[R-lang] lmer() and polarity of IVs

[Martin Corley]

Hi all,

I'm looking at a model where there are two IVs -- each has a
value "change" and "nochange" (and I want to fit a simple 2x2 model).

The problem is that depending on the intercept I use, the models
aren't symmetrical -- in fact the interaction term doesn't change, but
everything else does, and I'm confused!  If I exclude the interaction
term the models are equivalent (but the interaction term is justified,
X2 is the same at around 22 in each case).  Here's some sample
output...

Any help greatly appreciated!

--Martin

#### original version

> summary(artic.lmer1)
Linear mixed-effects model fit by REML 
Formula: norm ~ place * voice + (1 | sub) + (1 | item) 
   Data: artic 
   AIC   BIC logLik MLdeviance REMLdeviance
 11004 11043  -5496      10978        10992
Random effects:
 Groups   Name Variance  Std.Dev.
 item          0.0495563 0.222613
 sub           0.0054327 0.073707
 Residual      0.5029739 0.709207
number of obs: 5037, groups: item, 64; sub, 5

Fixed effects:
                        Estimate Std. Error t value
(Intercept)               0.9726     0.1230   7.908
placeplace+               0.9501     0.1280   7.425
voicevoice+               0.7181     0.1367   5.251
placeplace+:voicevoice+  -0.7688     0.1528  -5.030

Correlation of Fixed Effects:
            (Intr) plcpl+ vocvc+
placeplace+ -0.892              
voicevoice+ -0.835  0.802       
plcplc+:vc+  0.747 -0.837 -0.895


#### releveled version

> summary(artic.lmerRLI)
Linear mixed-effects model fit by REML 
Formula: norm ~ relevel(place, "place+") * relevel(voice, "voice+") +      (1 | sub) + (1 | item) 
   Data: artic 
   AIC   BIC logLik MLdeviance REMLdeviance
 11004 11043  -5496      10978        10992
Random effects:
 Groups   Name Variance  Std.Dev.
 item          0.0495563 0.222613
 sub           0.0054327 0.073707
 Residual      0.5029739 0.709207
number of obs: 5037, groups: item, 64; sub, 5

Fixed effects:
                                                              Estimate
(Intercept)                                                    1.87198
relevel(place, "place+")place-                                -0.18124
relevel(voice, "voice+")voice-                                 0.05074
relevel(place, "place+")place-:relevel(voice, "voice+")voice- -0.76883
                                                              Std. Error
(Intercept)                                                      0.05845
relevel(place, "place+")place-                                   0.08361
relevel(voice, "voice+")voice-                                   0.06828
relevel(place, "place+")place-:relevel(voice, "voice+")voice-    0.15285
                                                              t value
(Intercept)                                                     32.03
relevel(place, "place+")place-                                  -2.17
relevel(voice, "voice+")voice-                                   0.74
relevel(place, "place+")place-:relevel(voice, "voice+")voice-   -5.03

Correlation of Fixed Effects:
             (Intr) rlvl(p,"+")- rlvl(v,"+")-
rlvl(p,"+")- -0.477                          
rlvl(v,"+")- -0.584  0.408                   
r(,"+")-:("   0.261 -0.547       -0.447      
> 


-- 
Martin Corley                                  Martin.Corley at ed.ac.uk
                                   http://homepages.ed.ac.uk/martinc/     
Philosophy, Psychology and Language Sciences, University of Edinburgh