Page 1 of 1

Preparation of continuous variables before MCMC (checking linearity and centring to the mean)

Posted: Tue Mar 09, 2021 6:55 pm
by ManuelDewez
Hello I am new to runmlwin.

I want to conduct a two-level random intercept multilevel analysis with a binary outcome variable, using the MCMC method to identify determinants of the availability of diagnostic tests in primary care surgeries.

I am using a multilevel approach because I want to examine the effects of the medical surgeries’ characteristics (level 1) and the effects of country (level 2) on the availability of the test in the surgery.

I am using 5 surgery explanatory variables (three are binary variables and two are continuous variables) and two country explanatory variables (both categorical). I have 847 observations (i.e. 847surgeries) and 19 countries.

I have 2 preliminary questions regarding the continuous explanatory variables:
1. Do I need to check if there is a linear association between each continuous variable and the outcome, and if not transform the continuous variable (e.g. quadratic or cubic transformation)?
2. Do I need to centre the continuous variable to the mean or to any other value?

Thank you in advance!