Anova two way non parametric....

[luciano basso]

Hi Team,

I would like to analyze the effect of two independent factors and their interaction in an ordinal variable. There is a  ANOVA two way non parametric? and a test of multple comparisons? Someone has syntax for this?

Suggestions are welcome....

Luciano Basso
University of São Paulo
Motor Behavior Lab

[Bruce Weaver]

Please provide more information.  E.g.,

1. What are the sample sizes?
2. What exactly is the dependent variable?  
3. What is the design -- i.e., both factors between-Ss, both with repeated measures, or one of each?

luciano basso wrote

Hi Team,

I would like to analyze the effect of two independent factors and their
interaction in an ordinal variable. There is a  ANOVA two way non
parametric? and a test of multple comparisons? Someone has syntax for this?

Suggestions are welcome....

Luciano Basso
University of São Paulo
Motor Behavior Lab

[Marta Garcia-Granero]

luciano basso wrote:
> Hi Team,
>
> I would like to analyze the effect of two independent factors and
> their interaction in an ordinal variable. There is a  ANOVA two way
> non parametric? and a test of multple comparisons? Someone has syntax
> for this?
>
> Suggestions are welcome....

Conover & Iman suggest using a two-way ANOVA on ranks instead fo the
original data. Shirley has a nice paper (Shirley EAC. /Applications of
Ranking Methods to Multiple Comparison Procedures and Factorial
Experiments/. Apps Statist 1987; 36(2): 205-213) with a modification of
the first method.

Anyway, I seem to recall that both approaches had been criticized, but I
don't remember where I read that.

HTH,
Marta GG (leaving my city, Pamplona, tomorrow morning, due to the
imminence of "San Fermines" week)


--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Marta Garcia-Granero]

This might help (code for Shirley's approach to the non parametric
two-way ANOVA):

* Replace "depvar" by the name of the dependent variable *.
RANK
  VARIABLES=depvar  (A) /RANK INTO Ranked.

* Replace "factor1" and "factor2" by the two independent factors *.
UNIANOVA
  Ranked  BY factor1 factor2
  /OUTFILE=EFFECT ('C:\TEMP\report_.sav')
  /DESIGN = factor1 factor2 factor1*factor2 .

* Next lines of code must be left unchanged *.
MATRIX.
PRINT /TITLE='KRUSKAL-WALLIS FOR FACTORIAL DESIGNS'.
GET depname /VAR=depvar_ /FILE='c:\temp\report_.sav'.
GET tnames  /VAR=source_ /FILE='c:\temp\report_.sav'.
GET ss      /VAR=ss      /FILE='c:\temp\report_.sav'.
GET df      /VAR=df      /FILE='c:\temp\report_.sav'.
PRINT depname(1)
 /FORMAT='A8'
 /TITLE='Dependent variable:'.
COMPUTE k=NROW(ss).
COMPUTE varri=ss(k)/df(k).
COMPUTE chi2=ss/varri.
COMPUTE chi2sig=MAKE(k-3,1,0).
LOOP i=1 TO k-3.
- COMPUTE chi2sig(i)=1-CHICDF(chi2(i),df(i)).
END LOOP.
PRINT {ss(1:(k-3)),df(1:(k-3)),chi2(1:(k-3)),chi2sig(1:(k-3))}
 /FORMAT='F8.3'
 /RNAMES=tnames
 /CLABELS='SS','DF','Chi^2','Sig'
 /TITLE='Chi-square tests & asymptotic significance'.
PRINT varri
 /FORMAT='F8.3'
 /RLABEL='VAR(Ri)='
 /TITLE='Ranks Variance'.
END MATRIX.

DELETE VARIABLES Ranked.

Best regards,
Marta GG

>
> I would like to analyze the effect of two independent factors and
> their interaction in an ordinal variable. There is a  ANOVA two way
> non parametric? and a test of multple comparisons? Someone has syntax
> for this?
>


--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Bruce Weaver]

Marta García-Granero-2 wrote

luciano basso wrote:
> Hi Team,
>
> I would like to analyze the effect of two independent factors and
> their interaction in an ordinal variable. There is a  ANOVA two way
> non parametric? and a test of multple comparisons? Someone has syntax
> for this?
>
> Suggestions are welcome....

Conover & Iman suggest using a two-way ANOVA on ranks instead fo the
original data. Shirley has a nice paper (Shirley EAC. /Applications of
Ranking Methods to Multiple Comparison Procedures and Factorial
Experiments/. Apps Statist 1987; 36(2): 205-213) with a modification of
the first method.

Anyway, I seem to recall that both approaches had been criticized, but I
don't remember where I read that.

HTH,
Marta GG (leaving my city, Pamplona, tomorrow morning, due to the
imminence of "San Fermines" week)


--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

Hi Marta.  Apparently, there are problems with the interaction term when factorial ANOVA is performed on ranks.  See the Wikipedia page for a brief explanation and some references:

    http://en.wikipedia.org/wiki/Analysis_of_variance#ANOVA_on_ranks

Cheers,
Bruce