Which one should I go for? My aim is to get a list of genes that are differentially expressed.
Usually you want a two-tailed test. You are better of with t-test, and you need to apply the Benferroni correction (see here http://en.wikipedia.org/wiki/Bonferroni_correction) as well. Be aware though that depending on your experiment design and number of replicates (you should have at least three for each of control and experimental) you will get very different answers.
t-test = comparison of two sets of variables. Anova = more than 2 sets and, as Sameet pointed, you will have options to make corrections if you use software like Prism.
t- test use to data less than 30 and anova more than and t -test for two sets and anova more than three
Dear if you have two groups go for t-test and if you have more groups go for ANOVA do not worry about the results because ANOVA just tells you that there is difference in groups but it does not point out where is difference while t-test will tell you where is the difference
Hi Rasheeba,
For two groups , go for unpaired t test.
For more than two groups go for anova test. If the null hypothesis is accepted in anova that means there is no difference between the groups.
But in case if the null hypothesis is rejected then follow it by post hoc test to see which of the pair of the goups are different among themselves.
Thanks
Hi Rasheeba,
Strictly speaking, as some colleagues say, the t test is for 2 populations (or groups), in the case of more than two groups or populations anova should be applied, since it is a joint procedure.
Mathematically, if the data do NOT present outliers, they behave approximately as a normal probability distribution and are homocedastic, so in the CASE of 2 populations, the results are the same, with respect to the approval or rejection of the hypothesis.
if the above does not occur, then there could be a variation depending on the size of the treatments
Something in the data is atypical!
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA