One sample t test

[Davidm]

I have a population of 11 samples. Can I take the mean from 2 of those 11 samples and use a one sample t test to determine if the mean of the 2 samples differs from the 11 sample population? Is there a better method? Thank you

 

[hlsmith]

What is the purpose/ question you are trying to address?

 

[Davidm]

Asking:
Do 2 geographically distinct samples/values differ from the 9 geographically distinct samples/values?

I have 11 samples/values to represent a population from the United States.

Thank you for your reply.

 

[hlsmith]

How were the two selected?

Why not two-sample t-test, since the one sample would not account for the missing variability around the proposed constant?

 

[Davidm]

How were the two selected?

Why not two-sample t-test, since the one sample would not account for the missing variability around the proposed constant?

The 11 were selected based on census criteria and infection rates. The two were then identified for being unique geographical regions (islands). The other 9 are on continents. The question is do the infection rates on the islands differ from the infection rates on the mainland.

Thank you!

 

[Dason]

I guess I echo the question on why you don't want to do a two sample t-test here?

 

[noetsi]

You can also run a contrast between the two means, but I think that would generate the same result as a two sample t test.

 

[Davidm]

Thank you all for your help! I will use a two sample t-test.