Conference Agenda

China has experienced unprecedented rural-to-urban migration over the past three decades, bringing a large number of rural children into urban environments. Peer networks are crucial for adolescent development, yet little is known about the extent of segregation between local and migrant children in friendship networks or its psychological impact. Existing theories offer conflicting predictions regarding how this segregation might influence migrant children’s well-being.

Using nationally representative data from junior high school students nested within classes and schools, this study adapts a racial segregation measure to assess the degree of separation between local and migrant students in peer networks. The findings reveal substantial segregation, particularly among boys. Migrant boys were more likely to befriend other non-local students, whereas migrant girls were similarly likely to make friends with local and non-local students, given the contextual distributions of local and non-local students. Furthermore, segregation from local peers was negatively associated with the psychological well-being of migrant boys.

This study is the first to systematically examine friendship network segregation among migrant children in China. The pronounced exclusion of migrant boys from local peer networks aligns with prior research on gendered barriers in social relationships. By highlighting the persistence of network segregation, this study contributes to a broader understanding of structural constraints in migrant children’s peer interactions. It highlights the need for targeted policies to mitigate the negative effects of segregation and promote greater social integration for migrant children.

The working conditions in nursing homes (EHPAD in French) are shaped by a combination of physical, organisational and psychosocial constraints, which are further exacerbated by the increasing dependency of elderly residents, many of whom suffer from neurodegenerative diseases (Balavoine 2022). A DREES study reveals that nearly 50% of hospital nurses leave their positions or change professions after ten years (Pora 2023), with sick leave being widespread in the healthcare sector, particularly among cleaning and orderly staff (Pollak et Ricroch 2017). Although solidarity mechanisms have made it possible to cope with the workload, the deterioration in working conditions also seems to have reduced opportunities to create bonds and support between colleagues (DREES 2016). The study utilises social network analysis to explore the relationship between interpersonal networks and the distress experienced by healthcare professionals in nursing homes. The investigation encompasses multiplex relationships: professional ties (colleagues, residents and their families) and personal networks, (including family members and friends). The research methodology involves the administration of field surveys, incorporating participant observations, the utilisation of questionnaires, and in-depth interviews with healthcare professionals in the Occitanie region of France. The present study has been designed to facilitate an understanding of how caregivers perceive and define work-related hardship, the role of social networks in either alleviating or intensifying these challenges, and the impact of workplace conditions on relational dynamics. Preliminary findings highlight the importance of solidarity networks, often centred around friendly groups, and reveal the role of residents in assisting caregivers.

Background: Individuals never work alone – especially not in the healthcare system, which is strongly characterized by multidisciplinary care. Patient-centred care requires appropriate support and guidance from healthcare providers as well as their cooperation and collaboration. This includes access to relevant services and the coordination of treatment and counselling appointments. Particularly in emerging areas of care, like exercise therapy for oncology patients, both maintaining existing relationships and expanding new network structures are essential. To ensure comprehensive access to exercise therapy for all cancer patients, a nationwide care structure is required. For this reason, a structured care pathway is currently being implemented and evaluated at seven CCCs in Germany.

Methods: Through semi-structured interviews and supplementary egocentric network maps, exercise experts from the seven CCCs were surveyed. As part of a context analysis, this study aims to analyse the relevant relationships and their influence on the activation of social capital in building care networks.

Results: Data collection was completed at the end of 2024, allowing the first results to be prepared in time for the conference. Initial findings indicate that exercise experts bring existing networks into their current roles. Experience from previous projects and shared facilities structurally contribute to the expansion of exercise therapy care pathways. The unregulated care structure in exercise therapy particularly relies on informal ties and collaboration, which are heavily influenced by uncertainty due to limited financial support and a general lack of awareness in practice.

Health beliefs related to vaccines, abortion, pharmacological treatments, and even conspiracy ideas do not exist in isolation—they are interconnected and form complex, dynamic systems shaped by social, cultural, and political factors. These belief systems not only influence individual health decisions but also impact public health outcomes and policy debates. Understanding their structure and interactions is crucial for addressing the growing influence of social and political polarization on public health attitudes.

This study examines the impact of polarization on health belief systems in the Spanish population using data from a nationally representative survey. Responses from 2,200 individuals were collected, assessing health beliefs through the COVID-19 Misinformation Scale (CMS12), along with additional measures of trust in institutions, ideological positioning, and exposure to misinformation. By leveraging mixed graphical models, we identify underlying structures and interconnections within these belief networks, exploring how different dimensions of polarization reinforce or challenge specific health narratives.

Our preliminary results indicate that belief systems among antagonistic groups are highly differentiated, yet common connections also emerge, helping to explain intermediate positions on specific health topics, such as vaccine hesitancy. By unveiling these complex interrelations, this research contributes to a deeper understanding of how social divisions shape health-related perceptions and behaviors. The findings provide valuable insights into the role of misinformation, trust, and ideological influences on health attitudes, offering implications for public health communication strategies and evidence-based policymaking.

The COVID-19 pandemic triggered significant lifestyle changes for many adolescents, including reduced in-person interactions, increased social media use, and heightened feelings of social isolation. Rural adolescents, who were already more likely to experience geographic and digital isolation, were among those most affected by the quarantine. While some prior research has focused on negative feelings as an impact of the pandemic, we focus on changes in the composition and structure of core networks. This study draws on a sample of rural students who had recently been accepted to a state university. The survey was administered to two separate cohorts, the first sample was drawn in 2018 (n=288), and the second in 2023 (n=248). A variation of the important matters name generator was used to elicit up to five unique names across five different support dimensions. The data were analyzed using multi-level modeling to account for both network and alter-level variables. We found that there were no significant differences between 2018 and 2023 in total support received or the number of close ties. However, in 2023 adolescents reported smaller support networks with higher bridging potential. The frequency of media use was not associated with social support, however, media multiplexity with alter, particularly social media use, was associated with feeling closer to core network members. These findings suggest that adolescent core networks remained resilient over a major societal event. It also highlights some of the evolving changes to the composition and structure of networks due to new media use.

Network creation games serve as valuable models for understanding the decentralized formation of economic and social networks, where self-interested agents aim to balance maximizing their connectivity while minimizing the number of links they maintain. Traditional models often simplistically assume that agents have complete visibility of the entire network or/and act synchronously, assumptions that are particularly unrealistic in real-world interactions. To address this limitation, our study embraces those constraints and proposes an Agent-Based computational model where individuals act asynchronously based on a partial view of the network. Despite these constraints, our model consistently converges towards the most efficient network structure. While a revised best response model achieves comparable structural results, it fails to replicate the temporal dynamics observed in a past empirical study, such as the frequency of actions. Our approach addresses this by assigning a utility to the action of 'inaction' dependent on the current network state, and by including an exploration bias. Our model's ability to capture complex network dynamics from past experiments suggests that seemingly complex behaviors in such games can emerge from decentralized, homogenous, payoff-driven agents.

This poster will present the methodological aspects of conducting an analysis of social networks within an Eco-citizen Observatory (Observatoire éco-citoyen in French) for monitoring environmental pollution in Senegal. The latter is the result of a collaboration between researchers, associations and residents and is based on local citizen mobilizations against the installation of iron and lead recycling plants, which cause nuisances and health concerns in a small city located in the suburbs of Dakar.

The data will be collected mainly using ethnographic methods in order to build up a whole network of the Eco-citizen Observatory and identify the relational chains and resource transfers (information, knowledge, skills, material or immaterial goods, etc.) present in this network. The methods used will serve to analyze the conditions that led to the creation of these chains of relationships, and in particular to examine the importance of sociability in the building of trust and the formation of the various groups involved. The role of new communication technologies, and more specifically the impact of online social networks, will also be taken into account in the development of this citizen participation.

Ultimately, the objective of this research, which is being carried out as part of a PhD in sociology, is to analyze how a biomonitoring experiment involving various publics not only transforms knowledge about but also creates opportunities for more citizens to mobilize to address the impacts of this pollution.

This study analyzes the investment relationships between startups and institutional investors in Japan. To examine how the investment network forms, we investigate its structural properties, focusing on two key statistical measures: the degree distribution and preferential attachment. Our findings indicate that the network follows a power-law degree distribution with an exponent of approximately –2.9 and exhibits a sub-linear preferential attachment rate with an exponent of 0.65. These results suggest that while highly connected startups are more likely to receive new investments, the effect is weaker than in classical preferential attachment models.

It is well established that previously formed personal and collective networks significantly influence an individual’s socio-economic standing. Several women's self-help groups have arisen from members’ prior interactions in community neighbourhoods or active and regular participation in socio-economic matters. Despite the available evidence, there is a lack of research regarding the impact of husbands' networks on the establishment of their spouses' self-help groups. The paper explores how the formation and sustainability of women's self-help groups in the Iringa region of Tanzania have been influenced by their spouses’ work networks. A qualitative method was employed to collect and analyse data that were derived from Focus Group Discussions (FGD) with groups identified through the snowball sampling technique. The findings indicated that the men's working relationships may have a positive impact on the formation, effectiveness and sustainability of their spouses' self-help groups. Deep-level similarities of duties in the workplaces facilitated strong social ties that in turn encouraged couples to create groups aimed at improving their economic conditions and enhancing positive social networks. Apart from improving members’ bonding social capital manifested in supporting each other in various social matters, groups also provided quick and affordable access to loans with low interest rates that facilitated women's economic empowerment.

This poster proposal is based on a sociological study devoted to the analysis and evaluation of a Cité Educative located in the priority neighbourhoods of a medium-sized town in Hauts-de-France. This research is funded through a contractual agreement with the University of Artois and the SHERPAS Laboratory.

The Cités Éducatives initiative aims to significantly reshape the relationship with educational practices by fostering collaboration and synergy between schools, the local ecosystem, and municipal services. These initiatives are led by a group of stakeholders from three main branches: representatives of the state, the municipality, and the national education system. Our study focuses on this small group of political, institutional, and operational decision-makers.

We have identified several networks that allow us to analyze two key variables: collaboration among actors and advisory relationships between them. These are small-scale networks (nodes = 16), for which we conducted various measurements, including weighted centrality, spectral geometry via Laplacian spectrum analysis, and community detection using Louvain clustering. Each of these networks has been analyzed comparatively over two years to assess their evolution. The statistical analysis was carried out using ‘R’.

We will also present the main findings of this study, highlighting potential hierarchical interferences and structural imbalances within the network that weaken both the implementation and administrative management of the Cités Éducatives initiative.

This study explores the long-term impacts of COVID pandemic on older adults’ social networks in the United States. In particular, I delve into the patterns of different types of social connections (e.g., in-person connections versus online connections) during and after the pandemic and the differential effects of those interactions on physical and mental health. Though previous literature discussed how the pandemic re-shaped our social networks in various ways as well as the peer effects on health, the mechanisms through which the pandemic exacerbate health inequalities through network effects have remained unexplored. Using data collected by National Social Life, Health & Aging Project (NSHAP), a pioneering nationally representative study of the intersection between social and intimate relationships and healthy aging, this study expands on this line of discourses by examining the relationship between older adults’ network structure and characteristics during the pandemic and their physical and mental health; the ways in which different dimensions of social network affect health on a different scale; the extent to which these patterns differ by people’s socio-demographic factors including age, gender, race and ethnicity, and socio-economic status; and the long term effects of COVID-19 pandemic on the structural disparities in health. Limitations and implications are discussed.

Introduction: Global elimination of the hepatitis C virus (HCV) is possible due to curative direct acting antivirals (DAAs). Curbing the spread of HCV through DAA treatment is essential to elimination plans, but many people who use drugs (PWUD) remain unaware of DAAs. This analysis examines diffusion of information about DAA treatment for HCV through the ego networks of rural PWUD enrolled in the Kentucky Viral Hepatitis Treatment (KeY Treat) project, an HCV DAA treatment trial. Methods: A name generating questionnaire elicited first name, last initial of drug, sex, and support egocentric networks. Participants (n=306) named network member(s) they told about the study and their relationship to the participant. Multivariable logistic regression was utilized to determine factors associated with diffusion of HCV treatment information. Results: Participants told an average of 1.42 network members about the study (range 0-7). Most (75%) of those told of KeY Treat were friends, followed by family (25%) and, according to alters, 15% (n=66) of egos enrolled in KeY Treat after being informed of the treatment trial. Treatment information diffusion was significantly more common among women (adjusted Odds Ratio [aOR]: 1.89, 95% Confidence Interval [CI]: 1.05, 3.41) and participants receiving buprenorphine treatment for opioid use disorder (aOR: 1.81, 95% CI: 1.01, 3.26). Conclusion: Results underscore the importance of networks in diffusion of interventions. Networks are vitally important in the spread of HCV but can and should be leveraged to encourage engagement in curative HCV treatment to achieve worldwide elimination the virus.

Opinion dynamics models are a versatile tool for investigating how a network of politically affiliated agents form perceptions collectively about media bias under exogenous (independent analysis of media outputs) and endogenous influences (peer pressure). Previous numerical studies of these models show that persuadable agents in politically allied networks are disrupted from asymptotically learning (learning in the long run) the true bias of a media organization, when the network is populated by one or more obdurate agents (partisans), who are not persuadable themselves but exert peer pressure on other agents. Partisan disruption occurs in two ways: agents asymptotically learn a false bias, or agents' opinions never settle and vacillate indefinitely between belief in a false bias and the true bias, a phenomenon called turbulent nonconvergence. In this paper, we derive (and validate with Monte Carlo simulations) an analytic instability condition that distinguishes these two modes of partisan disruption, in terms of the learning rate and key network properties, for an idealized model of media bias featuring a biased coin. We interpret the condition as expressing a balance between the exogenous influence of the media organization’s published outputs, and the endogenous influence of

the partisans. We explore the partisan influence as a function of network size, sparsity, and partisan fraction and find that the network is less likely to experience turbulent nonconvergence as the learning rate increases, size decreases, sparsity increases, and partisan fraction increases. These results and their social implications are interpreted briefly in terms of the social science theory of structural balance.

From the mid-1970’s to the late-1980’s, French energy policymakers implemented a massive nuclear power scale-up which resulted in an industrial, economic and political predominance of this technology in the country. Both political discourse and scholarly accounts often argue that this unhampered policy was made possible by the social proximity of bureaucratic and industrial elites. Such individuals are said to have dominated the energy policy-making process for years and are therefore referred to as “nucleocrats”. Based on an original database developed within a larger multi-method project, we assess the validity of this broad view, as well as the precise empirical claims it implies, using social network analysis. Data consists of individual-to-institution affiliations of over 200 individuals pre-identified, through detailed historical work, as core members of an administrative elite made of government advisors and implementors related to the nuclear energy program from 1974 to 1984. Questions to be answered from this database include: What was the extent of interlocking ties within this network and how did they evolve during industrial implementation? Did the social network patterns change over the 1970’s decade under the right-wing government? How did the 1981 swing to a socialist government, along with subsequent bureaucratic appointments, reshape the nucleocrat’s network? As this research is ongoing, the poster will present the method (case selection, individuals’ selection, data collection), hypotheses as well as preliminary results drawn from exploratory analyses.

Mackems, inhabitants of Sunderland in the Northeast of England (NEG), perceive their dialect to be different from that of Geordies, who live in Newcastle, another NEG city 20 miles apart (Burbano-Elizondo 2008). NEG morphosyntactic dialect features (DF) are perceptually salient around Newcastle specifically (Childs et al. 2020). Are these perceptual differences the result of actual linguistic borders within the dialect continuum?

Building on Dodsworth and Benton’s (2017, 2020) recent introduction of bipartite networks to sociolinguistics, I suggest that it is possible to find out by looking at how NEG DFs circulates within geographically localised networks of speakers. The bipartite network is inferred from dialect corpora (DECTE, Corrigan et al. 2012) and links 88 speakers to 52 places where they live, have lived, studied or commuted, as proxies for the social ecologies influencing the regionalised nature of their speech. The network is analysed through Exponential Random Graph Models, new to sociolinguistics.

Results suggest that NEG DFs users do not cluster anywhere in the network. However, there is homophily between users of nonregional DF aye, net of other factors. They are mainly linked to urban centres and rarely associated with rural locations, contrary to the expectation that rural speakers are more conservative. Overall, more central cities with salient dialect-based identities seem to be key sections of the network regarding this declining DF’s persistence, but Mackems and Geordies do not have morphosyntactically different dialects.

A stylized fact in modern economies is the more developed a country is, the greater the weight of the service sector. In this sense, the study of economic complexity through the standar measure of complexity index presents an increasingly relevant omission to understand the economic process and its growth. This paper proposes a new methodology to retrieve information on economic complexity in services. For this purpose, the US input-output matrix is used. This work is novel because, thanks to the structure of the data as a network, it is possible to infer the missing information of complexity of services at a level of disaggregation that is strikingly higher than in other works. Using the k-NN method is possible to learn 146 services sectors complexity index. The index recuperated by this method are consistent with previous works and this index is highly correlated with the GDP of States and US economy.

The relationship between the improvement rate and patent citation network structure of a technological domain represents a dynamic network model. Characteristics of the dynamic patent citation network determine a constant in the technological improvement rate. These dynamic network models can be seen as survival models, with the hazard rate in these models being the probability for a patent to be cited by a new patent given the time since the last citation. We name these ’knowledge production’-models since they capture the rate of output (new patent) given the input (existing patents).

Some important economic implications follow from these models. As an illustration, we analyze all patents on the evolving technological frontiers associated to inventions in hydrogen technologies. Our analysis reveals that of the four key technology subdomains, ’production’, ’storage’, ’distribution’, and ’fuel-cells’, patents in ’distribution’ exhibit the lowest rate of knowledge production. From this we deduce that ’distribution’ related costs will be a key constraint to the emergence of a hydrogen market for retail consumers. Other results reveal that knowledge acquired by exploitative learning (within subdomain citation) suffer higher (opportunity) costs of patenting. In fact, this could offer an explanation for the slower development of hydrogen ’distribution’. As the investments needed in this subdomain are high there is less incentive to share and more incentive to monopolize new knowledge.

As the effects of climate change accelerate, shifts towards more sustainable consumption patterns is needed. However, in France, while a large majority of individuals have pro-environmental attitudes and 82 % of them say they would be willing to change their way of life under certain conditions, only 23 % declare they have adopted more sustainable habits. Indeed, the gap between attitudes and behaviours towards sustainable consumption practices is still largely unexplained. Moreover, state-of-the-art statistical models explain no more than 20 % of the variance in behaviour, and there is a lack of understanding of the dynamics of sustainable consumption patterns. We hypothesise that social networks are a missing variable of these models. Indeed, previous work has shown that social networks, through peer pressure and influence, can explain behavioural changes such as reducing one’s electric consumption. However, these studies lack long-term empirical data, they do not take into account the cognition of individuals, and the generative mechanisms are not properly described.

In light of preliminary statistical results using data from periodic national social surveys on the environment, this study aims to explore methodological means (e.g. statistical analysis, experiments, social network analysis, agent-based modelling) to collect empirical data on the evolution of sustainable behaviours and social networks, to model cognitively complex agents, in order to investigate whether social networks have a significant effect on shifts towards more sustainable consumption and sufficiency practices.

Background: Capacity-building is important for increasing the quality of cancer research. The Winship Cancer Institute’s shared resources support the cancer research efforts of investigators across Emory University by providing them with subsidized services. The Intervention Development, Dissemination, and Implementation (IDDI) Shared Resource provides members with expertise in behavioral science research methods and supports them in developing, testing, or disseminating behavioral and systems interventions to prevent cancer, detect cancer early, and/or improve survivorship. The aim of this project is to conduct an ego network analysis of IDDI to examine its connections with researchers and understand how IDDI supports the cancer research infrastructure at Emory.

Methods: We obtained data from Winship’s tracking system, PPMS. We analyzed data from June 2020 until December 2024. We compiled the data into Excel and utilized UCINET and NetDraw for all network analysis and SPSS for all descriptive statistics.

Findings and Implications: We had 92 unique projects from 50 users (45.7% repeat users). Users of IDDI were predominantly from the School of Medicine (57.6%) and School of Public Health (35.9%). We will examine unconnected departments within each school to identify isolated departments in our network and therefore gaps in outreach, such as faculty who may be unaware of our resources and potential for collaboration. We also explored other attributes of our users, such as early career professors versus tenured professors, to examine our impact on the cancer research pipeline. These types of analyses will show trends in our capacity-building efforts and recommended areas for outreach.

Objective: To characterize and compare the ego-network structures of individuals by N. gonorrhoeae and C. trachomatis (NG/CT) status, identifying network patterns that may contribute to the transmission dynamics of these infections and inform targeted public health interventions.

Methods: From May 2023 to June 2024, individuals completed a survey on demographic, clinical, behavioral, and social network characteristics and provided biological samples for NG/CT diagnosis if symptomatic or exposed. Social network analysis metrics were calculated, including degree (number of connections), network density, and the proportion of alters with specific characteristics relative to the ego. Comparisons by infection status were conducted using the Mann-Whitney U test.

Results: 107 ego-networks containing 187 individuals were analyzed. The median degree was 1 and median density 1.83. The median ties between named friends was 3, and ¼ of the networks reported having sex with people outside of Clark County in the previous 6 months. Networks of individuals diagnosed with NG/CT had a greater proportions of males p= 0.05), showed greater homophily by gender (p= 0.02) and had a greater proportions of named friends diagnosed with NG/CT (p<0.001). No differences were observed in tie strength or drug use.

Conclusion: Ego-networks of individuals with NG/CT exhibited greater gender homophily and higher NG/CT prevalence among named friends, suggesting that network structures influence transmission. Network-based patterns could inform targeted STI prevention strategies.

The recall of the Loudoun County School Board, prompted by debates over the inclusion of Critical Race Theory (CRT) in educational curricula, has spurred extensive discourse on Twitter (now X). Although previous research has highlighted that tweeting behaviors—such as topic adoption and sentiment expression—are significantly influenced by interactions within networks, and figures like Ian Prior have been noted as key influencers, there remains a lack of quantitative assessment of the impact these political entrepreneurs have on public conversations.

This study will this gap by utilizing egocentric network analysis to quantify the influence of political entrepreneurs on their followers, with a particular focus on topic engagement and sentiment adoption. Moreover, this research will examine the role of bots within these dynamics, assessing the likelihood that both influencers and their followers are bots. We collected and analyzed tweets related to the school board recall from March 1, 2021, to May 31, 2022, focusing on three prominent topics: the school board recall, CRT, and transgender student rights (TSR). Using a supervised machine learning model, these tweets were categorized based on their relevance to each topic and the sentiments expressed, thereby laying the groundwork for comprehensive network analysis. By delving into the mechanisms of influence within these networks, this research aims to provide deeper insights into how political entrepreneurs and bots may shape discussions about CRT on Twitter, thereby enriching our understanding of digital discourse and networked influence.

Inclusive education is vital for the rights of all children and teacher aides (TAs) are key players in these efforts. However, aspirations of inclusive education policies, like others, often fail to be realized. As such, two key purposes guide this work: 1) explore the relational space between TAs, teachers, and school leaders within a Community of Learning in the context of a national education policy that highlights the importance of relationships and knowledge sharing; and, 2) extend current research on inclusive education that takes a social network perspective to include an examination of social networks at a systems level. Drawing on social capital theory and social network analysis, we investigate the extent to which TAs and teachers can access and leverage each other’s knowledge, expertise, and support using data from a survey of 561 educators across two Communities of Learning in New Zealand. We found that knowledge sharing among TAs and between TAs and teachers is constrained by weak relational patterns. While TAs are accessible to others from a network perspective, people do not access them. They are rarely considered valued sources of knowledge and expertise or identified as collaborators. In most schools, like disabled students, TAs are on the periphery of the network. Ambitious and well-intentioned inclusive education policies are not working as intended. Educational leadership has a key role in improving the relational space amongst TAs and teachers, and creating classroom, school, and system conditions that support and sustain inclusive education.

Urban infrastructure facilitates connectivity within cities. However, some elements, such as urban highways, are massive, concrete constructions that can become a barrier to walkable, accessible and human-scale neighborhoods. Between 2022 and 2024, the Reconnecting Communities Pilot (RCP) Program and Reconnecting Communities and Neighborhood (RCN) Grant in the US awarded approximately four billion dollars to “carry out a project to remove, retrofit, mitigate, or replace an existing eligible facility with a new facility that reconnects communities”. However, there are few existing quantitative tools to evaluate the applications and awarded grants according to their impact on social connections. Here, we aim to offer an easy-to-use toolset to quantify the social connectivity impact of barriers identified in 52 highway redesign or capital construction projects funded by the RCP and RCN. To do so, we overlay the project sites with spatial social networks constructed from Twitter mutual followership data in the 25 largest U.S. metro areas. We calculate the Barrier Score for each site which compares actual social networks to randomized social networks that do not account for the highway infrastructure. This enables us to assess whether these sites exhibit reduced connectivity compared to a randomized baseline. We demonstrate that some projects show strong potential to enhance social connectivity, while others appear to have limited impact according to our network-based evaluation. This approach represents a first step toward a scalable, social network data-based toolset for policymakers assessing the social impact of urban infrastructure projects.

Recent work has placed emphasis on how people frequently avoid their strong ties on important issues (Small 2017, Small et al. 2024). This planned study aims to explore how young adults in Hong Kong find confidants and moments of *Gemeinschaft* in the context of intense political polarization. The focus lies on the question how individuals regulate their relationships when political divisions cut across their personal networks. Inspired by the Caen Panel Study, the methodology of the study aims to capture an extensive range of alters that allows to measure the structure of ego-networks, additionally to the content of relationships. Initial results of a pilot study with young Mainland Chinese students in Hong Kong indicate that many young adults deliberately avoid many of their strong ties regarding their most pressing life problems, especially family members. At the same time, diverse circles of friendships provide an important source of community and mutual emotional support.

While network scholars are interested in understanding whom people turn to for support during challenging times, limited research has focused on core support networks in total institutions, such as prisons. This study uses egocentric network data from 379 people who are incarcerated, encompassing 1368 network members, to answer: 1) What are the significant correlates of having an incarcerated person in your core support network? 2) How do ties with other incarcerated persons differ from ties with community-dwelling persons in the core support network? About 20% of participants include at least one fellow incarcerated person in their core support network (average network size=3.61). In the ego-level logistic regression, men are less likely (OR=0.216, p<0.001) and those with longer sentences are more likely (OR=1.006, p<0.05) to include a fellow incarcerated person in their network. Also, having a lower proportion of ties with prior incarceration experience (OR=0.112, p<0.001), a lower proportion of kinship ties (OR=0.235, p<0.01), and less trust with network members (OR=0.551, p<0.01) are related to higher odds of naming an incarcerated person. In the tie-level multilevel linear analysis, participants are less close to and have less trust in fellow incarcerated persons compared to community-dwelling ties, but there are no significant differences in emotional or advice support. Findings confirm the contact opportunities theory, unmet social needs theory, and experiential homophily. By examining the dynamics of core networks in prisons, this study contributes to a deeper understanding of resilience, trust formation, and the critical role of “weak” ties in navigating high-stress, constrained environments.

While past research has underlined the potent influence of peers on various behaviors in childhood, less is known about the continued influence into adulthood. Emerging adulthood is a transitional period in which people navigate through an increasingly complex social world, while taking on an increasing number of adult roles and responsibilities. To successfully navigate through these times and adjust well, self-regulation is an important skill, heavily shaped by one’s social environment. With our project, we want to fill this gap by extending the notion of homophily in human social networks to neural networks.

This longitudinal, multidisciplinary study uses a sample of a peer network of emerging adults that are part of a large Dutch student organization. Through group nominations that are collected every few months, selection and influence processes will be studied through social network analysis. Network data will be combined with multiple waves of neuroimaging data collected from a subsample. During the scanning session, participant engage in behavioural tasks tapping into self-regulation and social reward sensitivity. Questions that will be investigated with this project are: Do people closer within a social network behave more similarly to their friend early on or does it change over time? If so, does this trend extend to a neural level by people closer within a network processing social information more similarly compared to people less close? The intention of this research poser is to share this novel project with fellow researchers in the field and to stimulate discussions and potential collaborations.

Trust is essential to the social cohesion of societies. It affects political stability it has been linked to better economic outcomes of societies, and, without trust, no cooperation can emerge. While it has been established that trust is rooted in interpersonal relationships, the interplay with the spatial/geographical context and its role in the creation of trust is much less clear.

Although multiple studies have looked at the role of neighbourhood diversity on trust, they show conflicting results and reduce the spatial dimension of social networks into a pre-defined neighbourhood that is handled uniformly and fail to consider the level of urbanity. Also, many studies do not actually contain data on social networks but only contact opportunities in the neighbourhood. Thus, these studies neglect important dimensions in the relationship between geographic context, social networks, and trust. Furthermore, with the relationships in previous studies only being simplistically embedded in the geographic context, the real potential of mobile actors being in contact with others outside their neighbourhood is ignored and the analysis is subject to ecological fallacies derived from the drawing of neighbourhood boundaries.

In this study, we analyse a unique data set of ego networks in Switzerland, including information on ego’s residence location, the area where they perform their everyday activities, as well as the locations where contact with the interaction partners usually takes place. Drawing on this data, our analyses contribute novel evidence towards a more comprehensive understanding of the interactions of geographical space and social networks in the making of trust.

This paper aims to determine the hydromorphological changes of selected meander sectors of the largest river in the Republic of North Macedonia, the Vardar River using GIS. The meandering process is a significant dynamic in the hydrological studies of river systems. Covering a period of 59 years (1964-2023), in the study of the meandering process of selected three meander sectors in river Vardar, topographic maps with a scale of 1:50000, geological maps with a scale of 1:100000 and satellite images from the Landsat mission (1983-2013) and the Sentinel-2 mission (2018-2023) were used, which resulted in their processing and analysis. The planimetric characteristics that were analyzed for meandering are the width of the river channel, sinusoidality, radius of curvature, width of the meander section, slope and migration of the riverbed. In the period between 1964 and 2023, all three meander sectors have experienced significant changes in all parameters, with large changes in the width and length of the meander sectors, accompanied with significant lateral erosion on both coastal sides consistently. The hydromorphological changes that have occurred also differ in the different time periods of action, where each period is characterized by certain spatial changes. Understanding the hydromorphological changes in the meandering process has a significant role in predicting future changes in the flow of the riverbed in order to reduce and possibly prevent future potential impacts on the space.

Geotagged social media data offer a valuable tool for identifying nature tourism hotspots and monitoring human-wildlife interactions. Previous studies have demonstrated the utility of hashtags in detecting popular tourist areas linked to wildlife, such as seal haul-out sites, both within and beyond designated Special Areas of Conservation (Mancini et al., 2018). Similarly, data mining techniques have been used to monitor recreational activities like fishing, providing insights into fisher behaviours and fish populations (Monkman et al., 2018). Integrating social media analytics with ecological monitoring presents a novel approach for assessing ecotourism’s impact on charismatic megafauna, such as seals and seal tourism.

This study analyses georeferenced Instagram© posts from 2014 to 2023 within the Ythan Estuary catchment, Aberdeenshire. The area lies within the Forvie National Nature Reserve, a protected and ecologically significant coastal site in Scotland, adjacent to Newburgh Seal Beach, a popular recreational area. Using hashtag analysis and data mining techniques, we examine whether visitor motivations, personal significance, and proximity to the site can be quantified from social media activity. Given the estuary’s history of human recreation, its growing seal population, and its designation as a critical seal habitat, this location provides a compelling case study for evaluating the sustainability of seal tourism.

We specifically extracted posts referencing seals to assess their relationship with visitor motivations and proximity patterns. Temporal trends in posting behaviour were analysed to determine the impact of the 2017 seal haul-out protection measures and the 2020 COVID-19 pandemic on visitor activity and social media engagement.

This poster will discuss my PhD research plan of bridging the theoretical gap between network and identity in the collective action research. I will focus on a case study of an ongoing panethnic social movement in the UK and adopt a relational framework to demonstrate how networks shape identity construction and movement mobilisation.

The COVID-19 catalysed a surge in racial discrimination against individuals with “Chinese-looking” features globally, leading to the emergence of an “East and Southeast Asian (ESEA)” social movement in the UK which witnessed the formation of the ESEA identity and ESEA organisations as the result of panethnic solidarity. While existing literature primarily interprets the panethnic movement through an identity politics lens, this research will draw on Harrison White’s perspective of conceptualising identity as a product of social formation embedded in the network system, and bring up the question of how panethnic identities are formed, negotiated, and shaped by the overlapping networks and social relations from pre-pandemic to post-pandemic time.

Methodologically, I plan to apply a mixed-method Social Network Analysis (SNA) approach. This includes collecting interorganisational network data through surveys, and adopting qualitative method of participatory observation, interview, and focus group for identity-related discussion. At this early stage, I seek feedback on the theoretical and methodological application of a relational perspective to networked activism and identity formation.

While much research has focused on between-school racial and socioeconomic segregation in the United States, research has also documented how that within-school racial and socioeconomic segregation contributes to broader social network segregation. Drawing on the under-utilized High School and Beyond dataset, in this study I explore school-level factors that shape within-school social network segregation. My main method innovates on traditional network analysis and observational techniques by extending conventional methodology to account for missingness in network data. This study contributes to the literature by providing new insights into how institutional structures and social dynamics within schools shape patterns of segregation in student networks.

Cette étude examine la manière dont les militants du Front des Forces Socialistes (FFS) en Algérie utilisent la plateforme Facebook pour construire leur légitimité et leur capital social dans un contexte politique tendu. Les militants s’appuient sur des stratégies numériques variées, oscillant entre une approche institutionnelle et la personnalisation des contenus, afin de répondre aux attentes parfois contradictoires du parti et du public. Leur participation s’inscrit dans un environnement façonné par les algorithmes numériques, lesquels créent ce que l’on appelle des « bulles de filtres », limitant ainsi leur exposition à des perspectives divergentes. Cela peut engendrer une compréhension erronée des orientations de l’opinion publique dans une période donnée. Dans ce cadre, la gestion de l’identité numérique apparaît comme un processus central dans la participation politique sur les réseaux sociaux, les militants naviguant entre divers rôles pour maintenir leur influence et renforcer leur légitimité.

For youth and young adults, social networks of peers and family influence decision making and present an opportunity to promote pre-exposure prophylaxis (PrEP). PrEP is available as an oral tablet or long acting injection and is an effective means of HIV prevention for individuals vulnerable to HIV. In the United States, HIV disproportionately impacts young Latino sexual minority men (SMM), in part due to low uptake of PrEP. Social networks are known to impact PrEP use overall, though research is needed on the networks of young Latino SMM which are shaped by unique cultural and developmental factors (e.g., immigration or migrant experiences, acculturation, family structure, language). In this study, we conduct ego-centric surveys to understand social network factors which impact PrEP use among young Latino SMM. We will analyze how network characteristics (e.g., homophily, proportion LEP, at least one member who would not disapprove of PrEP, parental figure in network) and structure (e.g., centrality, bridging) relate to the outcome of PrEP use. We will use these results to identify who within these networks might be well suited to become peer leaders in future community-engaged PrEP interventions. Data collection is ongoing.

The specific focus of the intended study is on the life management and social relationships of cloudworkers. The special aspect of cloudworking is that the work is carried out in a so-called cloud, a digital workplace, and is mediated via online platforms. As a result, the boundary between work and private life is often blurred, as there is no specific firm or predetermined workplace. Cloudworkers are usually self-employed and not tied to any company or fixed organization. The aim of the study is to find out how cloudworking influences the life management of employees and their social relationships . Using a mixed method design, problem-centered interviews with employees are used to survey existing networks. “Network maps” are used to record the social relationships in the various areas of life.

The poster serves to present the methodological approach and is open for discussion.

U.S. senior centers (SCs) focus on social engagement to reduce older adults’ social isolation and loneliness. During the COVID-19 pandemic, 96% of U.S. SCs closed to in-person programming, leaving members at risk for increased social isolation and loneliness. Based on theory that purports that chronic loneliness leads to avoidance of social connections, we hypothesized that members experiencing loneliness after the COVID-19 pandemic would be less likely to have companionships at a newly re-opened SC.

Using data collected 09-12/2021, we modeled the likelihood of companionship ties among SC members using an ERGM and found that loneliness (Θ = -0.69, p = .036) and high social support (Θ = -0.58, p = .014) were negatively associated with the likelihood of a companionship tie between SC members. Members with high depressive symptoms were more likely to have companionships (Θ = 0.57, p = .040) Transitivity was positively associated with the likelihood of a companionship tie. Demographic factors, including homophily on age, gender, marital status, and living alone, were not significantly associated the likelihood of companionship ties.

Companionships among SC members may help reduce loneliness but may not lead to increased social support. It is possible that older adults with high levels of social support are less likely to seek out companionships with other members because they already have beneficial relationships in other settings. Members with depression may find support for fostering and maintaining companionships in the programming offered at the senior center.

Purpose: Networks can be created from language tasks, including verbal fluency where participants name as many words as possible (e.g., animals) in one minute. Nodes correspond to commonly produced words across participants, and edges are placed between words that co-occur within a specified distance (i.e., window size). People produce words in groups called clusters (e.g., fish, pets). The methodological problem we investigate here is understanding which window size should be chosen for networks from a new morphological fluency task. Following previous research, the window size capturing the cluster size’s mean and standard deviation should be used.

Methods: Thirty-four participants (aged 20-62) were given 24 Dutch words (e.g., instrument) and asked to produce as many morphologically related words as possible (e.g., instruments, instrumental), one minute per cue. Responses were classified according to morphological type, based on which clusters were defined. The overall cluster size’s mean and standard deviation were 2.61 and 0.69. Therefore, we compared modularity (Q), clustering coefficient (CC), and numbers of nodes (N) and edges (E) when calculated for window size 3 (overall) versus each network’s window size, ranging 2-6 (tailored).

Contributions: Generally, overall and tailored window sizes produced similar networks, without statistical differences in Q, N, and E. For CC, a significant interaction between tailored window size and the window used for calculation (tailored versus overall) revealed that the former did not affect CC obtained with overall window size. However, tailored window size affected Q, N, and E regardless of the window used, thus requiring careful consideration.

This poster proposition aims to contribute to the analysis of professional and residential trajectories through the analysis of social relations in times of crisis, thanks to my thesis work. It focuses on the mobility of individuals who have made a residential bifurcation, from urban to low-density rural areas, towards self-employed status, in France.

The macro context, marked by major health, social, economic, environmental and housing crises, reinforces uncertainties about the future, which are likely to modify households' professional and residential projects. Initial work on the effects of the Covid-19 crisis has shown that this context may have led households to reconsider their housing conditions in the city by moving to less densely populated areas. In addition, research into the sociology of work has shown that some people make voluntary career changes for a variety of reasons, whether they are chosen or not.

Using a social network analysis approach, the aim of my thesis will be to understand how professional and residential mobilities modify the personal networks of people, and, conversely, the importance of relational circles in the construction of projects and concrete actions put in place for these bifurcations, mixed with uncertainty as well as irreversibility.

Using individual interviews and mixed methods, the goal is to better understand how these new ways of settling are implemented through various concrete projects, and what they do to sociabilities. This poster will present my methodology and the first results of my field surveys, focusing on the evolution of networks after these bifurcations.

This study investigates gender disparities in the field of communication through a comprehensive analysis of citation metrics, authorship patterns, team composition, and faculty salaries. We explore how gender influences research impact, collaboration dynamics, and career advancement within the field. Using a dataset of 62,359 papers from 121 communication journals, we examine trends in authorship and citations over time, revealing persistent gender-based differences. While female authors are increasingly represented in communication research, their work tends to receive fewer citations, particularly in sole-authored papers, compared to those written by men. However, this citation gap appears to narrow in larger and more diverse research teams.

Our findings also highlight a notable pattern of gender homophily in team formation, with single-gender research teams being more common than mixed-gender teams. This tendency raises important questions about collaboration networks and the inclusivity of research environments. Analyzing the top 10 communication journals, we observe a persistent underrepresentation of women and a citation advantage for male authors, particularly in high-impact journals. These disparities suggest structural challenges that may hinder women's visibility and influence in the field.

Additionally, faculty salary data from leading U.S. public universities reveal gender-based pay disparities, particularly at the Assistant Professor level, though these gaps decrease at higher academic ranks. Our study underscores the need for targeted efforts to promote gender equity in communication research, emphasizing inclusive collaboration, fair citation practices, and equitable compensation to create a more diverse and just academic environment.

Traditional social network research in education has primarily focused on positive and neutral networks, overlooking the dynamics of negative networks and their impact on individual well-being. This study applies egocentric network analysis to investigate the structural and psychological consequences of teachers’ networks involving aggressive and violent students. By integrating network theory with the Job Demands-Resources (JD-R) model, we examine how negative networks function as stress-inducing structures, influencing burnout, psychological distress, and turnover intentions among teachers. Using data from 507 K-12 teachers across 42 U.S. states, who collectively identified 1,703 aggressive or violent students, we analyzed network centrality, density, and homophily as key predictors of teacher outcomes. Findings indicate that teacher victimization tends to be more individualized rather than structurally embedded within student peer groups, as reflected in low network density. Path analysis revealed that teachers with higher centrality (more aggressive students) experienced greater burnout and distress, leading to stronger turnover intentions. However, network density (connections between aggressive students) and racial homophily did not significantly predict teacher outcomes. Furthermore, the frequency of aggression played a more critical role in psychological distress than severity, indicating that chronic exposure to low-severity aggression may have cumulative negative effects. These findings contribute to network theory by advancing our understanding of negative networks in educational settings. We discuss methodological innovations in applying egocentric network analysis to school-based victimization research and propose new directions for integrating negative network frameworks into studies of organizational stress, workplace violence, and teacher retention.

This study investigates the role of key authors in shaping early AIDS discourse on Usenet, a decentralized network that became an unexpected yet vital platform for public health communication between 1982 and 1986. Drawing on over 15,000 threads and 43,000 comments from six Usenet newsgroups—including net.med, net.motss, and net.singles—the research employs computational methods—such as sentiment analysis, Latent Dirichlet Allocation (LDA), Structural Topic Models (STMs), and network analysis—and historical interpretation to analyze the thematic, emotional, and networked dimensions of early digital health communication. Findings reveal that discussions on AIDS were marked by heightened negative sentiment, reflecting the pervasive fear, stigma, and emotional volatility that accompanied the epidemic’s initial spread, and that the discourse was predominantly shaped by scientific and medical themes, a characteristic shaped by the platform’s academically oriented user base and their need to navigate complex and rapidly evolving health information amidst widespread uncertainty. Crucially, several key authors emerged as central nodes in this communication network, acting as mediators who bridged specialized technical knowledge with broader social narratives. Their influence extended beyond simple content generation to shaping thematic priorities, emotional tone, and the adoption of critical terminologies, and as a result, fostering collaborative knowledge production and establishing trust within the community. These findings underscore the dynamic interplay between grassroots communication and emergent digital infrastructures in mediating public health crises. By situating these observations within broader sociological frameworks of power, influence, and digital publics, this study highlights the enduring significance of early online networks in shaping public health discourse.

Multivariate time series are ubiquitous objects found in a variety of domains. Generalised Network Autoregressive (GNAR) models attempt to capture the network structure of such objects, whilst avoiding the curse of dimensionality by using a small number of parameters. Applications in economics, epidemiology and transportation, among others, have shown its power and convenience. However, the model requires a predetermined network structure to define the dependencies of the multivariate time series.

The aim of the presentation will be to show theoretical and empirical evidence to prove the utility of genetic algorithms in detecting such underlying network structures. We will explore applications in macroeconomics, especially looking at their utility in unemployment forecasting.

A promising step towards addressing both the climate crisis and the obesity epidemic is to understand how healthy and pro-environmental food intake can be improved in the formative years of young adulthood. As young adults often eat and drink in the presence of others, it has been acknowledged that the social context is an important determinant of their food intake. However, research shows that previous aggregated, group-level findings may not generalize to personal processes within individual young adults across time and contexts. Therefore, in this three-year project, I will adopt an idiographic approach to test the within-person effects of social influence on young adults’ food intake.

To achieve this aim, I aim to employ a mixed-methods design including qualitative and quantitative data. Using a daily diary study, I will study which youth are influenced by their social relationships on the intake of several relevant food types. Innovatively, the most important social relationships of these young adults are invited to join this diary study as well. General mechanisms of social influence are identified by a review study and focus groups. Whether these mechanisms apply to individual young adults will be reflected upon with the participating youth while reviewing their social influence dynamics on their own food intake after the diary study. Finally, focus groups will be held to co-create suggestions to improve food intake.

I look forward to talking and brainstorming with you about the design of this three-year mixed-method study.

This presentation shows how ideanet, a recently released R package, remedies common issues related to working with egocentric network data. These tools include a function - nc_read - for reading in data exported from Network Canvas, making Network Canvas more accessible to new adopters.

Researchers often collect egocentric data when efforts to capture sociocentric networks are impossible or highly impractical. However, methods used for egocentric network collection and storage are less standardized than those used for sociocentric networks, posing risks for data sharing and reproducible analysis. The Network Canvas suite of software tools - and more recently Network Canvas Fresco - offers methods for egocentric data capture that create efficient and intuitive survey instruments. However, users have encountered barriers in the analysis of Network Canvas data, as Network Canvas stores data in its own unique format, that often must be restructured for analysis. Therefore, analyses of Network Canvas data may remain difficult for scholars less familiar with this format and scholars unaccustomed to the restructuring of network data. To address this problem, researchers can use ideanet to read in and reshape data stored by Network Canvas so it can easily be adapted for analysis. Moreover, ideanet includes comprehensive tools for summarizing local networks once they are read into R, creating a seamless bridge between data cleaning and analysis. ideanet promises greater efficiency, reliability, and reproducibility for studies using egocentric data. These benefits are amplified when ideanet is paired with Network Canvas, and studies using both tools suggest a future standard for egocentric network analysis.

Life adversities create stress, challenging both survival and mental health recovery. They can disrupt personal, work, and academic identities but may also foster resilience, particularly in youth. Research indicates that individuals facing catastrophic events utilize resources such as personal traits (e.g., cognition, personality), biological factors (e.g., immune system health), and social support networks. However, the relative impact of these factors and their interactions in influencing both well-being loss and recovery remains underexplored. This study proposes a longitudinal design, piloted with 40 university students (average age 23, 72% women). Buccal swab samples will be collected to analyze cytokine and chemokine profiles related to long term chronical stress-induced inflammation. Participants will fulfill psychosocial questionnaires assessing the nature of 12 adversities, individual differences, and aspects of their personal support networks (size, density, centrality, relationship quality, and social tie diversity). This poster will analyze personal networks and will presents some findings on personality traits, psychosocial variables, immunocompetence, and transcriptional responses to adversity, considering the 12 types of adversities evaluated. It aims to discuss and refine hypotheses regarding the interplay between individual traits, biological factors, and social influences in coping with adversities prevalent in local population. These preliminary conclusions will be further tested in a larger longitudinal study, part of the Biobanco Iztapalapa macro project.

Rural teachers often work in isolated settings, limiting collaboration and professional growth opportunities. This study examines the professional networks of rural science teachers in a western U.S. state, focusing on collaboration, advice-seeking, and friendship ties while exploring a professional development process known as Technology-Mediated Lesson Study (TMLS) as a driver of network formation. Using exponential random graph models (ERGMs) and separable temporal ERGMs (STERGMs), we analyzed network formation and evolution among 330 rural secondary science teachers (13 TMLS participants) at T1 (2023) and 195 rural secondary science teachers (20 TMLS participants) at T2 (2024). ERGMs revealed that network ties in collaboration, advice-seeking, and friendship were influenced by factors such as reciprocity, transitivity, and grade-level specialization. For example, rural science teachers were more likely to establish ties with colleagues teaching the same subject. We also found that propinquity—indicated by employment within the same school district—was a strong predictor of all three types of network ties. However, TMLS participants were able to broaden their professional connections beyond their districts, thus enhancing their networks for collaboration, advice-seeking, and friendship. This indicates that their participation in professional development, such as TMLS, enabled them to become knowledge sources beyond direct involvement in TMLS. STERGMs showed that rural science teachers’ networks evolved through reciprocity and transitivity, with established ties facilitating new connections. Participation in TMLS significantly contributed to expanding these networks for rural science teachers. This study provides valuable insights into the influence of technology-enhanced professional development on the professional networks of rural teachers.

Advice-seeking networks play a crucial role in organizational knowledge exchange, allowing individuals to access valuable information and expertise across intraorganizational boundaries. Traditionally, advice relationships are influenced by social and psychological factors, such as status-based selection and the costs associated with exposing one’s ignorance. However, the emergence of Generative Artificial Intelligence (GenAI) introduces a transformative layer of informational support that may redefine these networks.

By generating responses resembling explanations and advice, GenAI acts as a cognitive assistant, mitigating the psychological and social barriers associated with seeking human advice. This shift has the potential to alter the structure of advice-seeking networks by reducing reliance on traditional human advisors, while simultaneously increasing the centrality of individuals with expertise in AI-related tasks, such as prompting and tool navigation, and domain-specific expertise, depending on the area of application (e.g., compliance specialists for regulatory processes supported by GenAI solutions).

This study investigates how GenAI impacts work-related advice-seeking networks, focusing on professionals working as consultants in IT consulting and digital services firms. Through Social Network Analysis (SNA), we examine the extent to which GenAI reduces intensity of advice seeking networks by providing direct informational support. Additionally, we explore whether increased perceived hallucinations in GenAI outputs lead to sustained reliance on human advisors, while further amplifying the centrality of domain specialists.

Our findings aim to provide actionable insights for organizations adopting GenAI, shedding light on how technology-driven shifts influence collaborative dynamics and professional interactions within knowledge-intensive environments.

The capital controls of 2015 and the COVID-19 pandemic have prompted Greece to engage more deeply with the digital transformation of its economy. In our research, situated within the field of health sociology, we aim to illuminate how the digital economy has altered working conditions within a collaborative workspace. We chose one of the most prominent and technologically advanced hospitals in Attica, as it provides a valuable opportunity to observe the impact of technological transformation on various factors, including the workplace environment, career stages, technological proficiency, and gender dynamics.

Our research is structured into three sections, each addressing a different aspect of our inquiry. In the first part, we examine the changes in labor relations in relation to the level of digitization among staff and services, using 2023 as a reference point. The second part presents our key conclusions, focusing on identifying and analyzing contradictory evidence. A significant concern regarding the hospital's social network economy is the considerable bureaucratic oversight imposed on remote workers and those who heavily rely on digital platforms.

In the third part, we synthesize our quantitative and qualitative results, leading to several policy recommendations for the hospital administration. These recommendations primarily address the reduction of working hours, the bureaucratic oversight of employees, the emergence of new employment forms, and measures to close the gender gap, which appears to have widened according to the data collected.

International collaboration in scientific research is vital for advancing knowledge, fostering development, and driving innovation, particularly in health and life sciences. Such partnerships enable the exchange of expertise, resources, and technologies, accelerating scientific progress and enhancing collective responses to shared challenges. As Latin America’s largest economy and leading research hub, Brazil plays a pivotal role in South-South cooperation, strengthening scientific ties across the Global South. This study analyzes Brazil's contributions to international health and life sciences research from 2010 to 2019, focusing on South-South and triangular (North-South-South) collaborations. Using network analysis of 362,860 articles and reviews from the Scopus database, we demonstrate that Brazil’s international collaborations nearly doubled during this period. While partnerships with high-income countries (HICs) remained dominant, triangular collaborations involving HICs and low- and middle-income countries (LMICs) tripled, highlighting Brazil’s strategic role in bridging HICs with LMICs to address global challenges. The United States of America and the United Kingdom emerged as Brazil’s primary HIC partners, while Argentina and Colombia were central to South-South networks. Research priorities diverged by collaboration type: BR-HIC partnerships spanned diverse health and life sciences disciplines; triangular cooperation emphasized clinical trials and microbiology, and South-South collaborations prioritized taxonomy, systematics, ecology, and applied biochemistry. These findings underscore Brazil’s leadership in fostering equitable global partnerships, advancing health, biodiversity conservation, and sustainable development innovation.

Against the background of flow space, the interaction and connection of element flows at the urban scale form intercity network (Beaverstock et al., 2000; Castells, 1996). The city is a complex system that evolves through its inherent social and economic interactions (Batty, 2008; Lee et al., 2017) . The development of a city is influenced not only by its size but also by its position and role within the urban network (Neal, 2011).

This study takes 87 cities (autonomous prefectures) in the Yellow River Basin as the basic unit, based on macroeconomic data on urban size, enterprise data derived from ownership linkages, and traffic data on intercity flights, railway, and highway schedules. And the rank-size rule, urban gravity model and social network analysis are integrated. To gain deeper insights into city system evolution, it is essential to examine its structure from both a hierarchical size and relational network (Pflieger， and Rozenblat, 2010; Rozenblat, 2010), while also analyzing the dynamic interactions between the two dimensions. This study integrates hierarchical size and network connectivity to develop an urban centrality evaluation framework, which characterizes urban positions and function by assessing the combined influence of urban size and network centrality. The cities in the Yellow River Basin are classified into five types, including regional central cities, sub-central cities, district central cities, general cities, and cities with lower levels of development.

The study not only enhances the understanding of the urban system from both functional and relational perspectives but also holds significant practical implications for optimizing resource allocation, strengthening regional economic connectivity, and further promoting high-quality development in the Yellow River Basin.

Economic inequality and income disparities across urban areas have long been central topics in urban economics and regional development. Traditional studies on income convergence have primarily focused on administrative boundaries, such as counties, districts, or neighborhoods. However, we also know from social network studies that information, job opportunities and knowledge flow along social connections, as such, income convergence facilitated by these processes might not align well with administrative divisions. In this work, we examine the relationship between administrative divisions, social network communities, and income convergence within the top 50 metropolitan areas of the US. Using data on the mutual followership of Twitter users from 2012-2013, we create weighted census tract networks for the metropolitan areas, in which we detect communities using a spatial Louvain algorithm. We compare the sigma and beta convergence of census tract incomes in administrative divisions and communities, while controlling for demographic characteristics and the initial income levels of the census tracts. Our results suggest that sigma convergence within network communities is significantly larger than within other administrative areas of cities. Also, there is significant beta convergence among tracts with higher community income levels enhancing tract income growth. These findings highlight the importance of social network structures when understanding economic dynamics, and they suggest that policies aimed at reducing income inequality may benefit from considering network-based boundaries.

Social networks play a crucial role in shaping women’s roles in household decision-making by influencing their relative bargaining power and expectations. Existing economic research often assumes that household bargaining power is static or shaped primarily by household and individual characteristics, overlooking the role of dynamic peer influence and social support. This study aims to address this gap by proposing a model where decision-making power is the result of a social diffusion process through information and social support within one’s social network. This allows for women to deviate from cultural norms or their expected role based on their relative bargaining power compared to their spouse. Using social network and household decision-making data from two villages in southern India, I will test whether women’s household decision-making power corresponds with this model. Results will provide useful context on the role of social capital and peer influence in shaping household gender dynamics, modelling how norms may shift through local networks, which should be considered in policy interventions designed to support women.

With the inauguration of the Trump administration in 2024, the diversification of protectionist tools became more pronounced, as it extensively employed both tariff and non-tariff measures, such as imposing high tariffs on Chinese imports and tightening regulatory barriers. Amid this trend of trade protectionism diversification, it is crucial to analyze the structural characteristics of different protectionist measures. This study employs Social Network Analysis to examine the network structure of protectionism measures among countries. Using Global Trade Alert data, it analyzes distinct networks formed by these measures. By treating count variables as weighted network data, it applies Stochastic ERGM to assess endogenous factors (e.g., reciprocity) and exogenous factors (e.g., UNGA voting patterns) to determine their significance in shaping trade protectionism networks.

The recent 2024 riots in Southport, England and the subsequence eruption of violent demonstration across the UK have highlighted the role of the far right in encouraging and leading these events. Worldwide, there is a resurgence of far right and neo-nazi thinking. Online influencers are key to this in spreading these views, often through lies and disinformation. What can we do about this? We can combat these lies online. or we can oppose the violent demonstrations with peaceful protest. But we can also map the network of far right individuals and organisations that are active in promoting these ideas and taking part in these violent events so that we understand the system we oppose.

Networks can be mapped and analysed. We can identify who is potentially most influential because of their position in the network. We.can also show how these networks cluster because of how they are connected. This gives us useful information on patterns of collaboration and mutual action - and that can lead to practical ideas on how to oppose extremism and disrupt its influence.

The poster will show a tentative first stage attempt to build a netmap from press reports and the UK's Hope not Hate database. Currently, this makes no attempt to show the historical development of the right. It crudely includes all the connections to organisations and individuals that may no longer exist, but which have had an influence on development. Later we will time-tag nodes so that we can animate the evolution of the movement.

This is very much a ‘hobby map’. Creating such maps helps me understand the complexities of the growth and spread of views and actions. A similar map of the Brexit campaign gained around 38,000 hits on the internet. Many of the actors on that map are also in the PR.esented map.

This study employs semantic network analysis to examine NATO-related discourse in Romania, a geopolitical battleground on NATO’s Eastern frontier.

The annulment of the November 2024 presidential elections amid allegations of Russian interference and anti-Western propaganda intensified debates on NATO’s role. We analyze shifts in media and social media narratives one month before and after this event. Using the Network Agenda Setting (NAS) model, we construct semantic networks from news media and Facebook sources, applying Named Entity Recognition (NER) to track entity co-occurrences, network structure, and community clustering.

Findings show a post-electoral increase in the visibility of anti-NATO clusters and entities (i.e., nodes). Before the election, the main clusters focused on the elections (329 nodes), the war in Ukraine (270), Moldova (69), NATO (51), and regional issues (11). Afterwards, the electoral cluster expanded significantly (634 nodes), now including as a prominent entity Calin Georgescu, the anti-NATO candidate, who was entirely absent in the pre-election network but later accumulated 1,500 occurrences. The Ukraine war cluster expanded (414 nodes) with increased Russia references (809 to 1,900). A separate Russia cluster (12 nodes) emerged, while NATO discourse tripled to 146 nodes.

Clusters also show significant shifts in sentiment, being dominated by anti-NATO entities. Mentions of Romanian military bases (Deveselu, Cincu, Kogălniceanu) rose from 31 to 162, largely in the context of anti-Western rhetoric from the candidate, who advocated for closures or budget reductions.

Our findings will contribute to a deeper understanding of media influence on public perception in regions vulnerable to information warfare.

Background: Sustaining the implementation of infection prevention and control (IPC) practices in neonatal care remains a continuous challenge. While this extends beyond neonatal care settings, the particular vulnerability of the neonatal population underscores the need for continuous, effective IPC. Existing research on implementation often does not integrate the complex, interrelated dynamics influencing implementation. This study combined systems thinking and implementation science, framing neonatal care settings as complex systems. It employed network analysis to explore the relationships of reported implementation determinants, strategies and IPC practices.

Methods: Implementation determinants and implementation strategies, coded according to the Consolidated Framework for Implementation Research (CFIR) and the Expert Recommendations for Implementing Change (ERIC) taxonomy, respectively, were extracted from a systematic review of 153 studies. The extracted CFIR constructs, and ERIC strategies were tabulated, and the pairwise cosine similarity estimated. Similarity networks were computed to visualize the associations of CFIR constructs, ERIC strategies and IPC practices.

Results: The similarity networks of CFIR constructs and ERIC strategies suggest that certain combinations of contextual conditions are consistently represented across various IPC practices, while implementation strategies exhibit an unspecific application pattern across these practices. Additionally, relationships among IPC practices vary based on their contextual conditions and applied implementation strategies.

Conclusion: This study extends the use of network analysis within implementation science by analyzing non-social entities and applying associative measures to examine underlying interactions. While further research is needed to validate this approach, the results highlight the need for tailored implementation strategies that address contextual dynamics influencing implementation.

School networks are commonly considered instruments for governance and school improvement but are rarely scrutinised as objects of study. Conversely, this perspective is central to Organisational Studies, a standpoint from which it is likely to think that schools may engage unevenly in inter-organisational networks, with some schools comparatively better or worse connected than others. This article empirically explores this assumption in a context where recent evidence shows that schools build inter-organisational relationships for multiple reasons, some vital for organisational sustainability and students’ educational opportunities. This study is underpinned by notions from Organisational Theory and carried out following a sequential Mixed Method Social Network Analysis approach (MMSNA). In the first stage, a weighted degree centrality measure was computed using data collected from mapping interviews with school principals to identify well-connected schools. In a second qualitatively-led stage, we aimed to respond to: What characteristics are associated with central schools? What is the content of relationships schools have with central schools? We describe three types of schools: i) symbiotically dependent, ii) exemplary, and iii) charismatic-led school – accounting for nearly a third of the total links in a local urban area – and whose centrality can be understood at various scales, driven by different motivations, and establishing inter-organisational relationships of diverse kinds. These insights not only contribute to the school networking scholarship but also illuminate broader issues that hold particular significance when understood at an inter-organisational level.

Since the first appearance of the Hegselmann-Krause bounded confidence model, many efforts have been made to include properties usually present in real-world systems into it. In particular, the inclusion of heterogeneous agents on systems with a subjacent network structure have shown prominent advances in recent years. In this work we use these studies as a preliminary background to understand the effect of networks with community structure on the steady state of the heterogenous Hegselmann-Krause opinion model. To this extent, we propose a novel benchmark of random networks composed of sparsely interconnected communities that can be used to generate a uniform ensemble of networks with the desired properties to study real-world social systems. Using this ensemble to create networks with high clustering, power-law degree distributions and small world behavior, we show that, when agents are divided into communities, having a large amount of individuals with high confidence bounds will make the opinion on each community converge to a weak consensus. Nevertheless, we also show that given that communities are sparsely connected, each community will have a different mean opinion, preventing the system to reach consensus as a whole. Finally, we observed that the system can also reach a state of polarization where each community becomes polarized between two different opinions.

As ageing populations grow and healthcare workforce shortages intensify, informal caregiving networks are becoming increasingly crucial in supporting older individuals. According to the European Commission (2021), the demand for informal care is expected to rise due to population ageing across Europe. In Fryslân, where this trend is accelerating, social networks play a vital role in healthcare utilization, particularly in rural areas with scarce formal healthcare services.

This study examines how social network structures—size, density, and centrality—influence informal and formal care utilization among older populations in urban and rural Fryslân. Grounded in Social Network Theory (SNT) and Social Support Theory (SST), it explores how network structure shapes healthcare utilization while considering emotional, instrumental, and informational support. SNT highlights the role of highly central individuals in influencing utilization patterns, while SST emphasizes caregiving networks' tangible assistance in utilizing healthcare services.

Using Social Network Analysis (SNA) and Exponential Random Graph Models (ERGM), this study assesses the extent to which social networks shape formal (professional) and informal (family/community-based) care utilization. Given that older rural populations often rely on smaller but denser networks, it examines whether they depend more on informal caregiving due to limited ways of utilizing professional healthcare services.

The study analyzes secondary data from the SHARE survey and Planbureau Fryslân datasets to identify network-driven disparities in healthcare utilization. Findings will inform policy recommendations to strengthen informal caregiving while promoting equitable utilization of professional care, advancing network-based interventions for ageing populations.

Background: Medications for opioid use disorder (MOUD, i.e., buprenorphine, methadone, naltrexone) save lives, but many individuals are unwilling to receive MOUD treatment. Few longitudinal studies have examined whether social network support for MOUD is related to willingness to receive MOUD.

Methods: Egocentric network data were collected from 395 adults with OUD who were incarcerated in 14 Kentucky prisons; follow-up data were collected about six months post-release. For each named alter, participants reported on support for buprenorphine, methadone, and naltrexone at baseline and follow-up; baseline network scores were averaged into a network support for MOUD score and a change score from baseline to follow-up was calculated. Multiple imputation by chained equations addressed missing data. A multivariate linear regression model of willingness to receive MOUD at follow-up was estimated that included baseline network support for MOUD, change in network support, baseline willingness to receive MOUD, lifetime history of MOUD, and demographics.

Results: Baseline network support for MOUD was significantly associated with willingness to receive MOUD at follow-up (b=.21, 95% CI: .10-.31, p<.001). Change in network support for MOUD over time was also significantly associated with willingness to receive MOUD at follow-up (b=.18, 95% CI: .10-.27, p<.001).

Conclusion: Consistent with social network theories, norms within networks appear to influence participants’ willingness to receive MOUD. A unique contribution is our consideration of the dual impacts of initial levels of network support and how improvements in network support over time had positive impacts on participants’ willingness to receive MOUD.

Funding: Supported by NIDA R01DA48876.

Without social capital, nothing flourishes: neither democracy, nor the economy, nor regions, nor people. We have witnessed this both in our experience modernizing public organizations during Chile’s return to democracy and in the current work of the Program of Innovation and Sociotechnology (PISCT) at the Universidad de Chile. Our theoretical, philosophical, and methodological approach is grounded in biologically based Radical Constructivism (Humberto Maturana and Francisco Varela) and the ontology of conversation, with a special emphasis on building trust—a particularly pressing challenge in a country where trust indices are notably low.

We have intervened in more than 50 organizations spanning the public and private sectors and civil society, from small grassroots groups to high-level executives. This has enabled us to observe the power of communication and collaboration in driving paradigm shifts, strengthening social capital, and fostering the development of socioemotional and leadership skills.

Our findings indicate that maintaining these transformations requires ongoing work, as temporary changes alone are insufficient. Trust and cooperation must be sustained over time. Thus, our experiences provide a theoretical and practical perspective on the relevance of social capital, underscoring the need for sociotechnological strategies that enhance collective well-being and advance the comprehensive development of territories.

With its many programs intended to welcome young foreigners, migration to Quebec is particularly young; the average age among newcomers was 29 in 2023-24 (ISQ, 2024). Leaving one's home to go elsewhere then takes place in a period of life that is pivotal to multiple transitions in the different spheres of life (e.g. family separation, pursuit of higher education, insertion in specialized employment). For these young people, the acquisition of autonomy and independence is then likely to be experienced far from the people around them (Atkin and Longo, 2022).

Based on a procedural theoretical framework (Mendez, 2010), this poster presentation questions influential people about the choice of young people to migrate to Quebec. Based on a longitudinal and qualitative methodology, the analysis is based on data collected in 3 successive waves of interviews over a period of two years. In total, 39 young migrants (22 women and 17 men) aged 21 to 35 and settled in Quebec on a voluntary basis (temporary or permanent) were interviewed.

The results tend to demonstrate that migration is an intrinsically relational choice. Different people influence this choice to go to Quebec, particularly the relatives with whom the young person has a strong tie (e.g. spouse, friend, family, teacher). Their forms of influence are mainly symbolic and vary according to the type of influential people. Moreover, the duration of the ties varies according to the opportunities, such as those of sharing time together or having common and concrete (e.g. starting a family, starting a business).