Correlates of research effort in carnivores: body size, range size and diet matter

Abstract

Given the budgetary restrictions on scientific research and the increasing need to better inform conservation actions, it is important to identify the patterns and causes of biases in research effort. We combine bibliometric information from a literature review of almost 16,500 peer-reviewed publications on a well-known group of 286 species, the Order Carnivora, with global datasets on species' life history and ecological traits to explore patterns in research effort. Our study explores how species' characteristics influenced the degree to which they were studied (measured as the number of publications). We identified a wide variation in intensity of research effort at both Family and Species levels, with some of the least studied being those which may need protection in future. Our findings hint at the complex role of human perspectives in setting research agendas. We found that better-studied species tended to be large-bodied and have a large geographic range whilst omnivory had a negative relationship with research effort. IUCN threat status did not exhibit a strong relationship with research effort which suggests that the conservation needs of individual species are not major drivers of research interest. This work is the first to use a combination of bibliometric analysis and biological data to quantify and interpret gaps in research knowledge across an entire Order. Our results could be combined with other resources, such as Biodiversity Action Plans, to prioritise and co-ordinate future research effort, whilst our methods can be applied across many scientific disciplines to describe knowledge gaps.

Figure 1. The number of Carnivore papers published per year from 1900–2010.

A number of notable dates are also shown.

Figure 2. The number of Carnivore papers published per year from 1900–2010 for species of Least Concern and at-risk species (IUCN Near Threatened, Vulnerable, Endangered and Critically Endangered).

The barplot shows the total number of papers per year.

Figure 3. Analysis of model residuals: Predicted Vs Observed values.

The line indicates where the predicted values  =  observed values i.e. the model is a perfect fit to the data. Species which receive more research attention than our model predicts based on their biology are above the line whilst those receiving less are below the line. Species with residuals greater than +2 or −2 are labelled.