At the Chair of Political Communication, we deal with the entire spectrum of political communication. A first focus is on the role of media coverage in the political process, for example, regarding news quality or the impact of media on political attitudes and behavior. A second focus is on political campaign communication: We investigate how parties and other actors communicate, how their messages affect both traditional and digital media, and what technical possibilities they use in election campaigns. Furthermore, we analyze how citizens themselves communicate politically and inform themselves. Our research thus shows how journalism, campaigns, and citizen communication intertwine and how the interrelationship between media, politics, and the population shapes public opinion formation and democratic processes.

Funded by the German Research Foundation (DFG; 512131520)

Duration: 2024-2026

Mass media political reporting focuses on problems and conflicts and is therefore heavily negatively biased. The consequences of this problem-centered reporting for the audience range from negative emotions such as fear and anger, to negative and polarized judgments towards politics, and a low willingness to participate in societal problem-solving. As an alternative to established, problem-centered journalism, constructive journalism has therefore been discussed for several years.
Based on a working definition of constructive journalism as journalism that reports on socially relevant problems and ways to solve them in a balanced and objective manner, the goal of this project is a) to theoretically model the effects of constructive journalism on recipients' emotions, cognitions, and behavior (intentions), and b) to empirically investigate these effects, as well as their prerequisites (content and use of constructive journalism) and boundary conditions (topic and person characteristics).
To this end, a manual quantitative content analysis will first be used to examine how media outlets that identify with constructive journalism report on socially relevant topics and how this differs from the reporting of established news media. Building on this, an experimental survey will assess what proportion of the audience prefers constructive journalism over problem-centered media reporting in a specific decision-making situation and on what boundary conditions this depends. Finally, the specific effects of the various dimensions of constructive reporting (balance, objectivity, solution-orientation) on different reporting topics and on people with different predispositions will be experimentally investigated.

Project Manager: Prof. Dr. Marcus Maurer
Project Staff: Matthias Mack, M.A.
Partner Institution: Prof. Dr. Olaf Jandura (Communication Research in the Faculty of Economics at Düsseldorf University of Applied Sciences)

Funded by the German Research Foundation (DFG; FKZ 543762082)

Duration: 2025-2027

The study of digital communication by non-institutionalized actors faces challenges regarding (1) the systematic definition, sampling, and identification of relevant actors, as well as (2) the specific classification of relevant actors and content based on ideological, typological, and other content-related characteristics (classification). This research project aims to contribute to the reliability and validity of identifying and classifying heterogeneous groups of actors on digital platforms. The design includes various identification, classification, and simulation studies based on manual and computational methods. It investigates how different procedures for identifying and classifying radical and extremist actors on Telegram differ in terms of reliability and validity (FF1), and how various decisions in identification and classification influence the identified actor constellations and content on Telegram (FF2). This involves a) systematizing strategies for identifying, sampling, and classifying radical and extremist actors in digital communication, b) identifying an approximate total population of this (German-speaking) target group using network analytical procedures, and c) classifying the actors using manual and automated procedures with various specifications for collecting actor characteristics. In d) evaluation studies, the effects of various identification and classification decisions are simulated and evaluated for their impact on actor constellations and content. Through e) documentation and archiving, long-term accessibility and reusability of the findings are ensured via a reference dataset, and best practices are derived, which, within the framework of f) transfer of knowledge, should contribute to the informed identification and classification of relevant actor groups by civil society and international research.

Project Managers: Dr. Pablo Jost & Prof. Dr. Annett Heft (Eberhard Karls University Tübingen)
Project Staff: Harald Sick

Funded by the German Research Foundation (DFG; FKZ 519731504)

Duration: 2023-2026

This research project (together with Sabine Trepte, Institute for Communication Studies at the University of Hohenheim) investigates the perception and impact of political online microtargeting by German parties in the 2024 European Election. Innovative methods are used to determine to what extent tailored advertisements on Facebook are perceived by voters and how they influence political attitudes, political knowledge, and participation. A particular focus is placed on the role of individual needs for privacy and citizens' informational self-determination in these perception and impact processes.
Firstly, qualitative methods are used for preliminary exploration of the perception of political online microtargeting by voters. Secondly, a combination of user tracking and a four-wave panel survey during the European election campaign is employed to analyze the effects of advertisements used by German parties on attitudes, willingness to participate, and political knowledge, depending on the sociodemographic factors and party affiliation of the surveyed individuals. Thirdly, experimental studies are conducted to investigate to what extent the use of personal data and the congruence between election advertising and the characteristics of the recipients influence the effects of online microtargeting. Based on the findings, the project provides instructions on where regulatory measures can be applied and examines at which stages of the perception process threats to democratic processes arise and for which user groups concrete action is needed.

Project Managers: Prof. Dr. Sabine Trepte (Institute for Communication Studies, University of Hohenheim), Prof. Dr. Marcus Maurer & Dr. Simon Kruschinski
Project Staff: Hanna Paulke, M.A.

Funded by the German Research Foundation (DFG; 462702138)

Duration: 2023-2027

The research group (among others, together with Goethe University Frankfurt, LMU Munich, and the Leibniz Institute for Educational Research and Educational Information (DIPF)) investigates how students use online sources for learning and to what extent their use influences the processing of generic and domain-specific tasks related to "Critical Online Reasoning" (COR).

This subproject specifically aims to describe the online information landscape in which students operate when solving generic or domain-specific tasks during their program of study. This is done through a quantitative content analysis of web pages that students visit when working on COR tasks. Special attention is given to analyzing the correctness (e.g., completeness and balance) and comprehensibility of the online information and its influence on performance in the tasks. Furthermore, the relationship between the information media used by students, their COR skills, and their learning outcomes throughout their course sequence is investigated using panel data.

Project Managers: Prof. Dr. Marcus Maurer & Prof. Dr. Christian Schemer

Project Staff: Tobias Scherer & Alice Laufer

Funded by the Federal Ministry for Research, Technology, and Space (BMFTR; FKZ 01UP2229A)

Duration: 2023-2026

In this collaborative project (together with the Ubiquitous Knowledge Processing Lab of the Computer Science Department at TU Darmstadt), consensus and polarization in the positions of different societal groups (science, politics, media, population) regarding measures to combat the COVID-19 pandemic are measured on the social network Twitter. Using innovative methods from the field of Natural Language Processing (NLP), opinion expressions are to be automatically captured, and opinion dynamics statistically modeled using time-series analytical procedures, in order to identify causes and developments of societal polarization processes. Specifically, among others, the following questions will be answered: How did various societal groups (e.g., politics or the media) and subgroups (e.g., different parties and media with different editorial lines) evaluate the Corona measures, how did this change over time, and how did the positions of the different groups mutually influence each other? Since the NLP models developed here can also be transferred to future crises, the project allows for the identification of general patterns in the emergence of consensus and polarization in crises and enables a kind of societal early warning system that can detect emerging polarization tendencies. In addition to this substantive goal, the project also pursues two methodological goals: Firstly, comparisons of opinion expressions measured on Twitter with representative population surveys are intended to provide information on how well the discourse on Twitter is suited as an indicator for public opinion. Secondly, the innovative NLP procedures are to be applied to social science questions and thereby further developed.

Further information can be accessed on the research project's website: www.kopocov.de

Project Managers: Prof. Dr. Marcus Maurer & Prof. Dr. Iryna Gurevych (Computer Science Department, TU Darmstadt)
Project Staff: Dr. Simon Kruschinski & Tilman Beck, M.A. (Computer Science Department, TU Darmstadt)

Funded by the Bertelsmann Foundation

Duration: 2024-2025

The aim of the project is to better understand and specifically improve political communication for young people aged 16 to 26 on social platforms such as TikTok and Instagram. The focus is on identifying what content, forms of presentation, and communication styles actually reach young target groups and what expectations they have of political actors in social media.
In the first step, the project team analyzes approximately 75,000 political short videos and images from around 2000 social media accounts of politicians, parties, and political influencers using automated content analyses. With the support of Large Language Models (LLMs), in addition to topics, positionings, and communication and mobilization strategies, visual characteristics such as gender or facial expressions of the appearing individuals are also examined. The analysis aims to systematically identify the success factors (e.g., virality, engagement) of political content.
In the second part of the project, the target group's perspective is captured: In focus groups and a representative online survey, the perception, usage behavior, and expectations of young people regarding political communication are investigated. Based on this, recommendations for action are developed for a more effective, democracy-strengthening approach to young people in the digital space – together with youth committees that accompany the project in an advisory capacity.

First Interim Report for Download

Project Lead: Dr. Pablo Jost, Project Staff: Hannah Fecher & Yannick Winkler.
Partner Institutions: Das Progressive Zentrum, Mercator Foundation

Funded by the Baden-Württemberg Foundation & Seed Grant of the Research Priority Area "AI & Politics" of the Amsterdam School of Communication Research at the University of Amsterdam

Duration 2025-2026

The "CampAIgn Tracker" project aims to increase transparency in dealing with AI-generated content in the political context. In digital election campaigns, AI-generated images and videos are increasingly used, which can potentially be used to influence public opinion. Especially on social media platforms like Facebook, Instagram, YouTube, TikTok, and in messenger services like Telegram, the dissemination of such AI content is difficult to trace, which can lead to a loss of trust in political actors and the democratic process.
The project solves this problem by developing a platform that collects and analyzes AI-generated posts and advertisements from political social media accounts. In the back-end, posts and advertisements are identified as AI-generated content through (semi-)automated classification. On the one hand, this is done by AI models trained for this purpose. On the other hand, this is validated by trained human coders. In addition, (automated) content analyses are performed to identify discussed topics, actors, or forms of negative campaigning in the AI content.
The results of these analyses – including the number of AI-generated content pieces, their content-related communication strategies, as well as the advertising budgets used or reach achieved – are made available to the public on www.campaigntracker.de. This platform provides insights into the scope and dissemination of AI-generated content in election campaigns, thus making digital AI campaigns transparent.

Project Lead: Dr. Simon Kruschinski and Dr. Fabio Votta (University of Amsterdam)

BA Inhaltsanalyse: Inhalte öffentlicher Kommunikation KF
Instructor: Dr. Nora Denner; Felix Valentin Dietrich; Prof. Dr. Simone Christine Ehmig; Hannah Fecher; Prof. Dr. Marcus Maurer; Prof. Dr. Michael Scharkow; Yannick Winkler
BA Politische Kommunikation - Seminar (mit Anwesenheitspflicht) KF/BF
Instructor: Dr. Pablo Jost; Prof. Dr. Marcus Maurer; Prof. Dr. Oliver Quiring; Prof. Dr. Christian Schemer; Tobias Scherer; Frank Schneider; Baruch Shomron
BA Politische Kommunikation - Vorlesung KF/BF (mit Studienleistung)
Instructor: Prof. Dr. Marcus Maurer
DK K Kolloquium zur Masterarbeit Wintersemester 2026/27
Instructor: Prof. Dr. Marcus Maurer; Prof. Dr. Oliver Quiring; Prof. Dr. Leonard Reinecke; Prof. Dr. Christian Schemer; Prof. Dr. Birgit Stark

WiSe 2026/27

DK K Kolloquium zur Masterarbeit Wintersemester 2026/27