Do you have independent samples or paired samples (repeated measurements)?
– OR –
More Information (if you need help deciding)
In order to answer this question you are going to have to plot a histogram of your variable of interest. When you look at your histogram, you will need to determine if it is normal (bell shaped with most of the data residing in the center) or skewed (leaning right or left with most of the data not in the center).
Independent samples: independent samples indicate that the two sets of group values on your variable of interest are not directly related at all. For example, if you measured the weights of 10 random men and 10 random women, these groups would be independent because the weights of the randomly chosen men did not have any effect on the weights of the 10 randomly sampled women.
Paired Samples (repeated measurements): paired samples indicate that your two groups are related in some way. For example, if students take a pre-test and a post-test, the results on those tests are dependent, meaning you have 2 measures for the same student, and some students may score well on the tests whereas other students may not. Paired samples usually occur when you have 2 measurements for the same research subject.