RAZD writes:
As yet there is questionable evidence that something exists (ie - footprints, reports of sightings), and now there is possible evidence of a bear related distantly to polar bears.
...
So it seems to me that a good "working hypothesis" is that the yeti is a bear ...
I think this is the problem. Right there in your last sentence.
The 'good working hypothesis' is not that the yeti is a bear.
The 'good working hypothesis' is that a bear exists in the area.
Notice how the good working hypothesis doesn't mention a yeti.
That's specifically because there is no evidence as of yet that points towards a yeti.
Science doesn't start with a conclusion and then form hypothesis from it:
"Let's look for a Yeti!!"
"Okay... first hypothesis... a Yeti would leave evidence..."
"We found some evidence... looks like a bear..."
"Next hypothesis... the yeti is a bear...."
This is horrible science. This is creation-science. This is science led by it's pre-made conclusion, not by it's evidence.
Good science goes from the data and moves from there... with no leading-pre-made-conclusion guiding it. Like this:
"Let's look for a Yeti!!"
"Why?"
"Because people say a Yeti exists!!"
"Do we have any evidence?"
"No."
"Okay... first hypothesis is that
something might exist... any something would leave evidence, let's go look.
"We found some evidence... looks like a bear..."
"Next hypothesis... a bear exists in the area..."
Notice how "Yeti" is left out of the hypothesis/science portion because there's no evidence for a Yeti.
Once you have actual "Yeti-unique-evidence" (evidence of a new species that is ape/man-ish)... then the evidence will be leading you towards a Yeti and you can start using "Yeti" in your scientific analysis.
Pre-maturely using the word "Yeti" in your science is allowing your pre-made conclusion to lead your work.
Good science lets the evidence lead. If there's no evidence of a Yeti... then there's no mention of a Yeti in the science.