> print(fm1, corr = FALSE)
Linear mixed-effects model fit by REML 
Formula: math ~ sx * eth + gr + cltype + (yrs | id) + (1 | tch) + (yrs |      sch) 
   Data: star 
    AIC    BIC  logLik MLdeviance REMLdeviance
 238810 239005 -119381     238837       238762
Random effects:
 Groups   Name        Variance Std.Dev. Corr   
 id       (Intercept) 1110.257 33.3205         
          yrs           45.028  6.7103  -0.301 
 tch      (Intercept)  235.240 15.3375         
 sch      (Intercept)  255.153 15.9735         
          yrs           44.975  6.7063  -0.775 
 Residual              335.342 18.3124         
Number of obs: 24578, groups: id, 10732; tch, 1374; sch, 80

Fixed effects:
            Estimate Std. Error t value
(Intercept) 561.5745     1.5754   356.5
sxF           2.4332     0.8590     2.8
ethB        -22.7079     1.3983   -16.2
ethA          5.7492     9.3765     0.6
ethH         -0.8045    12.7543    -0.1
ethI        -23.4432    16.7808    -1.4
ethO         -5.5812    10.5379    -0.5
gr.L         96.2172     1.9424    49.5
gr.Q         -4.4521     0.8910    -5.0
gr.C         -3.3736     0.8739    -3.9
cltypereg    -8.0250     1.2377    -6.5
cltypereg+A  -7.2302     1.2328    -5.9
sxF:ethB      1.0180     1.4298     0.7
sxF:ethA     -5.4603    13.3146    -0.4
sxF:ethH     12.2617    19.4519     0.6
sxF:ethI    -13.9292    27.6507    -0.5
sxF:ethO     22.6353    17.1675     1.3
> dim(fm1@A)
[1] 23016 23016
> rev(sort(sapply(slotNames(fm1), function(nm) object.size(slot(fm1, nm)))))
       L        A     ZXyt    frame    flist   offset  weights    ranef 
 4103968  3594864  3547072  3180528   978952   196664   196664   184024 
   terms     call   cnames       ST     dims deviance    fixef       Gp 
    6168     4128     2112     1072      632      528      176       72 
>