I'm replying, no, there's nothing dubious about it. Which doesn't mean it's accurate. Biology has always prominently featured observation from life, and this is, in general, how the field first starts to suspect that, say, bees can sense electric fields.
Experiments on crabs might and should follow, but your skepticism isn't grounded in anything here.
Yes, exactly, because at least counting the crabs and comparing the count with other clusters of crabs would show whether there 'is something'. Missing accuracy is not my problem, it's missing 'there is something' vs. 'there is nothing'. Without any statistical data, there is nothing to talk about, not in biology either, which is also science.
Skepticism is very important. I challenge them to count crabs on cables and compare with crab counts not on cables, and then make a case. That's all.
This is the follow-up experiment.
https://www.mdpi.com/2077-1312/9/7/776
As usual with modern science journalism, it's necessary to click through and read the actual research paper. Otherwise, one will be totally lost.
As for samatman's concerns, the paper uses standard experimental controls and statistical analysis, as expected.