# Seems so simple: differences... and yet but not between groups... Which test to use?

#### Becky

##### New Member
Hi there

This seems so simple I'm hoping someone can help me!

I've got the mean values for no. effective alleles in 5 different populations of fish.
So, 5 rivers, each with a mean value... and I'm trying to give a statistical value as to whether there are significant differences between the values of the 5 groups.

River 1 = 5.89
River 2 = 5.53
River 3 = 3.56
River 4 = 3.69
River 5 = 6.25

I've tried doing one-way ANOVA, but as each "group" or factor only has the one value, it will not work... and yet I can't use other tests for "groups".. as it is just 1 value per population (or sample).

I've been told not to use the full list of values, only these means (I've gone down that route don't worry!)

Please, desperate, someone tell me I'm missing the obvious and it's simple test?
Thank you for any help gratefully received!
:yup:

#### Londoner

##### New Member
Re: Seems so simple: differences... and yet but not between groups... Which test to u

Hi Becky,

I'm not really a stats expert, but it sounds to me like you could think about doing t tests to test for differences between the river means. However, to do this I think you would need to know the standard deviations for each river as well as the sample size.

http://stattrek.com/hypothesis-test/difference-in-means.aspx

#### rogojel

##### TS Contributor
Re: Seems so simple: differences... and yet but not between groups... Which test to u

hi,
I guess the best would be to get the raw data from which the means came from and then do the ANOVA. Failing this, maBe you can get the standard deviations and sample sizes together with the means, the you could at least do pairwise comparisons between the groups.

Unfortunately mean values without any information on their variation are useless IMHO.

regards
rogojel

#### bugman

##### Super Moderator
Re: Seems so simple: differences... and yet but not between groups... Which test to u

I've been told not to use the full list of values, only these means (I've gone down that route don't worry!)
Why would you be told that?