R help - Jun 2011 - Help with Median test and Coxon-Mann-Whittney test

Hi All,

I have the following dataset (triplicates values of 5 independent
measurements and duplicate vaues of a control):

1            2          3           4           5          C
181.8      58.2     288.9     273.2    290.9    53.9
120.3     116.8    108.9     281.3     446      39.6
86.1      148.5     52.9      126       150.3

My aim was to find if mean values of Samples 1 - 5 are significantly higher
than the mean value for C (control).

 At first, I carried out mean, SD and t-test (one-tail). Although SD error
bars are large, two of the samples have mean values that are significantly
higher than that of C.

Second, I carried out median and Coxon-Mann-Whittney test because of my
worry about the size of my data and the high variations in the replicates. I
was surprised however, that median and Coxon-Mann-Whittney tests did not
reveal statistical significant results.

I will be happy if anyone could adivce me on the best way to analyse this
dataset.

Regards,

Rob

	[[alternative HTML version deleted]]

Hi Rob,

This list is primarily intended for questions about how to do things in R,
so you're more likely to get a helpful response elsewhere. You might want to
try some place like the Cross-Validated web site (
http://stats.stackexchange.com/) for general statistics and data analysis
questions.

Before you do that, you might want to clarify your question. It sounds like
you ran a couple tests on a very small dataset, and they came up with
different results. They're not even tests for the same thing, since the
t-test is testing whether the means are different, and the Mann-Whitney is
testing whether the medians are different. And in any case, it's hard to
imagine coming up with any kind of useful inference on this tiny sample.
When you post on Cross-Validated (or other site), people are going to want
to know the problem you are trying to solve, which isn't at all clear at
this point.

Best,
Ethan

On Fri, Jun 10, 2011 at 12:26 PM, Robert Johnson
<robjohn7x@gmail.com>wrote:
> Hi All,
>
> I have the following dataset (triplicates values of 5 independent
> measurements and duplicate vaues of a control):
>
> 1            2          3           4           5          C
> 181.8      58.2     288.9     273.2    290.9    53.9
> 120.3     116.8    108.9     281.3     446      39.6
> 86.1      148.5     52.9      126       150.3
>
> My aim was to find if mean values of Samples 1 - 5 are significantly higher
> than the mean value for C (control).
>
>  At first, I carried out mean, SD and t-test (one-tail). Although SD error
> bars are large, two of the samples have mean values that are significantly
> higher than that of C.
>
> Second, I carried out median and Coxon-Mann-Whittney test because of my
> worry about the size of my data and the high variations in the replicates.
> I
> was surprised however, that median and Coxon-Mann-Whittney tests did not
> reveal statistical significant results.
>
> I will be happy if anyone could adivce me on the best way to analyse this
> dataset.
>
> Regards,
>
> Rob
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> 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]]