Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

Session Overview
B4: Social Media and Networks
Thursday, 08/Sept/2022:
5:00pm - 6:00pm

Session Chair: Florian Keusch, University of Mannheim, Germany
Location: A 025

HTW Berlin, Campus Treskowallee, Treskowallee 8, 10318 Berlin

Show help for 'Increase or decrease the abstract text size'

Differences in Online behaviour of men and women: large N network analysis

Linda Coufalová

Faculty of Social Sciences, Charles University, Czech Republic

Relevance & Research Question

The Internet has been considered a male-dominated space for many years, but since 2009 research suggests that the proportion has been shifting. That is likely to be connected to the progressive shift of interpersonal communication to the online space. Nevertheless, there is still a lack of clarity on the difference between the behaviour of men and women. Most of the research done in this area has been conducted in niche online spaces such as technology or health online forums. My research focuses on observing gender differences in creating connections between two dominant genders in general online space.

Methods & Data

I use large N social network available at Stanford large network data collection. This directed network, collected in 2012 by scraping user profiles of Slovak biggest social media has 1 632 803 nodes and 30 622 564 arcs and therefore represents one-fifth of the Slovak population. I use the Auto-logistic actor attribute model (ALAAM) to test for gender differences, specifically using the Equilibrium Expectation algorithm to fit the model on large N data. ALAAM is developed to explain dichotomous individual attributes based on connections an individual has with others and of attributes of connected others. Therefore, it is a suitable method to evaluate hypotheses about different ways in which men and women navigate connections in social media.

Although previous research has suggested differences in the online behaviour of two genders, I have not been able to confirm these. The only observed differences were suggesting that women tend to have slightly more connections and that men tend to create connections with other men more often than women with other women. Nevertheless, both of these differences were found to be very small.

Added Value

The findings of this research are bringing new understanding into the online behaviour of people because I tested niche differences found in previous research in a general population online space. I have used the Equilibrium Expectation algorithm developed only recently, therefore I was able to test hypotheses on large N network data which is not yet common in sociology.

StakeX - Organizational Networks from Web Research

Swen Sieben, Paul Simmering

Q | Agentur für Forschung GmbH, Germany

Relevance & Research Question

Large infrastructure projects are bound to address challenges in public transportation and the energy transition. Additionally, they require support from local and influential stakeholders and their associates. To construct, for example, new subways, train- or electricity lines, decision-makers must be convinced that the construction provides a lasting benefit and takes their interests into account. The organizers of these projects need to know who those stakeholders are, who can mobilize others and how they can be best addressed in communication.

Data acquisition for Social Network Analysis is challenging and often bound to one population and data source, however, stakeholders are embedded in multiple social contexts. We introduce an online research method and network analysis approach for identifying stakeholders by capturing personal and organizational networks.

Methods & Data

Each approach starts with a small number of diverse seed stakeholders, representing different hierarchical and topical areas of the network. The network is expanded iteratively. For each stakeholder we use multiple publicly available online resources, e.g., databases, reports of companies and institutions, websites and publications, to capture professional affiliations and collaborators and add them to the network. Data entry, analyses and visualizations are performed with R and two R Shiny apps.

We present a typical regional stakeholder network that started with 140 seed stakeholders and expanded to 6,378 stakeholders with a total of 139,545 relationships, defined by memberships in 606 different organizations.


On a visual and statistical level, the resulting network reveals social groups, network segments, critical influencers and network brokers. We also show a technique to reduce visual clutter by selectively merging nodes in the dual layer network of people and organizations.

Added Value

Our approach transforms a vague target group description of influential stakeholders into a precise dataset suitable for operational and planning purposes. Our clients use analyses like this to e.g., decide on preparatory talks or inform and communicate before and during project launches. By understanding key stakeholders and their networks, they can increase a project’s efficacy, acceptance and support.

Contact and Legal Notice · Contact Address:
Privacy Statement · Conference: GOR 22
Conference Software - ConfTool Pro 2.6.145
© 2001–2022 by Dr. H. Weinreich, Hamburg, Germany