tag:blogger.com,1999:blog-3522311.post6607014737196837499..comments2024-03-12T03:23:42.976-04:00Comments on NeuroDojo: Should sprinting shape scorpion’s stingers?Zen Faulkeshttp://www.blogger.com/profile/07811309183398223358noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-3522311.post-69087199380880470272012-01-11T20:09:54.753-05:002012-01-11T20:09:54.753-05:00Good point, although one could always argue that t...Good point, although one could always argue that the same thing could happen with p=0.04 - a few more data points may change everything. I agree with you that confidence intervals are probably the way to go. All the benefits of p-values but much more useful.Bradley Carlsonhttps://www.blogger.com/profile/16362855916687257328noreply@blogger.comtag:blogger.com,1999:blog-3522311.post-29337399720750956922012-01-11T16:07:03.726-05:002012-01-11T16:07:03.726-05:00Thanks for your comments!
On the p values, I'...Thanks for your comments!<br /><br />On the p values, I've had cases where early p values were (say) 0.06, but that that went, when I added a few more data points, the "wrong" way to something not even close to statistically significant. <br /><br />I'm becoming more convinced that confidence intervals are a better way of conveying the degree of certainty or not.<br /><br />Would be happy to hear of any follow-ups!Zen Faulkeshttps://www.blogger.com/profile/07811309183398223358noreply@blogger.comtag:blogger.com,1999:blog-3522311.post-22793221237652510652012-01-11T14:40:37.885-05:002012-01-11T14:40:37.885-05:00Sorry to comment over a year after your initial po...Sorry to comment over a year after your initial post - I just found it now by chance. I'm the Carlson in the paper, which was research I performed as an undergrad. I appreciated your comments (even the critical ones) and interest. I will point out 2 quick things to address a couple of your concerns: 1) we didn't test dehydration effects on stinging because the dehydration experiment was an afterthought that turned out to be the most interesting element. Following it up would be interesting. 2) We were cautious about our interpretation of sex effects on sting rates. The p-value was 0.07, and of course 0.05 is an arbitrary cut-off point - is it completely meaningless to say that there is a 93% chance that there is a difference between sexes? I think too many interesting biological patterns may be missed by dogmatic understandings of what really amounts to "rules of thumb".<br /><br />That being said, I wanted to let you know that I am following up this work to explicitly look at the effects of sex, mass, limb length, and tail morphology on stinging and sprinting to better tease apart what is going on here. My newest suspicion - females are actually just more aggressive. If you want, I can let you know when I've got some results. Best, BradBradley Carlsonhttps://www.blogger.com/profile/16362855916687257328noreply@blogger.com