Developing the model

Complete the steps in the modelling process and develop a final model for interpretation with explanatory variables at both levels. Summarize the main results.

  1. The first step is to estimate the null model and compute the intraclass correlation.
  2. The second step is to develop the full level 1 model, normally the individual level model.
  3. The third step is to develop the final model with level 2 explanatory variables.
SPSS solution
** The null model.
mixed poltrust with agec agec2 female seced terted copeinc stfeco stfgov stfdem ppltrst
/fixed =
/method = ml
/random = intercept | subject(cntry)
/print = solution.

** Full individual level model.
mixed poltrust with agec agec2 female seced terted copeinc stfeco stfgov stfdem ppltrst
/fixed = agec agec2 female seced terted copeinc stfeco stfgov stfdem ppltrst
/method = ml
/random = intercept | subject(cntry)
/print = solution.

** Models with country level variables HDI + CPI **.
mixed poltrust with agec agec2 female seced terted copeinc stfeco stfgov stfdem ppltrst c_hdihdr_2009 c_ticpi_2008
/fixed = agec agec2 female seced terted copeinc stfeco stfgov stfdem ppltrst c_hdihdr_2009 c_ticpi_2008
/method = ml
/random = intercept | subject(cntry)
/print = solution.
Stata solution
** The null model
xtmixed poltrust || cntry: , mle variance, if nmiss==0

** Full individual level model
xtmixed poltrust agec agec2 female seced terted copeinc stfeco stfgov stfdem ppltrst || cntry: , mle variance, if nmiss==0
** Models with country level variables HDI + CPI
xtmixed poltrust agec agec2 female seced terted copeinc stfeco stfgov stfdem ppltrst c_hdihdr_2009 c_ticpi_2008 || cntry: , mle variance, if nmiss==0
Go to next page >>