scholarly journals A Bayesian networks approach to infer social changes in burials in northeastern Taiwan during the European colonization period

2021 ◽  
Author(s):  
Li-Ying Wang ◽  
Ben Marwick
Keyword(s):  
Network Analysis ◽  
Iron Age ◽  
Network Formation ◽  
Exponential Random Graph Models ◽  
Network Ties ◽  
Archaeological Data ◽  
Robust Model ◽  
Exponential Random Graph ◽  
European Colonization ◽  
Colonization Period

Burials provide valuable information to study social structures based on the assumption that burials and associated grave goods can represent social roles and relations in a society. To study social relationships, network analysis has been increasingly applied to archaeological data to infer interactions and relationships between entities. Statistical approaches to network analysis, such as exponential random graph models (ERGMs), provide a way to test hypotheses about dynamic processes of network formation. However, computational difficulties and sensitivity to uncertainties limit the application of ERGMs. In this paper, we introduce a Bayesian framework on ERGMs that enables an efficient computational process, effective quantification of uncertainty, and robust model evaluation of network properties. We tested a hypothesis of social change relative to the arrival of Europeans by studying burial data from Kiwulan, an Iron Age site in northeastern Taiwan. The results indicate a transition among the burials from network ties based on ritual objects to wealth objects, and a more centralized structure with increased social differentiation after the European presence was established in the 17th century. Our case study demonstrates the effectiveness of Bayesian network analysis for archaeological data, and expands the use of burials in understanding the impacts of colonial presence on Indigenous groups in a pericolonial context.

scholarly journals Investigation of Rater Effects Using Social Network Analysis and Exponential Random Graph Models

2017 ◽  
Vol 78 (3) ◽  
pp. 430-459 ◽  
Author(s):  
Iasonas Lamprianou
Keyword(s):  
Social Network ◽  
Social Network Analysis ◽  
Network Analysis ◽  
Exponential Random Graph Models ◽  
Graph Models ◽  
High Stakes ◽  
Rating Data ◽  
New Approach ◽  
Exponential Random Graph ◽  
Rater Severity

It is common practice for assessment programs to organize qualifying sessions during which the raters (often known as “markers” or “judges”) demonstrate their consistency before operational rating commences. Because of the high-stakes nature of many rating activities, the research community tends to continuously explore new methods to analyze rating data. We used simulated and empirical data from two high-stakes language assessments, to propose a new approach, based on social network analysis and exponential graph models, to evaluate the readiness of a group of raters for operational rating. The results of this innovative approach are compared with the results of a Rasch analysis, which is a well-established approach for the analysis of such data. We also demonstrate how the new approach can be practically used to investigate important research questions such as whether rater severity is stable across rating tasks. The merits of the new approach, and the consequences for practice are discussed.

Why Legislative Networks? Analyzing Legislative Network Formation

2017 ◽  
Vol 7 (3) ◽  
pp. 505-522 ◽  
Author(s):  
Stefan Wojcik
Keyword(s):  
Political Parties ◽  
Network Formation ◽  
Contemporary Literature ◽  
Party Organization ◽  
Exponential Random Graph Models ◽  
Graph Models ◽  
Personal Traits ◽  
Random Graph Models ◽  
Exponential Random Graph ◽  
The Social

Are the social networks of legislators affected more by their political parties or their personal traits? How does the party organization influence the tendency of members to work collectively on a day-to-day basis? In this paper, I explore the determinants of the relationships of legislators in the Brazilian Chamber of Deputies. I use exponential random graph models to evaluate the relative influence of personal traits versus party influence in generating legislator relationships. Despite a focus on personalism in Brazil, the analysis reveals that the effects of political parties on tie formation are roughly equal to the effects of personal traits, suggesting that networks may make political parties much more cohesive than contemporary literature would lead us to believe.

Ecological Influences on the Formation of the Hiring Network in the Communication Job Market, 2015 to 2019

2021 ◽  
pp. 009365022110346
Author(s):  
Yu Xu
Keyword(s):  
Ecological Model ◽  
Ecological Factors ◽  
Niche Overlap ◽  
Formation Mechanisms ◽  
Hiring Decisions ◽  
Interorganizational Network ◽  
Exponential Random Graph Models ◽  
Network Ties ◽  
Past Behavior ◽  
Exponential Random Graph

This study investigates ecological factors that drive hiring decisions in the academic marketplace. Faculty hires between institutions are conceptualized as interorganizational network ties. Drawing on theoretical insights from network inertia and niche processes in organizational ecology, the current study builds an ecological framework to explain the formation mechanisms of the faculty hiring network among 81 U.S. institutions granting PhDs in communication. Consistent with the predictions of the ecological model of hiring decisions, the empirical results of exponential random graph models (ERGMs) revealed that past behavior (or the presence of previous ties), niche width (or the number of research specializations), and niche overlap (or the degree of shared research specializations) significantly constrained the likelihood of tie creation during the 2015 to 2019 period. These effects held true even when traditional explanations such as network self-organization and status-based hiring patterns were taken into account. Theoretical and practical implications are discussed.

Network Analysis: A Systems Framework to Address Grand Challenges in Plant Pathology

2018 ◽  
Vol 56 (1) ◽  
pp. 559-580 ◽  
Author(s):  
K.A. Garrett ◽  
R.I. Alcalá-Briseño ◽  
K.F. Andersen ◽  
C.E. Buddenhagen ◽  
R.A. Choudhury ◽  
...  
Keyword(s):  
Plant Pathology ◽  
Network Analysis ◽  
Disease Management ◽  
Agricultural Development ◽  
Research Priorities ◽  
Exponential Random Graph Models ◽  
Graph Models ◽  
Exponential Random Graph ◽  
Plant Disease Management ◽  
Systems Framework

Plant pathology must address a number of challenges, most of which are characterized by complexity. Network analysis offers useful tools for addressing complex systems and an opportunity for synthesis within plant pathology and between it and relevant disciplines such as in the social sciences. We discuss applications of network analysis, which ultimately may be integrated together into more synthetic analyses of how to optimize plant disease management systems. The analysis of microbiome networks and tripartite phytobiome networks of host-vector-pathogen interactions offers promise for identifying biocontrol strategies and anticipating disease emergence. Linking epidemic network analysis with social network analysis will support strategies for sustainable agricultural development and for scaling up solutions for disease management. Statistical tools for evaluating networks, such as Bayesian network analysis and exponential random graph models, have been underused in plant pathology and are promising for informing strategies. We conclude with research priorities for network analysis applications in plant pathology.

Μeasuring and Visualizing Coders’ Reliability: New Approaches and Guidelines From Experimental Data

2020 ◽  
pp. 004912412092619
Author(s):  
Iasonas Lamprianou
Keyword(s):  
Experimental Data ◽  
Social Network ◽  
Social Network Analysis ◽  
Network Analysis ◽  
Random Graph ◽  
Exponential Random Graph Models ◽  
Graph Models ◽  
New Approach ◽  
Random Graph Models ◽  
Exponential Random Graph

This study investigates inter- and intracoder reliability, proposing a new approach based on social network analysis (SNA) and exponential random graph models (ERGM). During a recent exit poll, the responses of voters to two open-ended questions were recorded. A coding experiment was conducted where a group of coders coded a sample of text segments. Analyzing the data, we show that the proposed SNA/ERGM method extends significantly our analytical leverage, beyond what popular tools such as Krippendorff’s α and Fleiss’s κ have to offer. The reliability of coding for individual coders differed significantly for the two questions although they were very similar and the same codebook was used. We conclude that the main advantages of the proposed SNA/ERGM method are the intuitive visualizations and the nuanced measurements. Detailed guidelines are provided for practitioners who would like to use the proposed method in operational settings.

Proximity and centrality in inter-organisational collaborations for innovation

2019 ◽  
Vol 58 (2) ◽  
pp. 239-254 ◽  
Author(s):  
Francesco Capone ◽  
Vincenzo Zampi
Keyword(s):  
Network Analysis ◽  
Inner Core ◽  
Preferential Attachment ◽  
Hierarchical Structures ◽  
Primary Data ◽  
Exponential Random Graph Models ◽  
Graph Models ◽  
Content Type ◽  
Exponential Random Graph ◽  
The Impact

Purpose The purpose of this paper is to investigate the impact of the different drivers of the establishment of innovation relationships in an aerospace cluster. In particular, the work investigates the impact of the various forms of proximity in the formation of inter-organisational collaborations for innovation. Design/methodology/approach The analysis is based on primary data collected through interviews and questionnaires on innovation collaborations, administered to all the firms operating in the aerospace cluster in Tuscany. The work applies social network analysis and Exponential Random Graph Models to analyse the forces that drive inter-organisational collaborations for innovation. Findings Results confirm the importance of geographical proximity in the formation of ties in the cluster, but social proximity is one of the main drivers for tie formation. Reciprocity shows that companies develop innovations in a reciprocal way and that most relationships are bidirectional. Triadic closure is also relevant, where the role played by trust and previous relationships is evident. Finally, hierarchy network processes are underlined, where the most central actors of the network are the most popular confirming a processes of preferential attachment. The central organisations gradually are more important, whereas the marginal ones are left in the periphery. Originality/value The work presents some novelties. First, it measures the different impacts of the various forms of proximity together with more advanced measures of network analysis. It allows pointing out the relevance of a firm’s network position in clusters and the fact that clusters assume hierarchical structures similar to centre-periphery networks, where most relevant nodes are in the inner core and marginal organisations are relegated.

scholarly journals Inferential Network Analysis with Exponential Random Graph Models

2011 ◽  
Vol 19 (1) ◽  
pp. 66-86 ◽  
Author(s):  
Skyler J. Cranmer ◽  
Bruce A. Desmarais
Keyword(s):  
Network Analysis ◽  
Random Graph ◽  
Graph Model ◽  
Network Data ◽  
Exponential Random Graph Models ◽  
Graph Models ◽  
Random Graph Model ◽  
General Acceptance ◽  
Random Graph Models ◽  
Exponential Random Graph

Methods for descriptive network analysis have reached statistical maturity and general acceptance across the social sciences in recent years. However, methods for statistical inference with network data remain fledgling by comparison. We introduce and evaluate a general model for inference with network data, the Exponential Random Graph Model (ERGM) and several of its recent extensions. The ERGM simultaneously allows both inference on covariates and for arbitrarily complex network structures to be modeled. Our contributions are three-fold: beyond introducing the ERGM and discussing its limitations, we discuss extensions to the model that allow for the analysis of non-binary and longitudinally observed networks and show through applications that network-based inference can improve our understanding of political phenomena.

scholarly journals Duality beyond Dyads: Multiplex patterning of social ties and cultural meanings

2020 ◽  
Author(s):  
Nikita Basov ◽  
Julia Brennecke
Keyword(s):  
Semantic Network ◽  
Exponential Random Graph Models ◽  
Cultural Meanings ◽  
Network Ties ◽  
Shared Meanings ◽  
Exponential Random Graph ◽  
The Social ◽  
Interpersonal Ties ◽  
Cultural Networks ◽  
Cultural Structures

Purpose: The social and cultural duality perspective suggests dual ordering of interpersonal ties and cultural similarities. Studies to date primarily focus on cultural similarities in interpersonal dyads driven by principles such as homophily and contagion. We aim to extend these principles for socio-cultural networks and investigate potentially competing micro-principles that generate these networks, taking into account not only direct dyadic overlap between interpersonal ties and cultural structures, but also the indirect interplay between the social and the cultural. Methodology: The empirical analysis utilizes social and semantic network data gathered through ethnographic studies of five creative organizations around Europe. We apply exponential random graph models (ERGMs) for multiplex networks to model the simultaneous operation of several generative principles of socio-cultural structuring yielding multiplex dyads and triads that combine interpersonal ties with meaning sharing links. Findings: The results suggest that in addition to the direct overlap of shared meanings and interpersonal ties, socio-cultural structure formation is also affected by extra-dyadic links. Namely, expressive interpersonal ties with common third persons condition meaning sharing between individuals, while meaning sharing with common alters leads to interpersonal collaborations. Beyond dyads, the dual ordering of the social and the cultural thus operates as asymmetrical with regard to different types of interpersonal ties. Research implications: The paper shows that in addition to direct dyadic overlap, network ties with third parties play an important role for the co-constitution of the social and the cultural. Moreover, we highlight that the concept of network multiplexity can be extended beyond its application social networks to investigate competing micro-principles guiding the interplay of social and cultural structures.

scholarly journals Social network analysis of lodging establishments in Tiradentes (MG): Profile and evidence of relationships genesis

2020 ◽  
Vol 22 (3) ◽  
pp. 424-445
Author(s):  
Geraldo Magela Rodrigues De Vasconcelos ◽  
Gustavo Melo-Silva ◽  
Velcimiro Inácio Maia
Keyword(s):  
Social Network ◽  
Social Network Analysis ◽  
Network Analysis ◽  
Random Graph ◽  
Exponential Random Graph Models ◽  
Graph Models ◽  
Random Graph Models ◽  
Exponential Random Graph

A análise de redes sociais (ARS) constitui um grande avanço na pesquisa em turismo ao revelar as características das relações estabelecidas, apresentando suas estruturas e propriedades. Este trabalho objetivou caracterizar e analisar a rede de cooperação formada entre proprietários de pousadas em Tiradentes-MG. Com o objetivo de explorar o contexto da pesquisa, foram realizadas entrevistas junto a sete proprietários de pousadas. Posteriormente, para a coleta dos dados, foi aplicado um questionário aos proprietários. Por meio da técnica da “bola de neve” foi gerada a rede de cooperação. A partir disso, utilizou-se das técnicas da ARS, com ênfase nas métricas descritivas e pela modelagem de grafos aleatórios da família exponencial (ERGM - Exponential Random Graph Models). Com isso, foi possível caracterizar a rede de cooperação e analisar suas propriedades. A rede possui 54 pousadas, com uma densidade geral baixa, pois há apenas 4,7% de relações possíveis. Foram identificadas 17 pousadas centrais e 37 periféricas. No subgrupo central, a densidade foi de 15,4% e no subgrupo periférico, de apenas 2,6%. A rede de cooperação observada apresentou homofilia por gênero e por procedência dos proprietários. Os resultados da modelagem ERGM permitiram explicações probabilísticas em termos de atributos endógenos dos proprietários, como gênero e procedência.

Sign in / Sign up

Export Citation Format

Share Document