I'm analysing a set of 400 ECG signals (each one of a duration of 30 sec). I want to get the mean signal of all my signals but I don't know how to prove if my mean signal is representative of all the signals that I averaged.

I know that I can use the standard deviation but I don't know how to use it in this case.

Thanks for your attention. ]]>

I want to test if the means are different between subsets (e.g early in the day versus late in the day) are different. If I had independent samples I'd just use a t-test. But since the samples are dependent, how do I deal with this? What stats tests are appropriate?

I think this is a case of "repeated measures", but I'm not sure how to approach it since I don't know which samples (really if any) were measured more than once.

Thanks! Any help would be amazing. ]]>

I would like an explanation with differences regarding the p value which i observed when i perform t-test and post hoc tests.

I have to compare the means between three groups. Specifically:

Group 1 vs Group 2 and

Group 1 vs Group 3.

So i performed a parametric one way analysis of variance testing (One-Way Anova) as my samples met all the criteria regarding normallity and variance testing.

The global p value of my One-Way anova is p=0.02 but the Bonferroni post hoc test gives non significant results. Is that possible?

Then i was confused when i performed unpaired t-test to compare the above groups. Specifically, unpaired t-test :

for the group1 vs group 2 gave me a p value at 0.008 and

for group 1 vs group 3 gave a p value at 0.03.

So my question is why the p value from post hoc test (non significan) and unpaired t test (significant) are different?

Is that possible taking into consideration that the compared groups are the same in both types of tests?

And if that is possible, which results should i trust?

Thank you in advance! ]]>