Installing the wiqid package

There is an introduction to wiqid here . The package is still getting major changes, meaning that your old scripts may not work with new versions of wiqid .

Some of the simpler Bayesian MCMC functions in wiqid are coded in R, others use JAGS and the rjags package.

Installing wiqid

wiqid is now on CRAN !

You can now install wiqid and the packages it requires from within R with the command install.packages("wiqid") . You can also go to the Packages > Install package(s)... menu item and select wiqid from the pop-up list, but with more than 11,000 packages on the list, I find typing easier.

After installing wiqid

For full functionality, you will need to have JAGS installed, as well as the rjags, secr and shiny packages.

1. JAGS: If you are using Mac or Ubuntu, check out the notes here or here . Go to the JAGS download page . A link at the top of the page gives you the latest installer; on Windows, this takes you to an HTML page with a choice of installers, depending on the version of R you are using. Install in the default directory.

JAGS 4 was released in October 2015; to use JAGS 4 you need version of rjags_4-4 or later.

2. rjags : Once you have installed JAGS, you can install the rjags package in R with

install.packages("rjags")

Check that it has been installed correctly and can link to JAGS by doing

library(rjags)

3. secr and shiny : You can install these in R with

install.packages(c("secr", "shiny"))

Both these packages depend on a number of other packages, and they will also be downloaded automatically.

Latest devel version

wiqid is hosted on the Github repository here . Changes will appear on Github before being made in the version released on CRAN. if you want to install the devel version, install the R package githubinstall then use:

githubinstall::githubinstall("wiqid")

As usual, devel versions should work properly but have not been fully tested: use with care!

The wiqid package sits within a broader ecosystem of R tools designed to make wildlife data analysis more approachable for researchers and students. It bundles together routines that bridge classical capture-recapture methods with Bayesian alternatives, allowing users to move between frequentist and probabilistic frameworks without leaving a familiar interface. Because it relies on well-established packages for numerical work, it fits naturally into existing workflows rather than replacing them. This makes it a practical starting point for anyone exploring modern approaches to animal population studies.

Installing the package has become considerably easier as the ecosystem of statistical software continues to mature. The standard repositories now host it alongside thousands of other contributed extensions, so users do not need to search for separate download locations or resolve dependencies by hand. Once installed, it loads alongside the rest of a user's R session and remains available across sessions. Keeping the package updated through the same channels ensures that bug fixes and new features reach users promptly. This kind of centralised distribution model reduces friction for newcomers.

The functions included cover a range of tasks that biologists commonly face when working with survey data. Detection probabilities, occupancy estimates, and density calculations form the core of the analytical toolkit, reflecting methods that have become standard in the field. Many routines accept covariates, allowing habitat or effort variables to be incorporated into models. By providing implementations of both single-season and more elaborate designs, the package accommodates studies of varying complexity. Users can therefore tailor their analyses to match the scale and ambition of their fieldwork without switching software environments.

Bayesian methods feature prominently throughout, reflecting a wider shift in how uncertainty is quantified in ecological research. Markov chain Monte Carlo techniques underpin many of the more advanced routines, with options to adjust chain length, thinning, and starting values. For those who prefer simpler approaches, conjugate distributions offer closed-form solutions that run almost instantly. This blend of computational strategies means users can match the method to the question at hand, balancing rigour against the need for timely answers. The flexibility encourages experimentation and careful model checking.

Documentation and introductory material accompany the code, helping users build intuition about the underlying statistical ideas. Worked examples walk through realistic scenarios, showing how raw detection histories translate into interpretable parameter estimates. Plots and summary tables are produced in standard formats that drop easily into reports or manuscripts. Because the package is maintained by someone who also teaches workshops, the design reflects common questions and confusions encountered by learners. This teaching-oriented ethos gives the software a welcoming character for those new to the field.