Advice: Paired t Test
Small to medium sample
PAIRED t TEST: This is listed in the parametric methods section of the analysis menu. For small samples you can use this method with the assumption that the differences between pairs are from a normal distribution. With paired studies each individual is acting as their own control therefore you have reduced inter-individual variability leaving most of the error due to measurement. Furthermore this error tends to normality when we are considering differences. If you resort to transformations to "normalise" your data then you are advised to seek statistical advice. Wilcoxon's signed ranks test is the nonparametric alternative.