quesion about lm function

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
project.org] On Behalf Of meng
Sent: Sunday, March 10, 2013 5:33 AM
To: Greg Snow
Cc: R help
Subject: Re: [R] quesion about lm function

Thanks for your reply.
But the mean y of sex(f) and sex(m) can't be negative since the min and
max of y is 1632.5 and 6410.6 respectively.

And your code's result:

sex(f): -1255.56
sex(m):-1118.73

The above result isn't correct since they are negative.

The tests are likely to be meaningless in this case and the means are
the predicted means at x1=0 and x2=0 (so the means may make more sense
if x1 and x2 are centered  first).

Thanks









At 2013-03-10 02:39:21,"Greg Snow" <538280 at gmail.com> wrote:

You can get the 2 means with:


result_lm <- lm(y~0+factor(sex) + x1 + x2)


The tests are likely to be meaningless in this case and the means are
the predicted means at x1=0 and x2=0 (so the means may make more sense
if x1 and x2 are centered  first).



On Sat, Mar 9, 2013 at 3:35 AM, meng <laomeng_3 at 163.com> wrote:
Hi all:
My data is in the attachment.
I want to analysis the mean difference of y between 2 sex.

My code:
result_lm<-lm(y~factor(sex) + x1 + x2)
summary(result_lm)

The result of "factor(sex)m" 136.83, is the mean difference of y
between 2 sex,and the corresponding p value is 0.07618.

My question is: how to get the mean y of sex(m) and sex(f) respectively
via lm function?

Many thanks for your help.
My best.