STRAND is an R package designed for simulation and analysis of network data. The package can be used to simulate network data under stochastic block models, social relations models, and latent network models. These tools allow for simulation not only of realistic human and animal social networks, but also allow researchers to simulate the effects of potential biases—like respondents falsely reporting ties or failing to recall real ties—on network level properties.
STRAND also allows users to conduct data analysis. Users can specify complex Bayesian social network models using a simple, lm() style, syntax. Single-sampled self report data can be modeled using stochastic block models or the social relations model, with or with-out covariates. Double-sampled network data can be modeled using a latent network approach that accounts for inter-respondent disagreement. Here we provide a brief overview of a workflow. For further details on any step, see our full publications at Pyschological Methods (for latent network models) and at Journal of Animal Ecology (for simple network models).
STRAND is part of an ecosystem of tools for modern social network analysis. DieTryin is a companion package designed to facilitate the collection of roster-based network data, and to run network-structured economic games. XLSFormulatoR is a package for automatically building name-generator network surveys for KoboToolbox.
Install by running on R:
################################### Install and/or load
library(devtools)
install_github('ctross/STRAND')
library(STRAND)
You will need to have cmdstanr and rstan installed.
Quickstart guides for these packages can be found here, for cmdstanr and here, for rstan.
STRAND calls Stan models in the background, so you will need a C++ compiler in your toolchain. Users frequently rely on RTools.
Example models, with test data, can be found here. Note that analysis of large network data sets can be quite slow using MCMC.
Binary/Bernoulli outcomes (runs in about 2 hours): Bernoulli models
Binomial outcomes (runs in about 2 minutes): Binomial models
Binomial outcomes with measurement error (runs in about 20 minutes): Binomial + ME models
Poisson outcomes (runs in about 20 seconds): Poisson models
Undirected networks (runs in about 20 minutes): Undirected networks
Double-sampled binary outcomes (find out for yourself): Latent network models
Multiplex Binary/Bernoulli outcomes (runs in about 3 hours): Multiplex Bernoulli models
Multiplex Binomial outcomes (runs in about 30 minutes): Multiplex Binomial models
Multiplex Poisson outcomes (runs in about 4 hours): Multiplex Poisson models
An example on both simulating and fitting networks (includes interactions): Simulating data and fitting interaction models
An example on both simulating and fitting multiplex networks: Simulating data and fitting multiplex models
An example on accounting for structural-zeros/censored-data. For example, there may be N groups of individuals in a dataset, where each group is in a separate enclosure, and thus only with-group ties can be modeled: Structural Zeros.
An example on changing default priors: Changing priors
Probit versus logit links for binary outcomes: Does anything depend on the choice of link function?
Probit versus logit links for multiplex binary outcomes: Does anything depend on the choice of link function?
Each of the models included in this package have been fit to real emprical datsets, and tested across a wide-range of simulated data to ensure their quality. However, this package is still rather new. If you come across any weird behavior, or notice any bugs, please open an issue, and we will work to address it!