Re: reporting categories of continuous variable in a regression model
- From: Jerry Dallal <gdallal@xxxxxxxxxxxxxxxx>
- Date: Mon, 27 Mar 2006 14:11:27 GMT
RCMR01@xxxxxxxxxxx wrote:
Thanks for your reply.
Sorry, for the confusion. AGE is an independent variable and is
significantly associated with the dependent variable (Y) in my model. I
could just report the regression coefficient of the variable AGE
because this represents the increase in Y for a 1 year increase in age.
However, this is not very meaningful and I would like to report the
values of Y by different age categories in my report. Any suggestions?
Thanks.
Raoul
I can imagine this question coming up in many contexts, but not this one. Presumably, the response makes some sense to your audience or there's nothing you could report that would be understandable.
How can difference in response *per year of difference in age* not be meaningful?! If it's because the difference in response is small per one year difference in age, one has the option to multiple by 10, say, to give the difference in response corresponding to a difference in an age decade, or something similar.
Also,
Be careful not to confuse "difference in age" (cross-sectional) with "increase in age" (longitudinal). I'd be willing to bet that the data are cross-sectional or, if longitudinal, not modeled correctly, given the nature of the questions.
For the purposes of display, epidemiologists are fond of dividing data into thirds/fifths/tenths/whatever, reanalyzing, and constructing plots of adjusted means to give their audience a feel for what the numbers "look like". Every so often, this exercise reveals that the response is not, in fact, linear in the predictor under the microscope.
.
- References:
- reporting categories of continuous variable in a regression model
- From: RCMR01@xxxxxxxxxxx
- Re: reporting categories of continuous variable in a regression model
- From: Reef Fish
- Re: reporting categories of continuous variable in a regression model
- From: RCMR01@xxxxxxxxxxx
- reporting categories of continuous variable in a regression model
- Prev by Date: Re: multicollinearity in regression
- Next by Date: Re: multicollinearity in regression
- Previous by thread: Re: reporting categories of continuous variable in a regression model
- Next by thread: differentiate between certain metropolitan areas
- Index(es):
Relevant Pages
|
