Imperfect detectability in animal studies has been an acknowledged problem for several decades. A number of analytical approaches (e.g., capture–mark–recapture, distance sampling, occupancy analyses), have been developed to deal with the problem and estimate biological parameters of interest. These different analytical approaches can be implemented through various software programs, each with its own graphical user interface (GUI), data input format, and graphing capabilities. Although GUIs and mouse clicks might be convenient to conduct analyses, mouse clicks are difficult to replicate and archive in the long term. Using programming environments such as the R open source project along with a text editor yields flexibility and leaves a paper trail after the analyses are conducted. Such environments also provide statistical and graphical tools for further treatment following the analyses. In this article, I use R to conduct classical analyses to estimate demographic parameters in four case studies. The examples feature interacting with certain software such as MARK, as well as using recent R packages designed to implement specific analyses in both a frequentist and a Bayesian framework. The case studies present all the steps to conduct the analyses, including data importation, model selection, and multimodel inference, as well as the graphical presentation of results. Such an integrative approach provides a flexible alternative to using GUIs, while keeping everything in the same environment and documenting precisely how an analysis was conducted. These are the basic ideas behind reproducible research.

You do not currently have access to this content.