I need to return to my blog because I need a place to organize my progress on the many different projects I’m working on.
(1) A paper comparing the HLSM with the multilevel p2 model.
(2) Bullying data (see below)
(3) Spillane new model for across teacher ties
(4) Madill data
(5) Simulation study investigating the correlation among ties
(6) SREE talk
(7) Edits to MM Chapter
(8) Working with Megan on fitting a p2 model
(9) New ideas (a spatial HNM, spatial component to an IRT model, network as mediator)
(10) Other odds and ends (emails, job decisions!, potential collaborations with psychologists)
*Bully Data*
I’ve fit separate fits and it’s not clear that there’s an effect. One last thing before I give up. I’m going to fit a Control group only and Treatment group only and estimate an overall mean of each group. We’ll see what happens.
*Correlation Sim Study*
It’s not at all clear that the LSM produces more or less correlated ties as network size increases. But the GEE results suggest that to be the case. What is going on? To answer this question (in part), I’ll look into correlation among valued networks (simulated).
To simulate a valued network: I’ll simulate some type of sparsity. So there’s a probability of a non-zero. And then I’ll simulate some count (poisson). Overdispersed poisson. And see what happens.
Oh right, I keep forgetting about these zeros. If one tie is always 0, it’s not possible to calculate correlation. So ties that are consistently 0 (which could be a good number) have undefined correlation. And that’s problem – there might be some kind of dependence that “causes” these ties to consistently zero. Tie-12 might be always 0 but that would affect Tie-13 if 2 and 3 are close together. Ugh.
I can’t seem to get my head oriented in the right direction. The LS positions (for this simulation) are fixed. So it’s really the distance that contributes to probability of a tie and that’s the same for each round. So maybe what I need to do is jitter/alter the intercept in the model and that will change the overall probabilities and I should compute the correlation among those. Let’s see…
I’m still not measuring what I want to measure. I want to somehow measure how a tie between 1-2 affects a tie between 1-3 or 2-4. The last attempt really just brought me back to correlating the linear functions of the two distances, which is not what I want.