Re: interpretting F & t-vals from regression


<snip>
duration500:intensity 0.008720 0.0533651 165 0.163403 0.8704
duration1000:intensity 0.006706 0.0533651 165 0.125671 0.9001

That indicates that "duration 1000" is the "most useless" in the
presence of other variables in the model.

REMOVE IT from your model, and see how the rest of them change.

Thankyou Bob, I appreciate your comments; I now understand what I'm doing a
lot better.

I've skimmed the posts and I know a small amount about these issues already
but I'm sure I should learn more:
I'm rather new to this game. Your information made me realise that I was
approaching the model building incorrectly rather than misinterpreting
coefficients. For anyone who's interested, here's my take on the situation
now.

You're right that the model was over-specified but I don't think removing
intensity1000 is possible. This is a designed experiment and in the context
of this analysis, that variable is needed to explain what subjects did at
sound durations of 1000 ms.

The model I was fitting wasn't intended to be predictive. I was coding
duration as a factor because I naively wanted to know what happened at each
duration. I suppose this is a lazy way of fitting a separate regression to
each subject at each duration. But this is hardly a parsimonious model.
Under these circumstances I suspect that the F-test is not meaningful other
than to tell me that I had too many parameters. To deal with this, I've run
the model again but with intensity as a continuous variable. This test
gives me one duration:intensity interaction which is not significant:
everything now ties together and is, I think, sensible.

p.s. I know this analysis, as posted to the Group, looks a little pointless
but it's actually part of a bigger test...



--
remove ferret to reply
.

Relevant Pages

  • Re: Project allocating duration/resources
    ... If Bill, Joe, and Bob all will do 8 hours of work ... >> and all of them are working on Friday, the duration is one day. ...
    (microsoft.public.project)
  • Re: Resource error message
    ... The message you note is caused when you attempt to assign a resource ... If the resource has no working time in the original task duration, ... calendar and see what working and non-working time is defined. ... let's call him Bob (I have ...
    (microsoft.public.project)
  • Re: Project allocating duration/resources
    ... What about the other two - what does their calendar say about ... If Bill, Joe, and Bob all will do 8 hours of work each ... and all of them are working on Friday, the duration is one day. ...
    (microsoft.public.project)
  • RE: Task scheduling with time to spare
    ... In the bottom pane, key all the information-- task name, change task type to ... Change the duration to anthing desired such as 3 days or 5 ... Bob could work on the task all day Wednesday ... This task as part of an overall project schedule and it's a predecessor ...
    (microsoft.public.project)
  • Re: Project allocating duration/resources
    ... If Bill, Joe, and Bob all will do 8 hours of work ... > and all of them are working on Friday, the duration is one day. ... > calendar for Bill and Joe says they work M-F while the calendar for Bob ...
    (microsoft.public.project)