R help - Feb 2011 - Perform one-way ANOVA using standard deviation and mean

Hi! I need to perform ANOVA on a couple of data sets. The only information I
have are N, Mean and Standard deviation. I am very new to R, so can someone
point me to the right direction on where to go? Thank you very much.


Sincerely
Niny Rao, PhD
Philadelphia University

On Feb 7, 2011, at 8:15 PM, Rao, Niny wrote:
> Hi! I need to perform ANOVA on a couple of data sets. The only  
> information I have are N, Mean and Standard deviation. I am very new  
> to R, so can someone point me to the right direction on where to go?
You need to go back and get the data.
> Thank you very much.
>
>
> Sincerely
> Niny Rao, PhD
> Philadelphia University

-- 
David Winsemius, MD
West Hartford, CT

On Monday2/7/11 9:09 PM, "David Winsemius" <dwinsemius at
comcast.net> wrote:
>
>On Feb 7, 2011, at 8:15 PM, Rao, Niny wrote:
>
>> Hi! I need to perform ANOVA on a couple of data sets. The only
>> information I have are N, Mean and Standard deviation. I am very new
>> to R, so can someone point me to the right direction on where to go?
>
>Here are my data
ID Mean SD Coef of Var
62L 121.60 6.21 5.10
94D 115.92 3.22 2.78
14F 86.51 3.17 3.66
11P 126.71 5.45 4.30
10L 116.67 2.17 1.86
96T 98.20 1.38 1.41
55U 125.93 3.69 2.93
79X 93.73 5.86 6.25
55L 84.24 2.52 2.99
59Q 128.72 0.74 0.58
56S 121.12 1.07 0.89



Sincerely
Niny Rao, PhD
Philadelphia University

Hi:

You need to check that you have the sufficient statistics necessary to
obtain an ANOVA table corresponding to the model you intend to fit.  If you
have that, you can use the vectorization and matrix capabilities in R to
manually get the ANOVA table, perform the F tests, etc. The only thing you
won't be able to get are residuals, so you'll be unable to check the
assumptions of the model. The pieces are all there, but you'll likely have
to do the calculations yourself. I don't believe there exists a function in
base R to do that reconstruction for you.

I just saw your most recent response - that is not the 'data' to which
David
referred, it's a *summarization* of the data. It's possible to construct
an
ANOVA from the summary information *if* your summaries include the
sufficient statistics of the ANOVA model you want to fit. If you don't know
what 'sufficient statistics' means, then you probably need to consult a
statistician in your neighborhood. For one-way ANOVA, the sample means,
sample standard deviations and sample sizes per group should be enough to
construct the ANOVA table and do the basic analysis. Many books that cover
one-way ANOVA contain the formulas to do the necessary calculations.

HTH,
Dennis

On Mon, Feb 7, 2011 at 5:15 PM, Rao, Niny <raon@philau.edu> wrote:
> Hi! I need to perform ANOVA on a couple of data sets. The only information
> I have are N, Mean and Standard deviation. I am very new to R, so can
> someone point me to the right direction on where to go? Thank you very
much.
>
>
> Sincerely
> Niny Rao, PhD
> Philadelphia University
>
> ______________________________________________
> R-help@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.
>
	[[alternative HTML version deleted]]