T-test is appropriate when variances of the two populations can be assumed to be equal, I am not sure your data holds that assupmtion. Perhaps you could find some clue at:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2789971/
The t-test has two variants: one with equality of variances assumed and one without this assumption. You should first check that your data is normally distributed using the Shapiro-Wilk normality test. If it is not normally distributed you can still run the t-test provided that your sample sizes are equal and greater than 25 - see:
Sawilowsky, S. S. and Blair, R. C. (1992) A more realistic look at the robustness and Type II error properties of the t test to departures from population normality, Psychological Bulletin, 111(2), pp. 352–360.
If this condition is not met, use the Mann-Whitey U test.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India