Hey all. You can find me at the other place instead. Same username and all that.

New blog post sharing updates on the latest version of the Diel.Niche R package. Interested in quantifying if a species is nocturnal with data? Well it is easier than ever now! #rstats #cameratrapping and now back to the other place... https://t.co/9R38RH4JiR

Making a BIG built in dresser in my bedroom and finally got all the drawers installed this weekend. More to do, but at least clothes can go in there now.

Here are some thoughts on teaching #statistics to non-statistical students, specifically those in #applied #ecology degrees. Thanks to @willpkay and @TiagoALOMarques for an enjoyable (and not last) collaboration. request pdf here 👉 https://t.co/7oDAM68DJG

Our preprint "Calibration, sensitivity and uncertainty analysis of ecological models -- a review" with Anne Malchow now available at https://t.co/i5KvFbsAkc This is our attempt at giving a broad overview about methods and issues surrounding

In this thread, @masonfidino mentions a similarly themed pre-print of ours. All the credit for this excellent work, which explores sample size requirements of the 'Rota co-occ' model, should go to the dynamic (and amazing) duo @AmberCowans and @abonetb 👇 https://t.co/lczU2HGfXC

I’m hiring for a super cool and super important role @calacademy I’m looking for a leader to work with our team to expand biodiversity community science in California!

@AmberCowans @chrissuthy Kinda sucks though that the suggestion may be logistically unfeasible for many. Fortunately, the autologistic representation of this model may get around that, check out this paper from Kass et al. (and I'm working on a blog post about it as well). https://t.co/4lJETLiCXq

@AmberCowans @chrissuthy As someone who has spent a great deal of time with this model (I generalized it to a dynamic model for my PhD and helped write some of the AHM volume 2 chapter on co-occurrence modeling), it's nice to see some solid suggestions to help guide study design!

While working on this we also found out that @AmberCowans was cooking up something similar with @chrissuthy! Overall we found complimentary results (i.e., you need data from hundreds of sites to estimate co-occurrence reliably).

How much data do you need to reliably estimate statistical interactions with the Rota. et al. co-occurrence occupancy model? Turns out A LOT. Check out this pre-print led by @LouvrierJulie! https://t.co/1OYkfIIFtX

Excited to bring Joint Species Distribution Models to {mvgam}: https://t.co/u8DLCsedvT. Throw any of {mgcv}'s or {brms}'s flexible predictor effects into the factor or observation models and give it a whirl #rstats #communityecology

Anyways, I'm waxing poetic at this point. But yeah, seeing that my blog is helpful to people really means a lot. So thanks to all of you who came up and said hi at TWS!

Certainly, using more complex statistical models does not mean you are doing better science. But misusing or misinterpreting complex statistical models is a surefire way to do bad science.

As the field of ecology is increasingly using more complex statistical methods, one thing I've always wanted to do is try to make them more approachable and remove barriers people may have to use them. That is the niche that my blog is trying to fill.

With the decline of scientists using this platform (and the fact that it seems like the algorithm for sharing stuff has changed here), I kinda felt like I was blogging to the void, so I stopped. It was a huge surprise to me learn otherwise!

Why did I decide to do these again? At the Wildlife Society Conference a couple weeks ago people thanked me for writing them EVERY SINGLE DAY. I had no clue people found them useful!

I decided to restart my ecological stats blog posts. Check out this new one on why you should probably stop using the number of survey days as a covariate in occupancy models. Check it out! https://t.co/JF83LLKpdD

I got to be the ‘there must be a more logical explanation’ scientist you often see in horror movies for this article. This was such a fun interview to do!

Doing research out of your field can be fun, but this is the first time I've really encountered a review from someone with an econometrics background and wow there is so much focus on how you we should be doing causal inference with observational data to make valid inference.