Choosing a sensible cut-off point: assessing the impact of uncertainty in a social network on the performance of NBDA

• DOI: 10.1007/s10329-018-0693-4
• Publication Date: 2018-10-09T12:56:56Z
• Authors: Sonja Wild; William Hoppitt
• Keywords: Animal ecology; Social network (sociolinguistics); Social Learning; Social Network Analysis

Network-based diffusion analysis (NBDA) has become a widely used tool to detect and quantify social learning in animal populations. NBDA infers if the spread of novel behavior follows network hence relies on appropriate information individuals' connections. Most studies populations, however, lack complete record all associations, which creates uncertainty network. To reduce this uncertainty, researchers often use certain threshold sightings for inclusion animals (which is arbitrarily chosen), as observational error decreases with increasing numbers observations. Dropping individuals only few sightings, can lead loss connecting are removed. Hence, there trade-off between including many possible having reliable data. We here provide R that assesses sensitivity given individuals. It simulates process through population then tests power reliably after introducing into network, repeated different thresholds. Our help using select threshold, specific their data set, maximizes study population.