I was invited to contribute to two papers in a recent issue of the Philosophical Transactions of the Royal Society A. Yes, you read that right, Transactions A: Mathematical, Physical and Engineering Sciences. While the B series (Life Sciences) is the more familiar stamping ground of ecologists and evolutionary biologist, this issue on Statistical Workflows, edited by Andrew Gelman, Aki Vehtari, Richard McElreath and Elizabeth Wolkovich, is highly relevant to ecologists, especially as our datasets have grown and our statistical analyses have got ever more complex. In these two papers, we have tried to keep the math to a minimum and emphasise the core concepts of developing robust (and reproducible) statistical workflows, starting from the research question:
A four-step simulation-based workflow for ecological analysis and science
Closing the gap between statistical and scientific workflows for improved forecasts in ecology
A major theme is simulating data from your (statistical) model as a routine and integral part of the workflow. As a bonus, the journal cover photo (supplied by Max Farrell) comes from our camera trapping work in the Kruger National Park)