07–09 August]

Henning Silber (GESIS), Sven Stadtmüller (HAWK Göttingen), Peter Schmidt (University of Giessen), Yannick Diehl (University of Marburg)    

Week 3 (12–16 August)

Introduction to Survey Design

Bella Struminskaya (University of Utrecht), Peter Lugtig (University of Utrecht)    

Data Science Techniques for Survey Researchers

Anna-Carolina Haensch (LMU Munich and University of Maryland)   

Introduction to Small Area Estimation

Angelo Moretti (University of Utrecht)   

Applied Multiple Imputation

Ferdinand Geißler (Humboldt-University Berlin), Jan Paul Heisig (WZB Berlin and Freie Universität Berlin)    

ECTS Credits & More

Thanks to our cooperation with the Center for Doctoral Studies in Social and Behavioral Sciences at the University of Mannheim, participants can obtain a certificate acknowledging a workload worth 4 ECTS credit points per one-week course. More information is available here.

You will find the full program, detailed course descriptions, and more information here.

Are you tired of staring at your data and wondering what caused what? Then, it is time to explore the fascinating world of causal inference methods! Get ready to unravel the mysteries of cause and effect with our upcoming online workshops. Whether you are a novice or a seasoned pro, there is something here to satisfy your analytical appetite.🧠📊

If you are just dipping your toes into the causal inference waters, our Introduction to Methods of Causal Inference workshop is the perfect starting point. Designed to lay a solid foundation, this course will guide you through fundamental concepts and principles of causal inference, putting you on the path to causal enlightenment.

For those already familiar with the basics of causality, our workshop program offers a diverse array of opportunities to delve deeper into specific methodologies and stay abreast of recent developments. You can explore the intricacies of Propensity Score Matching, unravel the power of Instrumental Variables, or dive into the world of Causal Mediation Analysis.

But that is not all! We have an exciting new lineup of workshops on the horizon:

• Applied Multiverse Analysis with Stata and R
• Synthesizing Evidence: Aggregating Support for your Hypothesis across Studies
• Introduction to Geospatial Techniques for Social Scientists in R
• Introduction to Bayesian Statistics
• Expert*inneninterviews
• Grundlagen und aktuelle Debatten der Regressionsanalyse

For additional details, registration, and our complete workshop program, visit our workshop website or have a look at the complete program below.

The Fall Seminar takes place from 30 August to 27 September 2024 and offers a variety of introductory and advanced courses in computational social science methods. It targets researchers who want to collect and analyze data from the web, social media, or digital text archives. All courses are taught in English and lectures in each course are complemented by hands-on exercises giving participants the opportunity to apply these methods to data. Some courses are held in-person in Mannheim, others online – keep an eye out for our brand-new blended learning format!

Please find our full course program below:

Introduction to Computational Social Science with R

[30 August-05 September | online blended learning]

Johannes B. Gruber (University of Amsterdam)   

Introduction to Computational Social Science with Python

[30 August-05 September | online blended learning]

John McLevey (Waterloo University)   

Web Data Collection with Python and R

[09-13 September | Mannheim]

Iulia Cioroianu (University of Bath)   

Introduction to Social Network Analysis

[16-20 September | Mannheim]

Philip Leifeld (University of Manchester)   

Introduction to Machine Learning for Text Analysis with Python

[16-20 September | Mannheim]

Marieke van Hoof (University of Amsterdam), Rupert Kiddle (Vrije Universiteit Amsterdam)   

Agent-Based Computational Modeling

[16-20 September | Mannheim]

Michael Mäs (Karlsruhe Institute of Technology), Fabio Sartori (Karlsruhe Institute of Technology)   

Advanced Social Network Analysis with R

[23-27 September | Mannheim]

Michal Bojanowski (Kozminski University & Autonomous University of Barcelona)   

From Embeddings to LLMs: Advanced Text Analysis with Python

[23-27 September | Mannheim]

Hauke Licht (University of Cologne), Lisa Maria Lechner (University of Innsbruck)   

Automated Image and Video Data Analysis with Python

[23-27 September | online]

Andreu Casas (Royal Holloway University of London), Felicia Loecherbach (University of Amsterdam)   

ECTS Credits & More

Thanks to our cooperation with the a.r.t.e.s. Graduate School for the Humanities at the University of Cologne, participants can obtain a certificate acknowledging a workload worth 2 ECTS credit points per one-week course. More information is available here.

All courses are stand-alone and can be booked separately – feel free to mix and match to build your own personal Fall Seminar experience that perfectly suits your needs and interests. There is no registration deadline, but places are limited and allocated on a first-come, first-served basis. To secure a place in the course(s) of your choice, we strongly recommend booking soon.

For detailed course descriptions and registration, please visit our website.

We're excited to announce the start of the first KODAQS Data Quality Academy Certificate Program cohort this fall! The six-month, hands-on Certificate Program aims to train early-career researchers on conceptualizing, diagnosing, and fixing social science data quality problems. The Program's innovative blended learning format combines self-study materials with weekly touchpoints and networking activities. Check our website after 22 July 2024 for more information once the Call for Applications for the first cohort opens.

Thomas' research interests include psychometrics and data analysis of multi-item rating scales and cognitive assessment tests. As part of his daily work for "KODAQS" – the Competence Center for Data Quality in the Social Sciences, he develops concepts and courses, tutorials, and packages in R or Stata, referring to data quality topics, especially in survey data. Previously, he worked in the Scale Development and Documentation team at GESIS and was part of ZIS, the Open Access Repository for Measurement Instruments.

Together with his co-lecturer Matthias Roth, he will teach the course "Data Quality Assessment for Survey Responses: Be Careful of the Careless" at GESIS Workshops. The online course will take place from 15–16 October 2024.

How did you become interested in your subject?

Thomas: My fascination with data quality in surveys stems from working extensively with various lengthy questionnaires in scale development and survey projects. I’ve often seen how fluctuating data quality in responses can dramatically impact results. Poor data quality can not only lead to inaccurate conclusions, but it also becomes glaringly obvious when substantial data loss occurs. Planned, in the best case, pre-registered analyses (e.g., group comparisons, latent models) can no longer be modeled properly, forcing you to switch to other, less robust methods. During the data preprocessing and analyzing process, you're faced with tough decisions about which data quality indicator is better suited for a certain task and where to start or stop excluding cases. These challenges make the topic both exciting and crucial for me, and I’d like to share some insights from survey research.

What lessons can participants draw from your GESIS course?

Thomas: My co-lecturer Matthias and I want to convey that the best way to dive into data quality is by thoroughly understanding your constructs and measurement instruments. And don’t forget to literally “Date your data”! Our course is based on KODAQS fundamentals, where participants will learn how to frame, find, and fix data quality issues, and bring data quality practices into a working flow. By the end of our course, they will gain a solid understanding of data quality issues, effectively evaluate survey responses, and handle low-quality data. We also provide numerous applied examples in R from everyday research and encourage participants to apply these to their studies. Ideally, this will be done in the spirit of open science, promoting transparency and reproducibility.

What do you enjoy most about being a social scientist?

Thomas: As a psychologist, I am very curious about human experience and behavior. As a quantitative social scientist, I can also relate to the metaphor of a qualitative researcher who’s working as a “miner” or “traveler.” While “mining” refers to a systematic approach, where the researcher meticulously digs through the data to uncover specific insights and patterns, “traveling” reflects a more explorative approach. I enjoy the most the tension between these more protocol-based approaches and the researcher’s degrees of freedom.

We thank Thomas for his insights and look forward to his and Matthias’ course in October.