R help - Oct 2007 - 2 Sample Confidence Interval - Formatting Data?

Hello all,

I have to run a 2 Sample Confidence Interval on some data; the command for such
intervals is "confint(...)", but in the help documentation it says
that you
need a "fitted model object" in order to run this command.  What does
that
mean?

The data is very small, it's:

x=c(8,12,10,14,2,0,0)
y=c(-6,0,1,2,-3,-4,2)


and I want to be able to run a Confidence interval for the difference of two
means (x-bar minus y-bar).  Do I have to fit these two pieces of data into some
kind of object?


A little confused,


Wayne

> -----Original Message-----
> From: r-help-bounces at r-project.org 
> [mailto:r-help-bounces at r-project.org] On Behalf Of Wayne Aldo Gavioli
> Sent: Tuesday, October 09, 2007 3:37 PM
> To: r-help at stat.math.ethz.ch
> Subject: [R] 2 Sample Confidence Interval - Formatting Data?
> 
> 
> 
> Hello all,
> 
> I have to run a 2 Sample Confidence Interval on some data; 
> the command for such
> intervals is "confint(...)", but in the help documentation it 
> says that you
> need a "fitted model object" in order to run this command.  
> What does that
> mean?
> 
> The data is very small, it's:
> 
> x=c(8,12,10,14,2,0,0)
> y=c(-6,0,1,2,-3,-4,2)
> 
> 
> and I want to be able to run a Confidence interval for the 
> difference of two
> means (x-bar minus y-bar).  Do I have to fit these two pieces 
> of data into some
> kind of object?
> 
> 
> A little confused,
> 
> 
> Wayne
> 
Wayne,

You may want to look at ?t.test instead.

Hope this is helpful,

Dan

Daniel J. Nordlund
Research and Data Analysis
Washington State Department of Social and Health Services
Olympia, WA  98504-5204

On Tue, 2007-10-09 at 18:37 -0400, Wayne Aldo Gavioli
wrote:
> 
> Hello all,
> 
> I have to run a 2 Sample Confidence Interval on some data; the command for
such
> intervals is "confint(...)", but in the help documentation it
says that you
> need a "fitted model object" in order to run this command.  What
does that
> mean?
> 
> The data is very small, it's:
> 
> x=c(8,12,10,14,2,0,0)
> y=c(-6,0,1,2,-3,-4,2)
> 
> 
> and I want to be able to run a Confidence interval for the difference of
two
> means (x-bar minus y-bar).  Do I have to fit these two pieces of data into
some
> kind of object?
> 
> 
> A little confused,
Try this, presuming that your data above are not paired:
> t.test(x, y)
        Welch Two Sample t-test

data:  x and y 
t = 3.0621, df = 9.264, p-value = 0.01308
alternative hypothesis: true difference in means is not equal to 0 
95 percent confidence interval:
  2.039893 13.388678 
sample estimates:
mean of x mean of y 
 6.571429 -1.142857 


See ?t.test for more information.


confint(), as per the top of help page which you were reading:

"Computes confidence intervals for one or more parameters in a fitted
model."

Said differently, it computes confidence intervals for the coefficients
in a linear model, not for the difference in means in a two sample test.

HTH,

Marc Schwartz

Hello Wayne,

Welch Test (as suggested by M.S.) can be a good
choice, but since your sample is very small you must
ask yourself what can you assume about the two samples
(i.e. do they come from more or less normal
distributions, etc.).

Regards,

Moshe.

--- Wayne Aldo Gavioli <wgavioli at fas.harvard.edu>
wrote:
> 
> 
> Hello all,
> 
> I have to run a 2 Sample Confidence Interval on some
> data; the command for such
> intervals is "confint(...)", but in the help
> documentation it says that you
> need a "fitted model object" in order to run this
> command.  What does that
> mean?
> 
> The data is very small, it's:
> 
> x=c(8,12,10,14,2,0,0)
> y=c(-6,0,1,2,-3,-4,2)
> 
> 
> and I want to be able to run a Confidence interval
> for the difference of two
> means (x-bar minus y-bar).  Do I have to fit these
> two pieces of data into some
> kind of object?
> 
> 
> A little confused,
> 
> 
> Wayne
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained,
> reproducible code.
>