help > Degrees of freedom using NBS
Showing 1-2 of 2 posts
Jul 27, 2021 03:07 PM | asinha
Degrees of freedom using NBS
We ran NBS using a one-sample t-test, 5000 permutations and
p<0.05 to identify statistically significant connections at each
time point such that we could reject the null hypothesis of a zero
mean. In total, we analyzed functional connectivity in 8 ROIs ->
28 total connections. If NBS found 15 statistically significant
connections in one connected component, is there a way to determine
what the degrees of freedom are? Is it 28-1 = 27 or 15-1 = 14? Or
is degrees of freedom not straightforward/relevant using NBS in
this context?
Jul 27, 2021 11:07 PM | Andrew Zalesky
RE: Degrees of freedom using NBS
Hi Anita,
the degrees of freedom will simply be the degrees of freedom for the t-test. The degrees of freedom won't depend on the number of components.
For a one-sample t-test the degrees of freedom are usually the total number of subjects minus 1
best,
Andrew
Originally posted by asinha:
the degrees of freedom will simply be the degrees of freedom for the t-test. The degrees of freedom won't depend on the number of components.
For a one-sample t-test the degrees of freedom are usually the total number of subjects minus 1
best,
Andrew
Originally posted by asinha:
We ran NBS using a one-sample t-test, 5000
permutations and p<0.05 to identify statistically significant
connections at each time point such that we could reject the null
hypothesis of a zero mean. In total, we analyzed functional
connectivity in 8 ROIs -> 28 total connections. If NBS found 15
statistically significant connections in one connected component,
is there a way to determine what the degrees of freedom are? Is it
28-1 = 27 or 15-1 = 14? Or is degrees of freedom not
straightforward/relevant using NBS in this context?