R help - Aug 2012 - significance of character (Anova)

Hi, 
I have a problem with the output of my anova.

> Tabelle2 <- data.frame(sb_ek_ber, JE, ABWHALT_C, ALLEINF,
Alter_Ältester_inklVN_C, Alter_Jüngster_C_inkl_AlterNutz, ALTERKAU_C, Wuerfel,
BEGKUNDE_C, Geoscore_C, GESCHL_VN,
+  NUTZART, NUTZKREIS, NICHTK_C_korrigiert, TARIFDAT_C, TARIFGRP_C, WOHNEIGEN,
WOHN_C, ZWEITFAHRZ, ZAHLWEISE, LASTSCHR)

 str(Tabelle2)
'data.frame':   1237838 obs. of  21 variables:
 $ sb_ek_ber                      : num  0 0 0 0 0 0 0 0 0 0 ...
 $ JE                             : num  0.1972 0.2222 0.0972 0.5694 0.5 ...
 $ ABWHALT_C                      : Factor w/ 3 levels
"0","1","2": 2 2 2 2 2 2 2 2 2 2 ...
 $ ALLEINF                        : Factor w/ 4 levels
"0","1","2","3": 3 3 4 4 4 4 3 4 4 4 ...
 $ Alter_Ältester_inklVN_C        : Factor w/ 7 levels
"0","59","60",..: 2 2 2 2 2 3 2 2 2 2 ...
 $ Alter_Jüngster_C_inkl_AlterNutz: Factor w/ 10 levels
"0","17","18",..: 10 10 10 10 10 10 10 10 10 10
...
 $ ALTERKAU_C                     : Factor w/ 8 levels
"0","1","4","7",..: 1 1 1 3 3 1 3 3 4 4
...
 $ Wuerfel                        : Factor w/ 6 levels
"1","2","3","4",..: 1 4 5 6 2 6 3 1 3 1
...
 $ BEGKUNDE_C                     : Factor w/ 11 levels
"0","1","2","3",..: 3 3 3 3 3 3 2 3 3 3
...
 $ Geoscore_C                     : Factor w/ 8 levels
"0","1","2","3",..: 6 6 8 1 8 7 8 2 4 4
...
 $ GESCHL_VN                      : Factor w/ 4 levels
"0","1","2","3": 3 3 3 2 2 3 3 2 3 3 ...
 $ NUTZART                        : Factor w/ 3 levels
"0","1","2": 2 2 2 2 2 2 2 2 2 2 ...
 $ NUTZKREIS                      : Factor w/ 4 levels
"0","1","2","3": 2 2 4 3 3 3 2 4 3 3 ...
 $ NICHTK_C_korrigiert            : Factor w/ 4 levels
"0","1","2","99": 1 1 1 1 1 1 1 1 1 1
...
 $ TARIFDAT_C                     : Factor w/ 9 levels
"0","1","2","3",..: 3 3 3 3 3 3 3 3 3 3
...
 $ TARIFGRP_C                     : Factor w/ 5 levels
"10","30","50",..: 1 1 1 1 1 1 4 1 1 1 ...
 $ WOHNEIGEN                      : Factor w/ 5 levels
"0","1","2","3",..: 5 5 4 5 5 4 5 3 4 4
...
 $ WOHN_C                         : Factor w/ 5 levels
"0","1","2","3",..: 5 5 2 5 5 2 5 2 2 2
...
 $ ZWEITFAHRZ                     : Factor w/ 4 levels
"0","1","2","3": 4 4 4 4 4 4 4 4 4 4 ...
 $ ZAHLWEISE                      : Factor w/ 5 levels
"0","1","2","4",..: 2 2 2 2 2 4 4 2 2 2
...
 $ LASTSCHR                       : Factor w/ 4 levels
"0","1","2","9": 2 2 2 2 2 2 2 2 2 2 ...


anova(ypoi, test="Chisq")
Analysis of Deviance Table

Model: poisson, link: log

Response: sb_ek_ber

Terms added sequentially (first to last)


          Df  Deviance Resid. Df Resid. Dev  Pr(>Chi)    
NULL                     1237837 4.4998e+10              
ABWHALT_C  2 244353722   1237835 4.4754e+10 < 2.2e-16 ***
Wuerfel    5  63772679   1237830 4.4690e+10 < 2.2e-16 ***

Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 
There were 50 or more warnings (use warnings() to see the first 50)

The thing is that the caracter "Wuerfel" is random generated  
and it shouldn' t  be significant.

It would be realy great if you could help me. 
	[[alternative HTML version deleted]]

On Aug 10, 2012, at 15:47 , Jennifer Kaiser wrote:
> 
> Model: poisson, link: log
> 
> Response: sb_ek_ber
> 
> Terms added sequentially (first to last)
> 
> 
>           Df  Deviance Resid. Df Resid. Dev  Pr(>Chi)    
> NULL                     1237837 4.4998e+10              
> ABWHALT_C  2 244353722   1237835 4.4754e+10 < 2.2e-16 ***
> Wuerfel    5  63772679   1237830 4.4690e+10 < 2.2e-16 ***
> 
> Signif. codes:  0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 
> There were 50 or more warnings (use warnings() to see the first 50)
> 
> The thing is that the caracter "Wuerfel" is random generated  
> and it shouldn' t  be significant.
> 
> It would be realy great if you could help me. 

The most obvious guess is that sb_ek_ber probably is not Poisson distributed. 

-- 
Peter Dalgaard, Professor,
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com

On Aug 13, 2012, at 15:38 , Jennifer Kaiser wrote:
> > Model: poisson, link: log
> > 
> > Response: sb_ek_ber
> > 
> > Terms added sequentially (first to last)
> > 
> > 
> >          Df  Deviance Resid. Df Resid. Dev  Pr(>Chi)    
> > NULL                    1237837 4.4998e+10              
> > ABWHALT_C  2 244353722  1237835 4.4754e+10 < 2.2e-16 ***
> > Wuerfel    5  63772679  1237830 4.4690e+10 < 2.2e-16 ***
> > 
> > Signif. codes:  0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 
> > There were 50 or more warnings (use warnings() to see the first 50)
> > 
> > The thing is that the caracter "Wuerfel" is random generated
> > and it shouldn' t  be significant.
> > 
> > It would be realy great if you could help me. 
> 
> 
> >The most obvious guess is that sb_ek_ber probably is not Poisson
distributed.
> 
> I tried it in another statistical  programm and I got the values that I
expected but for my bachelortheses I need to do it in R.
> 
And what do you expect us to do with that bit of information?

Did you check the distribution of the response variable? For all we know, it
might have values of 99999 indicating missing values to "another
statistical program".


-- 
Peter Dalgaard, Professor
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com