Overcoming the liability of distance? An exploratory study of the associations between social networks, sense of community and spatial colocation

Purpose This study aims to examine the associations of social networks with the sense of community (SOC) construct and spatial colocation or having an office. The study site was an institute for health-care policy research formed in 2011 by bringing together scientists from more than 20 different university units. Only 30% of the scientists were had an office or physical presence at the institute. Therefore, the institute was an ideal site to examine whether SOC was correlated with different dimensions of network position – connectedness, reachability and brokerage – even when the authors account for the lack of spatial colocation for the off-site scientists. Design/methodology/approach A two-part (sociometric and workplace) internet survey instrument was administered in 2014 to the institute’s population of 411 individuals. The sociometric data were used to create an undirected interaction network and the following dependent variables (DVs) or network centralities: normalized degree to measure connectedness; average reciprocal distance to capture reachability; and normalized betweenness to proxy brokerage. Separate node-level network regressions were then run with random permutations (N = 10,000) and listwise deletion for each of the DVs with SOC and spatial colocation as the independent variables, and variables that controlled for gender, organizational affiliation and job category. Findings SOC and spatial colocation are both positively and significantly correlated with network connectedness and reachability. The results suggest that both SOC and spatial colocation have a larger impact on reachability than connectedness. However, neither SOC nor spatial colocation are significantly associated with network brokerage. Finally, the findings show that SOC and spatial colocation are more reliable predictors of network connectedness and reachability than are key individual- and unit-level control variables, specifically the individual’s sex, job category and organizational affiliation. The controls were not significantly associated with any of the three network centralities, namely, connectedness, reachability and brokerage. Originality/value This exploratory study used social network analysis and node-level network regressions to examine the associations from SOC and spatial colocation to dimensions of network position. SOC is positively and significantly associated with network connectedness and reachability, suggesting that SOC is an important consideration when individuals are disadvantaged from the absence of spatial colocation. The findings have implications for work in the context of the COVID-19 pandemic as they imply that interventions based on the SOC construct could potentially lessen the negative effects of remote work on workplace social networks due to factors such as the reduction of social contacts.

Directed closure coefficient and its patterns

The triangle structure, being a fundamental and significant element, underlies many theories and techniques in studying complex networks. The formation of triangles is typically measured by the clustering coefficient, in which the focal node is the centre-node in an open triad. In contrast, the recently proposed closure coefficient measures triangle formation from an end-node perspective and has been proven to be a useful feature in network analysis. Here, we extend it by proposing the directed closure coefficient that measures the formation of directed triangles. By distinguishing the direction of the closing edge in building triangles, we further introduce the source closure coefficient and the target closure coefficient. Then, by categorising particular types of directed triangles (e.g., head-of-path), we propose four closure patterns. Through multiple experiments on 24 directed networks from six domains, we demonstrate that at network-level, the four closure patterns are distinctive features in classifying network types, while at node-level, adding the source and target closure coefficients leads to significant improvement in link prediction task in most types of directed networks.

Portable Node-Level Parallelism for the PGAS Model

The Partitioned Global Address Space (PGAS) programming model brings intuitive shared memory semantics to distributed memory systems. Even with an abstract and unifying virtual global address space it is, however, challenging to use the full potential of different systems. Without explicit support by the implementation node-local operations have to be optimized manually for each architecture. A goal of this work is to offer a user-friendly programming model that provides portable performance across systems. In this paper we present an approach to integrate node-level programming abstractions with the PGAS programming model. We describe the hierarchical data distribution with local patterns and our implementation, MEPHISTO, in C++ using two existing projects. The evaluation of MEPHISTO shows that our approach achieves portable performance while requiring only minimal changes to port it from a CPU-based system to a GPU-based one using a CUDA or HIP back-end.

Exploring Network Measures of Social Capital: Toward More Relational Measurement

Although social capital is a relational concept, existing studies have focused less on measuring social relations. This article fills the gap by reviewing recent studies that used network measures grouped into three types according to the measurement level. The first group defined social capital as an individual asset and used node-level measures to explain personal benefits. The second group defined social capital as a collective asset and used graph-level measures to describe collective properties. The third group used subgraph-level measures to explain the development of social capital. This article offers a link between the concepts and measures of social capital.