[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]
re: Trailestes cherry picking
The more evidence the merrier, Tim. The only cherry-picking I've done, I
confess, is avoiding various foramina as characters, other than the big pineal,
on the grounds that very few references include them. Very few braincases are
exposed. And I've seen many cases where some openings are considered damage or
skull rot. With regard to Trialestes, if I have just a little, I'll use it. If
I have more (which is why I requested pix) I'll use more.
I've seen cherry-picking at its best when dissecting the matrices of others.
So, I'm learning. Aren't we all?
That's why I'm here.
If anyone has any trouble with the way I use PAUP, I'd be glad to hear about it
and share files. I'm learning as I go. So far I've learned to avoid polytomies,
avoid using ingroup taxa at the base of the cladogram and avoid small studies
of wide gamuts.
All the best,
David Peters wrote:
That's what I like to see. Corroborating evidence.
I know what you mean here - it's always good to receive affirmation of one's
own hypotheses. You get a nice warm fuzzy feeling inside.
However, I'd advise you to be very careful here. You could be of accused of
'cherrypicking'. Just because you find corroborating evidence (or what you
interpret as corroborating evidence), doesn't necessarily make you correct.
Look at *all* the evidence - "corroborating" and otherwise. The best hypotheses
are those that stand up to challenges, instead of pretending that such
challenges don't exist. It's much more important to account for contrary data
than simply looking for corroborating data. In other words, don't restrict
yourself to studies that (you think) back you up.
Would love to see some pix. Thanks for the reference Tim!
You're welcome. But I would strongly advise you to do some background reading
on the fundamentals of cladistics. I'll add to the chorus of DMLers who have
noted (or strongly implied) that your grounding in cladistics appears to be