Unifying spatial and social network analysis in disease ecology

• Published in: The Journal of animal ecology Vol. 90; no. 1; pp. 45 - 61
• Main Authors: Albery, Gregory F., Kirkpatrick, Lucinda, Firth, Josh A., Bansal, Shweta, Farine, Damien
• Format: Journal Article
• Language: English
• Published: England Blackwell Publishing Ltd 01.01.2021
• Subjects: animal ecology; Animals; Covariance; disease ecology; Ecology; Heterogeneity; Inference; methodology; Models, Biological; Network analysis; parasite transmission; Reliability analysis; Reproducibility of Results; Social analysis; Social behavior; Social Network Analysis; Social networks; Social organization; Spatial Analysis; Spatial data; Spatial heterogeneity; spatial variation; Spatial variations; spatial analysis; disease ecology; parasite transmission; methodology; social network analysis
• ISSN: 0021-8790; 1365-2656; 1365-2656
• DOI: 10.1111/1365-2656.13356

Social network analysis has achieved remarkable popularity in disease ecology, and is sometimes carried out without investigating spatial heterogeneity. Many investigations into sociality and disease may nevertheless be subject to cryptic spatial variation, so ignoring spatial processes can limit inference regarding disease dynamics. Disease analyses can gain breadth, power and reliability from incorporating both spatial and social behavioural data. However, the tools for collecting and analysing these data simultaneously can be complex and unintuitive, and it is often unclear when spatial variation must be accounted for. These difficulties contribute to the scarcity of simultaneous spatial‐social network analyses in disease ecology thus far. Here, we detail scenarios in disease ecology that benefit from spatial‐social analysis. We describe procedures for simultaneous collection of both spatial and social data, and we outline statistical approaches that can control for and estimate spatial‐social covariance in disease ecology analyses. We hope disease researchers will expand social network analyses to more often include spatial components and questions. These measures will increase the scope of such analyses, allowing more accurate model estimates, better inference of transmission modes, susceptibility effects and contact scaling patterns, and ultimately more effective disease interventions. Social network analysis is increasingly widespread in disease ecology, but is often carried out without investigating spatial processes. The authors describe how uniting spatial and social network analyses can augment disease ecology investigations, outlining tools and methodology to do so.