[R] [r] ANCOVA method
gunter.berton at gene.com
Fri Dec 3 17:23:13 CET 2010
1. You need to seek local statistical help.
2. The answer to your question is: it depends in how you define
"influence significantly." If you define it as "the interaction term
is significant" then, by definition the answer is yes. If you want to
understand what is going on and make meaningful scientific statements,
then the answer is no. At the least, you need to plot your data
informatively. To determine what this means, see (1) above.
On Fri, Dec 3, 2010 at 8:13 AM, Francesco Nutini
<nutini.francesco at gmail.com> wrote:
> Dear [R] Users,
> I have implemented a linear model with this syntax:
> model<- lm (var_dependent ~ var_indipendent + factor + var_indipendent : factor, dataframe)
> anova (model)
> Response: var_dependent
> Df Sum Sq Mean Sq F value Pr(>F)
> var_indipendent 1 20.5522 20.5522 87.8701 1.167e-14 ***
> factor 1 0.1060 0.1060 0.4530 0.50277
> var_indipendent:factor 1 1.3861 1.3861 5.9261 0.01706 *
> Residuals 83 19.4132 0.2339
> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
> If I read the line "var_indipendent:factor" can I understand if the factor influence significatvly the regression between dependent-indipendent variable?
> Thanks a lot!
> Francesco Nutini
> P.S. numbers have no significance, it's just an example
> [[alternative HTML version deleted]]
> R-help at r-project.org mailing list
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
Genentech Nonclinical Biostatistics
More information about the R-help