Relevance & Research Question

The literature identifies different types of masculinity. Hegemonic masculinity refers to the type of masculinity legitimating unequal gender relations (between men and women, between masculinity and femininity and among masculinities). Hegemonic masculine norms for the "ideal man" include traits and practices like being strong, successful, independent, unemotional, in control. Hypermasculinity, aggressivity, sexual prowess are generally glorified.

Differently, caring masculinity refers to the idea that, without rejecting masculinity, men are able to adopt what (traditionally) is seen as a feminine characteristic. Here, the emotional dimension is crucial.

Recent research refers of men negotiating masculinity e.g., not feeling challenged ‘as men’ because of their caregiving roles but reinforce interest in ‘manly’ hobbies/sports suggesting processes in which they negotiate what aspects of masculinity do not fit with their identity and what they do.

Extreme-right and manosphere ideologies tend to privilege hegemonic masculinity, which legitimizes unequal gender relations and anti-democratic stands. But how to measure these norms to investigate their spread in the public opinion?

This contribution describes the development, both conceptually and empirically, of the measurement of hegemonic masculinity norms and explore their spread among the general population of six European countries.

Methods & Data

The contribution reports on the development of the new “Hegemonic Masculinity Norms Scale” drafted in the context of the MEN4DEM (Masculinities for the future of European democracy – Horizon Europe, GA n. 101177356), on the cross-cultural comparability challenges (including use of advance translation), results from the online survey experiment conducted to assess reliability and validity. The final measurement is then used to show the spread of hegemonic masculinity norms in Greece, Germany, Italy, Netherlands, Poland, and Sweden. (N=800 in each of the 6 countries; representative samples randomly assigned to exp.setting)
Results
Fieldwork will finish in November 2025
Added Value
The “Hegemonic Masculinity Norms Scale” brings the following innovation: it combines a multidimensional perspective to masculinity with a continuum approach in its measurement, allowing for in-depth analysis of public opinion on masculinity, an aspect still largely unexplored. Methodology applied to develop the instrument in a cross-national setting can be of interest for cross-cultural survey scholars.

Relevance & Research Question

In emerging risk contexts and situations requiring contingent data collection, social media surveys can play a crucial role in analysing population needs and stress levels, offering an innovative monitoring tool. Bradyseism is a phenomenon of slow ground deformation that occurs in the Campi Flegrei area, in the Naples proximity (South of Italy). This phenomenon can cause shallow earthquakes, damage to structures, and an increase in seismic activity. The Campi Flegrei area is characterized by high population density and strong socioeconomic inequalities. This study aims to assess stress levels in the population affected by bradyseism in the Campi Flegrei area.

Methods & Data

An online survey was conducted during a period of increased bradyseism activity in the Campi Flegrei area. Meta advertisement platform was used to recruit participants exposed to the phenomenon. The survey, lasting 11 days, collected over 600 completed questionnaires with an investment of 720 Euros. This approach allowed for targeted spatial sampling of the population. The study included residents in the arounds of Pozzuoli (high-risk area) and Ottaviano (control area).

Results

The survey provided valuable information on stress levels among the affected population. The propensity to respond to a survey on social networks is proportional to the interest in the survey topic. This resulted in a higher number of responses from those affected by the earthquakes, despite our efforts to achieve a balanced sample. The data collected offered insights into the differential impacts of the bradyseism phenomenon across various demographic groups, with a particular focus on gender-based vulnerabilities. Moreover, the stress level, measured using a ten-item psychometric scale, showed a high correlation with both the number of perceived earthquake tremors and their intensity.
Added Value
This study demonstrates the effectiveness of social media-based surveys in rapidly gathering data during emerging risk situations in a specific area. It provides a cost-effective and time-efficient method for assessing population needs and stress levels in volcanic contexts. The findings contribute to a better understanding of the socio-economic and gender-based impacts of bradyseism, potentially informing more targeted and inclusive public policies for disaster preparedness and response.

Relevance & Research Question

This study addresses the growing need for transparent, theory-based methods to analyze political text using artificial intelligence. It asks whether Large Language Models (LLMs) can reliably identify political stances—such as ideological orientation, support for liberal democratic values, and populist rhetoric—without the need for manually labeled data.

Methods & Data

The paper employs a Natural Language Inference approach leveraging a retrained model, DEBATE (Burnham et al. 2024). In so doing, we build on expert survey codebooks from reputable sources in political science. The model is validated by comparing its outputs with expert-assigned scores to assess accuracy.

Results

The findings show no statistically significant differences between the model’s classifications and those of human experts. The model also demonstrates strong multilingual capabilities across English, German, Italian, Portuguese, and Spanish.

Added Value

This study introduces a cost-effective, replicable, and theoretically grounded approach for stance detection in political texts. By eliminating the need for data labeling and integrating the method into the forthcoming Automated Political Stance Identification (APSI) platform, it provides an accessible tool for researchers, policymakers, civil society, and the public seeking evidence-based insights into ideological and rhetorical patterns in politics.

Relevance & Research Question
Addressing complex market research questions often requires the integration of data from multiple sources, the development of reliable indices, and coordinated collaboration across organizations. In syndicated studies, even market competitors may work together within a shared research framework. Achieving high quality outcomes demands a deep understanding of the underlying data as well as the expectations and needs of all stakeholders. Only then can indices be developed that are robust, transparent, and capable of supporting strategic decision making.
Methods & Data

For a syndicated study, data from two competing companies have been pooled to date. A new questionnaire was developed specifically for the study in which internal employees were surveyed. In addition, data from various customer satisfaction studies were integrated. Financial performance indicators from the partner companies and their suppliers were also included.

From these inputs, distinct non-overlapping indicators were constructed, complemented by a single comprehensive index that synthesizes all individual measures. The development process took place in close collaboration with the syndicated study partners. Moreover, suppliers and customers were invited to contribute their perspectives and assessments, ensuring that the resulting indicators reflect a broad and balanced understanding of the market context.

Results
The analysis highlights several factors that are essential for the development of meaningful indicators. Although large volumes of data are often available, careful dimensionality reduction is necessary to maintain clarity and to avoid overly complex structures. Indicators tend to overlap, which can create unintended effects and lead to misplaced analytical focus. In addition, integrating data from various empirical surveys and KPIs requires appropriate statistical procedures. In our case, we apply, among other methods, inverse-variance weighting to ensure high reliability of the indicators.
Added Value
We illustrate how heterogeneous data, differing organizational perspectives, and collective development processes can be brought together to create a coherent and reliable measurement framework. The insights gained expand the understanding of how indices can be designed in collaborative settings and show how transparent methods increase acceptance and long-term usability. We offer practical guidance to develop robust indicators for complex syndicated studies.

Relevance & Research Question

When news of Ozzy Osbourne’s death broke in July 2025, millions of Facebook users brought collective grief together in a global digital ritual. This study examines how mourning, nostalgia, and fandom unfolded on social media, asking: how does celebrity death evolve into a networked media event? Building on media event theory (Dayan & Katz, 1992) and the concept of affective publics, the research explores how emotions, algorithms, and cultural memory intersect when a legendary figure dies "live" in the networked sphere.

Methods & Data

We collected 46,390 public Facebook posts published between 22 and 30 July 2025 that mentioned "Ozzy Osbourne". We then preprocessed the texts through tokenization, lemmatization, and stopword removal, and extracted bigrams using the gensim Phrases model. We applied Latent Dirichlet Allocation (LDA) topic modeling (Coherence = 0.73) to identify dominant themes and conducted sentiment analysis with TextBlob to measure polarity on a –1 to +1 scale. Finally, we visualized temporal and emotional dynamics to trace the evolution of discourse over time.

Results
Four dominant topics emerged: Death & Legacy, Memorial & Tribute, Music & Fan Tributes, and Tribute & Farewell. Posting peaked on 23 July (13,866 entries) and declined sharply thereafter, mirroring the lifecycle of an online event. Sentiment was predominantly positive (mean polarity = 0.105), with Tribute & Farewell (50% positive) and Music & Fan Tributes (49% positive) dominating. Frequent collocations such as "black_sabbath", "prince_darkness", "rest_peace", and "farewell_concert" illustrate how users merged grief, admiration, and mythology into a shared cultural narrative.
Added Value
This study conceptualizes celebrity death as a hybrid of media spectacle and participatory ritual. By integrating computational text analysis with cultural interpretation, it shows how algorithmic publics collectively reframe loss into commemoration, blending journalism, fandom, and emotional performance. The findings extend Dayan and Katz’s framework to the digital sphere, demonstrating how networked mourning transforms moments of death into global acts of mediated memory.