# [R] Multiple regression intercept

Daniel Malter daniel at umd.edu
Mon Sep 12 22:54:27 CEST 2011

```This suggests that this is a dangerous office to be in because this is a
basic question. I am sure somebody in your office knows this. Anyway, the
baseline gives you the average value of the group that constitutes the
baseline when all other covariates are zero. Let's say you measure whether
men or women are happier.

set.seed(23423)
sex<-sample(c("male","female"),100,replace=T)

#Now assume that men reach on average 1 point on the happiness scale and
women 2 points on the happiness scale, and some random noise

y<-1*(sex=="male")+2*(sex=="female")+rnorm(100)

summary(lm(y~sex))

The summary tells you that the intercept, i.e., females, reach an average of
1.98 happiness points and that this value is significantly different from
zero. The male coefficient tells you that men are, on average, 0.75 points
less happy on the happiness scale and that this is significantly different
from the average happiness of women. If there were more groups, then the
coefficients compare the group for which the coefficients is estimated with
the baseline group.

HTH,
Daniel

burdy wrote:
>
> Hi I am having difficulty interpreting the multiple regression output. I
> would like to know what it means when one of the factors is assigned as
> the intercept?
>
> In my data I am looking at the relationship between environmental
> parameters and biological production.
>
> One of my variables in the analysis is substratum type and gravel is
> identified as the intercept and the P-value is significant,...
> Does this mean that I can talk about the relationship which it has with
> production as being significant or because it is the intercept can I not
> use it because it has been selected and used as the basis for the
> relationships between the other factors in the variable substratum?
>
> No one in the PhD office can answer this
>
> I look forward to any replies
>
> Thanks
>
> Matt
>

--
View this message in context: http://r.789695.n4.nabble.com/Multiple-regression-intercept-tp3808045p3808599.html
Sent from the R help mailing list archive at Nabble.com.

```