Maddie Kushner
Hi!
I noticed another quirk in my output and was wondering if anyone has insight into how to handle this. Starting a new post because the topic of this question is different than my previous one.
I have noticed that my output separates the contemporaneous paths by direction, for example, there might be 12 individual networks have neuroticism 'predicting' agreeableness, but then there are 5 more that have agreeableness 'predicting' neuroticism (all non-lagged). It makes sense that they'd be separated out for lagged associations, but for contemporaneous paths, this is challenging to interpret. I get that this may occur because gimme is estimating betas, which can be tested in both directions, as opposed to something like correlation coefficients, but should these be interpreted as equivalent associations regardless of direction? Or is there some kind of distinction between them?
In addition, I have noticed some individual networks which seem to have significant contemporaneous paths going in either direction. For example, one individual network exhibits contemporaneous path in which neuroticism 'predicts' antagonism and another where antagonism 'predicts' neuroticism. In some cases both are positive paths and in some cases one is positive and one negative. Also perplexed about how to interpret these kinds of associations.
I have attached my individual path estimates and path summary spreadsheets, as well as attached plots of some models with the bidirectional contemporaneous associations for reference.
Thanks in advance!
Maddie