Density Dependence Without Resource Partitioning: Population Ecology on Change.org

E-petitioning is a prominent form of Internet-based collective action. We apply theories from organizational population ecology to investigate whether similar petitions compete for signatures. We use latent Dirichlet allocation (LDA) topic modeling to identify topical niches. Using these niches, we test two theories from population ecology on 442,109 Change.org petitions. First, we find evidence for density dependence, an inverse-U-shaped relationship between the density of a petition's niche and the number of signatures the petition obtains. This suggests e-petitioning is competitive and that e-petitions draw on overlapping resource pools. Second, although resource partitioning theory predicts that topically specialized petitions will obtain more signatures in concentrated populations, we find no evidence of this. This suggests that specialists struggle to avoid competition with generalists.