Re: interpretting F & t-vals from regression
- From: Rob Campbell <news@xxxxxxxxxxxxxxxxxxxxxxxxxx>
- Date: Thu, 23 Mar 2006 21:40:49 GMT
First of all, just because you designed some level in your study
doesn't
imply what you said is correct! If a level within your design turned
out to be useless in the light of observed data, then it's a RESULT of
your design!
The removal of any redundant variable is always POSSIBLE and appropriate.
The fact that I see a zero slope for intensity at certain durations is, as
you say, a result. Potentially it might be an important result. I don't
understand what you mean by "removing" it. Removing it from what? The
model? Which duration levels I see a slope and which levels I do not is
something I'm interested in. That is why I have now taken the approach of
fitting separate regressions to these data at each duration and reporting
the coefficients and their significance. That makes sense, surely? Where
does the removing come into it.
That's a small step toward porgress. :-) But I think you're still
missing quite a bit -- in both the design and the analysis.
Bob, your comments are helpful and knowledgable and I appreciate them. I'm
quite aware that I have more to learn. If you think I'm missing things over
and above what we have discussed directly, being more specific would be
helpful.
--
remove ferret to reply
.
- References:
- interpretting F & t-vals from regression
- From: Rob Campbell
- Re: interpretting F & t-vals from regression
- From: Brett Magill
- Re: interpretting F & t-vals from regression
- From: Rob Campbell
- Re: interpretting F & t-vals from regression
- From: Reef Fish
- Re: interpretting F & t-vals from regression
- From: Rob Campbell
- Re: interpretting F & t-vals from regression
- From: Reef Fish
- interpretting F & t-vals from regression
- Prev by Date: Re: interpretting F & t-vals from regression
- Next by Date: Re: interpretting F & t-vals from regression
- Previous by thread: Re: interpretting F & t-vals from regression
- Next by thread: Re: interpretting F & t-vals from regression
- Index(es):
Relevant Pages
|