R: Examples for 'MASS::nlschools'

### ** Examples

## Don't show: 
op <- options(digits=5)
## End(Don't show)
nl1 <- within(nlschools, {
IQave <- tapply(IQ, class, mean)[as.character(class)]
IQ <- IQ - IQave
})
cen <- c("IQ", "IQave", "SES")
nl1[cen] <- scale(nl1[cen], center = TRUE, scale = FALSE)

nl.lme <- nlme::lme(lang ~ IQ*COMB + IQave + SES,
                    random = ~ IQ | class, data = nl1)
## IGNORE_RDIFF_BEGIN
summary(nl.lme)

Linear mixed-effects model fit by REML
  Data: nl1 
    AIC   BIC  logLik
  15120 15178 -7550.2

Random effects:
 Formula: ~IQ | class
 Structure: General positive-definite, Log-Cholesky parametrization
            StdDev  Corr  
(Intercept) 2.78707 (Intr)
IQ          0.48424 -0.516
Residual    6.24839       

Fixed effects:  lang ~ IQ * COMB + IQave + SES 
             Value Std.Error   DF t-value p-value
(Intercept) 41.370   0.35364 2151 116.985  0.0000
IQ           2.124   0.10070 2151  21.088  0.0000
COMB1       -1.672   0.58719  130  -2.847  0.0051
IQave        3.248   0.30021  130  10.818  0.0000
SES          0.157   0.01465 2151  10.697  0.0000
IQ:COMB1     0.431   0.18594 2151   2.317  0.0206
 Correlation: 
         (Intr) IQ     COMB1  IQave  SES   
IQ       -0.257                            
COMB1    -0.609  0.155                     
IQave    -0.049  0.041  0.171              
SES       0.010 -0.190 -0.001 -0.168       
IQ:COMB1  0.139 -0.522 -0.206 -0.016 -0.003

Standardized Within-Group Residuals:
      Min        Q1       Med        Q3       Max 
-4.059387 -0.631084  0.065519  0.717864  2.794540 

Number of Observations: 2287
Number of Groups: 133

## IGNORE_RDIFF_END
## Don't show: 
options(op)
## End(Don't show)

Examples for 'MASS::nlschools'

Eighth-Grade Pupils in the Netherlands