D. W. Franks

Group size modulates kinship dynamics and selection on social traits

This study shows that group size affects kinship dynamics and age-related social behaviors in social mammals, with smaller groups promoting more extreme helping/harming behaviors, especially in younger individuals.

Bayesian inference of toothed whale lifespans

This study develops a Bayesian mortality model to estimate lifespans and age-specific mortality in toothed whales using age-structured data from mass-strandings and other sources.

The Evolution of Democratic Peace in Animal Societies

This study applies evolutionary game theory to test the democratic peace hypothesis across animal societies, finding support for the idea that collective decision-making reduces intergroup conflict.

BISoN: A Bayesian Framework for Inference of Social Networks

BISoN provides a Bayesian framework to quantify uncertainty in social network edge weights and supports downstream analyses using observational data.

Common Permutation Methods in Animal Social Network Analysis Do Not Control for Non-independence

This paper shows that common permutation methods in social network analysis do not properly account for non-independence, suggesting parametric models may be more reliable.

Accuracy and Power Analysis of Social Networks Built From Count Data

The paper develops a method to estimate network correlation and power analysis for social networks built from count data, showing that network correlation impacts statistical power and providing an R package for implementation.

Calculating effect sizes in animal social network analysis

The paper demonstrates that relying solely on permutations to control for confounding variables in social network analysis can lead to biased and uninterpretable effect sizes, and recommends controlling for confounders in statistical models.

Common datastream permutations of animal social network data are not appropriate for hypothesis testing using regression models

The paper shows that common datastream permutations are not suitable for testing regression models in animal social networks due to high false-positive rates.