When considering whether or not to use the two sample t test, does the data need to be before and after an event at the same location or is the location irrelevant in certain scenarios.
I refer to the below as an example where i think the two sample t test would be appropriate. I have no issue with the maths involved here rather the interpretation of the question.
Cheers for any help
A further survey of 40 households were undertaken in another region and the mean was found to be 16,500 miles per year added to the main car clock in 2006, whilst the standard deviation was 4,870 miles per year.
(i) Perform a test to decide whether there is evidence of a significant difference in the mean miles added to the main car clock between the two regions.
(ii) Clearly state the key assumptions and results underlying your analysis in part (i).
Jan 11th 2012, 07:13 AM
Re: Two Sample t test
If I understood you correctly, yes you can use a two sample t-test from two independent samples they aren't just used for correlated samples.