ANOVA with Optimal Scaling technique and unequality of variance

[Sebastien-6]

Hi, I'm a PhD student in Ecology. I've got experiments that follow a two-
way ANOVA design. My variances are unequal even after classic
transformation (sqrt, Log). I would like to use ANOVA using the Optimal
Scaling technique available on SPSS. Does ANOVA with optimal scaling
require equal variances?
Thank you.
Sébastien.

[Kooij, A.J. van der]

I guess you mean regression using optimal scaling (CATREG)? with nominal scaling level voor dependent variable this is ANOVA. No equal variances required.

Regards,
Anita van der Kooij
Data Theory Group
Leiden University

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Sebastien
Sent: 15 June 2006 13:42
To: [hidden email]
Subject: ANOVA with Optimal Scaling technique and unequality of variance


Hi, I'm a PhD student in Ecology. I've got experiments that follow a two- way ANOVA design. My variances are unequal even after classic transformation (sqrt, Log). I would like to use ANOVA using the Optimal Scaling technique available on SPSS. Does ANOVA with optimal scaling require equal variances? Thank you. Sébastien.

**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
**********************************************************************

[Sebastien-6]

Thank you for your answer. I've got some further questions:
1) Do you have any reference I can use to justify that equal variances is
not required?
2) Are there any assumptions behind optimal scaling?
3) What are the risks of type 1 and 2 errors with the optimal scaling
technique when the data have not equal variances?
It has been suggested to me to use Aligned Rank Transformation to study
interaction effect in a two-way ANOVA design, but it appeared that this
method is not suitable when variances are not equal, which is the case of
my data. That's why I'm looking for another method and I want to make sure
that this method can be used with confidence.
Sébastien.

[Kooij, A.J. van der]

this should be: ... with nominal scaling level for INdependent variables (and numerical level for dependent), sorry.

________________________________

From: Kooij, A.J. van der
Sent: Thu 15/06/2006 16:55
To: '[hidden email]'
Subject: RE: ANOVA with Optimal Scaling technique and unequality of variance



I guess you mean regression using optimal scaling (CATREG)? with nominal scaling level voor dependent variable this is ANOVA. No equal variances required.

Regards,
Anita van der Kooij
Data Theory Group
Leiden University

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Sebastien
Sent: 15 June 2006 13:42
To: [hidden email]
Subject: ANOVA with Optimal Scaling technique and unequality of variance


Hi, I'm a PhD student in Ecology. I've got experiments that follow a two- way ANOVA design. My variances are unequal even after classic transformation (sqrt, Log). I would like to use ANOVA using the Optimal Scaling technique available on SPSS. Does ANOVA with optimal scaling require equal variances? Thank you. Sébastien.


**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
**********************************************************************