|
|
Estimates random slopes per cluster |
|
Message/Author |
|
|
Is it possible for multilevel logistic regession analyses to obtain the estimates of the random slopes per cluster? For example, I know that the slope of IV1 is positively associated with DV1, and varies across countries, but I don't know whether the direction is the same for all countries. Thanks in advance! |
|
|
Yes, just request FSCORES (factor scores). |
|
|
Thank you for your quick reply. I have a follow up question: Is it possible to test whether these cluster-estimates significantly differ from 0 (or each other)? Or does such a test not make sense because the parameter estimates are realizations of the random effect? |
|
|
Right, it doesn't make sense because you are considering a random variable values, not parameters. The interest should instead be of the variance of the random effect is zero or not. |
|
Back to top |
|
|