Social Network Analysis Collaboration (SNAC)
Information Systems. Knowledge Networks. Criminal Networks.
PROJECTS
Currently we are focusing on the following two projects:
1. Network Analysis of Knowledge Production Systems in Security Studies, IPE, and Health Sciences
2. Iran's Nuclear Capability: Dynamic Network Analysis
Security
Who Shapes Research on Afghanistan
The purpose of the project is to systematically assess all the publications on United States’ security strategy in Afghanistan from 2001 through the present. We employ Citation Network Analysis (CNA) to map out all relevant research. First, we gather all the relevant publications through a keyword search in the Web of Science and Scopus. Second, we use the publications and their associated references to compile a dataset. The dataset is processed through ORA to create three types of citation networks: author-author network, journal-journal network, and paper-paper network. Together these three networks comprise the knowledge system we analyze.
The system is examined for multidiciplinarity, heterogenuity and stakeholders representativness. The results of the analysis enable us to quantitatively assess the strengths and weaknesses of the intellectual structure that guides how research is produced in security studies. The insights are used to design recommendations aimed at ensuring that output produced by the security studies field is of the highest quality and uses best available research strategies.
Iran's Nuclear Trade Networks
In this project, we employ dynamic social network analysis to map out the suppliers of nuclear material and expertise to Iran and their contacts inside Iran as a network that evolves over time. We created a comprehensive dataset that captures nuclear technology, material, and knowledge that was supplied to Iran.
The data is processed through ORA. Once the network of suppliers is elicited, we employ centrality measures (out-degree degree centrality and betweenness centrality) to assess the most influential members of the network . Particular attention is paid to the ‘opinion leaders’ (nodes that have the highest out-degree centrality scores) and ‘brokers’ (top nodes in terms of betweenness centrality). Additionally, we evaluate the members of the supplier network based on the material or expertise they supply. Considering that there are two phases of the nuclear cycle–Enrichment and Weaponization–that specifically contribute to the development of nuclear weapons, the emphasis is on the members of the network that are most central to empowering those two phases.
Determining Iran’s most critical suppliers provides necessary insights for any policy directed at disrupting its nuclear capability.
Measuring Iran’s Nuclear Capability
The purpose of this study is to develop a classification-visualization technique that captures a country’s nuclear activities in a manner accessible to policymakers. The full extent of a country’s nuclear activities explicitly outlines its “supply-side” technical capability in the nuclear field. While this technique is generalizable, we use Iran’s nuclear program as a case study to assess Iran’s nuclear posture. To achieve this end, first, we build a comprehensive database that contains information on the equipment, materials, technology, and expertise that contribute to Iran’s nuclear program. Second, we use the Nuclear Threat Initiative’s (NTI) classification scheme of a nuclear program’s stages as a basis to develop a generalizable visualization tool that allows us to create a map of “nuclear capability.” By applying the scheme to the data we have collected on Iran, we use the resultant visualization to clarify points relating to the controversial aspects of Iran’s nuclear posture.
Terrorism in Afghanistan and Pakistan 1990-2011:
Dynamic Network Analysis
The aim of this study is to elicit the most lethal terrorist organizations in Afghanistan and Pakistan. Using data provided by the Global Terrorism Database, we conduct a dynamic network analysis (DNA) of one-mode and two-mode terrorism activity networks from 1994 to 2011. These include terrorist collaboration, target and weapon type networks. The lethality of each organization is assessed by determining its centrality in collaboration networks, expertise exclusivity in types of weapons and targets, activity level in regards to all types of weapons and targets and ability to consistently attack the most challenging targets. Binary variables such as success and suicide are incorporated in the form of attributes. Results are presented as dynamic sociograms and the findings are used to provide recommendations for the United States Armed Forces and relevant stakeholders. The evidence suggests that terrorist organizations undergo considerable transformation over time and there are critical differences between terrorist organizations based in Afghanistan and those in Pakistan.
Evaluating Terrorism Reserch
How research on terrorism changes over time? Who are the top authors and journals publishing on terrorism and what is the quelity of scholarship they produce? Is there any international scientific colaboration on terrorism issues? How did 9/11 affected terrorism research and its quality? Drawing on normative attributes of the knowledge production system delineated by Gibbons et al. (1994), heterogeneity and organizational diversity are operationalized as measurable indicators. A list of hypotheses is developed and tested in regards to terrorism research. Data consists of 5121 publications on terrorism between 1992 and 2012. Each publication is assigned to a quality variable (based on the 3-point Lum et al. scale) and several structural variables, such as author’s subject area, language of publication and type and country of the publishing organization. The data is longitudinal and relational, with publications connected to one another based on co-authorship. Additionally, all locational variables are assigned latitude and longitude coordinates. Geospatial and dynamic network analyses of collaboration networks of authors, organizations, countries and funding agencies are performed (1) to explore how the system varies over geographical space and time, and (2) to test the relationship between the structural attributes of knowledge production and the quality of knowledge that is being generated. The findings demonstrate that the quality of research varies considerably across countries, organization types and time. Analyzed dynamically, the terrorism research knowledge system grows more heterogeneous and organizationally diverse. Finally, while most variables measuring heterogeneity and organizational diversity correlate with quality, a portion of the indicators does not display a statistically significant relationship.
Health Policy and International Development
Assessing Research on Rural Access to Primary Care
Access to primary care in rural areas of the United States has been a pervasive issue for decades. Many policies and programs have been formed and implemented to address the problem, particularly in the current period of health reform. To better understand the nature of this issue, we consider the knowledge production system around rural access to primary care. The knowledge production system consists of all publications, including journals, books, whitepapers, and websites, cited on this topic. We propose that the Mode 2 framework, as defined by Gibbons, Scott, Nowotny, et al, is the normative understanding for knowledge production. Utilizing citation network analysis, we analyze two constructs from this framework, heterogeneity and multidisciplinarity, to understand their representation within the author and source networks. Our findings suggest that the knowledge production system for rural access to primary care is geographically concentrated and does not reflect a diversity of sites. The network, however, appears to be incorporating a wider scope of disciplines over time, such as dentistry, nursing, and mental health, suggesting an increase in multidisciplinarity. Funding for research that stresses involvement of multiple stakeholders across settings will foster this trend to develop sustainable solutions for this vulnerable population.
Development Aid in Sab Saharan Africa: Evaluating Invisible College
The aim of this study is to investigate whether high quality research positively correlates with an author’s role in the knowledge production system. The system consists of all research published on development aid to Sub-Saharan Africa from 1990-2012. We employ Citation Network Analysis (CNA) to elicit the knowledge production system. Using a role analysis framework based on Ronald Burt’s concept of brokers (1992), we classify members of the network (authors) according to their specific roles (opinion leaders, gatekeepers, and marginal authors). Centrality measures are used to determine and assign roles. Next, publications by opinion leaders, bridges, and a sample of marginals are evaluated for quality. Considering that half of the papers in any knowledge system are produced by one half percent of all the contributing authors, understanding the relationship between author’s role and quality is essential for any knowledge society that expects valid knowledge from the experts.
IUU Crab Fishing at the RFE:
SNA and Systems Thinking
Illegal, unreported and unregulated (IUU) crab fishing in the Russian Far East is a complex policy issue reflective of the policy challenges faced by policy-makers around the world. The Russian federal government attempted several policies and programs to address it, but the measures implemented up to date have been unsuccessful. In this study we combine a systems thinking approach with social network analysis to assess crab fishing in the RFE as a policy-issue system. Drawing on Churchman (1971), we determine the system’s goal and actors. Next, the system is analysed as an evolving communication network, where communication structure is represented by the actions actors perform towards one another and rules that govern each action (Ostrom, 1990, 2005). The rules are characterized as explicit rules, conforming to the law or implicit norms, deviating from the law. Analysing the system in terms of its actors and the explicit rules/implicit norms that govern interactions among the actors surfaces problematic areas in the system. Additionally, we employ social network analysis to determine the most central actors in the system (opinion leaders and brokers). Understanding the implicit rules and degree of influence of each actor in the system informs a strategy to stop illegal crab fishing and turn it into a lawful industry.
We will make the datasets available shortly.