Dear Bethany
Sorry for the delay in replying. I recommend that for simplicity, report the parameters as they come - i.e., don't translate to units of Hz.
You can explain that the more positive the self-connection parameter, the more self-inhibition in that region, and the more negative the parameters, the less self-inhibition. The same applies to covariates in the PEB model - a positive effect of a covariate on a self-connection means indicates a positive relationship between the covariate and the level of self-inhibition, and vice versa.
All the best
Peter
On 14/05/2021, 20:53, "Bethany Sussman" <[log in to unmask]> wrote:
Hello Peter,
Thank you for your response, I think I understand, but I want to double check.
It sounds like the output for the self connection diagonal is still in the unitless log-scale format, so I should put it in to Hz format using the formula A_Hz = -0.5 * exp(A) before reporting.
If you run a PEB analysis and put a covariate in to the model (e.g. a clinical score), should you also do this transformation for the covariate A-matrix?
Best,
Bethany
|