Subjects do not provide evidence for either H0 or H1

SBrol

Dear all

I am trying to do a Bayesian independent t-test, where I compare HC vs Patients. However, as you can see in the plot, it seems that the first group do not provide any evidence..

In figure above my grouping variable is sorted on HC=0 and PT=1. However, previously my data was sorted on another variable, group, and it was like this in the file:

PT=2 (for 40 cases)

HC=3 (10 cases)

PT=2 (~50 cases)

HC=3 (~70 cases)

When the data was sorted like this, the sequential analysis plot looked like this:

So now it seemed like those first 40 cases do not provide evidence. But it is the same data, and outcome measure, and I would assume that it should not matter how the subjects are sorted?

Does anybody maybe have an idea what I might try to fix this?

Thanks!

Sophie

EJ

Hi Sophie,

When you do an independent samples t-test, the sequential BF will remain at 1 until you feed it data from the other group. Without having seen any data from the other group, there can be no evidence of an effect of the manipulation. So if the first x people are all in group A, the BF will be 1 for the first x people.

EJ

SBrol

Ah, that makes sense:) Thanks!