Analysing social network data is difficult. How do you reflect uncertainty in your measurements when all you have is a single value? E.g., this network's density is X
And even if you have quantified uncertainty, how do you pull that uncertainty into your downstream statistical analyses?
E.g., into, say, a statistical model that compares the density of network A to the density of network B
Together with colleagues, members of my research group have been working to try and solve these and other common challenges in social network analysis. This includes BISoN, a new framework for the analysis of network data, and its associated R package, bisonR. BISoN allows users to quantify uncertainty in the network measurements (be they topological, at the level of the individual, etc) and carry these through to, say, a linear regression that contains other (non-network) predictors and confounds.
And even if you have quantified uncertainty, how do you pull that uncertainty into your downstream statistical analyses?
E.g., into, say, a statistical model that compares the density of network A to the density of network B
Together with colleagues, members of my research group have been working to try and solve these and other common challenges in social network analysis. This includes BISoN, a new framework for the analysis of network data, and its associated R package, bisonR. BISoN allows users to quantify uncertainty in the network measurements (be they topological, at the level of the individual, etc) and carry these through to, say, a linear regression that contains other (non-network) predictors and confounds.
BISoN-related and other publications on social network methodologies:
Hart JDA, Weiss MN, Franks DW, Brent LJN (2023). BISoN: A Bayesian framework for inference of social networks. Methods in Ecology and Evolution. doi.org/10.1111/2041-210X.14171. Hart JDA, Franks DW, Brent LJN, Weiss MN. bisonR – Bayesian inference of social networks with R. Prepint - https://osf.io/ywu7j/ Hart JDA, Weiss MN, Brent LJN, Franks DW (2022). Common permutation methods in animal social network analysis do not control for non-independence. Behavioural Ecology & Sociobiology. 76, 151. doi.org/10.1007/s00265-022-03254-x. Hart JDA, Franks DW, Brent LJN, Weiss MN. Why datastream permutations need diagnostics. Preprint - https://osf.io/xkvcu Weiss MN, Franks DW, Brent LJN, Ellis S, Silk MJ, Croft DP (2020) Common datastream permutations of animal social network data are not appropriate for hypothesis testing using regression models. Methods in Ecology and Evolution. doi:10.1111/2041‐210X.13508. Hart JDA, Franks DW, Brent LJN, Weiss MN (2022). Accuracy and power analysis of social networks built from count data. Methods in Ecology and Evolution. doi.org/10.1111/2041-210X.13739. |