Re: Binary MLM assumptions
Posted: Wed Sep 06, 2017 8:44 am
Hi Jessica,
In answer to your questions:
a) Yes at higher levels you can check the normality assumption via residual plots - the residuals at level 1 should not be normal
b) Not sure what you want to do here - you can plot the transformed fitted line to your data for example or you could test a quadratic term to see if that is significant
c) Again you could look for outlying clusters at higher levels e.g. a cluster with more 1s or 0s than would fit a normal distribution at level 2 - level 1 outliers are more challenging as the data is 0 or 1.
Hope that helps,
Bill.
In answer to your questions:
a) Yes at higher levels you can check the normality assumption via residual plots - the residuals at level 1 should not be normal
b) Not sure what you want to do here - you can plot the transformed fitted line to your data for example or you could test a quadratic term to see if that is significant
c) Again you could look for outlying clusters at higher levels e.g. a cluster with more 1s or 0s than would fit a normal distribution at level 2 - level 1 outliers are more challenging as the data is 0 or 1.
Hope that helps,
Bill.