Skip to main content

Data from: Visualizing connectivity of ecological and evolutionary concepts – an exploration of research on plant species rarity

Cite this dataset

Boyd, Jennifer et al. (2020). Data from: Visualizing connectivity of ecological and evolutionary concepts – an exploration of research on plant species rarity [Dataset]. Dryad.


Understanding the ecological and evolutionary factors that influence species rarity has important theoretical and applied implications, yet the reasons why some species are rare while others are common remain unresolved. As a novel exploration of scientific knowledge, we used network analysis conceptually to visualize the foci of a comprehensive base of >800 studies on plant species rarity within the context of ecology and evolution. In doing so, we highlight existing research strengths that could substantiate novel syntheses and gaps that could inspire new research. Our results reveal strong integrated foci on population dynamics with other ecological concepts. In contrast, despite the potential for ecological and evolutionary processes to interact, few studies explored the interplay of environmental factors and microevolutionary patterns. The cellular and molecular biology, physiology, and plasticity of rare plant species within both ecological and evolutionary contexts similarly provide avenues for impactful future investigations.


Visualizing connectivity of ecological and evolutionary concepts –  literature base, keywords & concepts, matrices

We used ISI Web of Science (Thompson Reuters, New York, NY) to screen the primary literature for studies of plant species in which their rarity or characterization as rare was a focus of the research. We curated 813 relevant articles and extracted their keywords. To capture the research foci of this literature base, we developed a network of broad 'parent' nodes and more specific 'child' nodes that represented ecology and evolution comprehensively as broad fields. We assigned each keyword to a parent node and child node (if applicable) based on general agreement among the authors. Titles, keywords, and abstracts from our curated base of 813 articles were searched with a Perl script to identify the number of articles that focused on each pairwise combination of concepts. Diagonal matrices were used to organize results according to the total number of studies that were identified for each pairwise concept combination. Connectivity values were calculated  as the ratio of the number of studies that focused on a given pair of concepts to the total keywords relevant to those two concepts. The median of all connectivity values was calculated and the importance of each pairwise connection was considered relative to the median.


National Science Foundation, Award: 1655762

University of Tennessee System, Award: Chattanooga, Center of Excellence in Applied Computational Science and Engineering