Reader Comments

Post a new comment on this article

Analysis question

Posted by eblomberg on 06 Dec 2014 at 01:10 GMT

Wolf populations have increased substantially throughout the west over the past 25 years. The results presented in this article seem to suggest that wolf abundance (indexed by number of pairs) are related to depredation rates. One would assume that increases in depredation rates are also correlated with the number of wolves that are culled in a given year. Following this train of thought, livestock depredation rates and the number of wolves culled have all probably increased through time as wolf abundance has increased. This would suggest that livestock depredation and wolf culls are likely to be serially autocorrelated (because both are ultimately driven by wolf abundance), which could confound the apparent relationship between livestock depredation in time t and wolf abundance in time t-1. I didn't see anywhere in the article where this subject was addressed? I was wondering if the journal or the authors care to comment?

No competing interests declared.

RE: Analysis question

eblomberg replied to eblomberg on 11 Dec 2014 at 03:06 GMT

What I meant to say above was that the apparent relationship between livestock depredation and wolf culling could be confounded by the collinear nature of wolf abundance, livestock depredation, and wolf culling. I am not sure if that was clear in what I wrote.

No competing interests declared.

RE: RE: Analysis question

davemech replied to eblomberg on 05 Mar 2015 at 16:30 GMT

Enter your comment...
You are exactly correct. Most of the findings in this article could more parsimoniously be attributed to the collinearity you described, rather than the conclusion that killing wolves increases livestock depredations. This erroneous conclusion is a great example of the common fallacy of misconstruing correlation as causation.

No competing interests declared.

RE: RE: RE: Analysis question

nmackey replied to davemech on 04 Jun 2021 at 00:20 GMT

Where exactly is this collinearity documented, in the absence of the wolf mortality? The only study I have seen supporting this was Michigan-based, and is quite undescriptive of the extremely low livestock mortality that the last few years have seen - on the order of a calf or hunting dog per 100 wolves per year, despite - or because of? - a virtual total lack of hunting and trapping of wolves in the Upper Peninsula. It becomes harder and harder by the year to use that "Michigan study" as a façade.

No competing interests declared.