Transforming into normal distro

[Marc Feuerstein]

Hello everybody.

I made the K-S test on the dependent variable and saw that it is not
normally distributed (Asymp. Sig. = 0,000).

Is there a data transformation or some magical incantation that exists
somewhere to transform this data into a normally distributed set ?

Bye,

Marc.

[Spousta Jan]

Hi Marc,

The best possible try is to use normal scores RANK ... /NORMAL.

GET FILE='C:\Program Files\SPSS\Cars.sav'.
NPAR TESTS
  /K-S(NORMAL)= weight
  /MISSING ANALYSIS.
RANK
  VARIABLES=weight  (A) /NORMAL /PRINT=YES
  /TIES=MEAN /FRACTION=BLOM .
NPAR TESTS
  /K-S(NORMAL)= Nweight
  /MISSING ANALYSIS.

But beware:
1) The transformation is generally very much non-linear and hard to
generalize
2) In some cases (discrete variables) it is impossible to obtain a
normal distribution even with this method.

Usually one rather tries "standard" transformations (log, 1/x, exp,
arctg...) to obtain something more Gaussian (even if not perfect
Gaussian) or uses non-parametric methods which does not require
normality.

Greetings

Jan

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Marc
Sent: Monday, July 31, 2006 1:13 PM
To: [hidden email]
Subject: Transforming into normal distro

Hello everybody.

I made the K-S test on the dependent variable and saw that it is not
normally distributed (Asymp. Sig. = 0,000).

Is there a data transformation or some magical incantation that exists
somewhere to transform this data into a normally distributed set ?

Bye,

Marc.

[William B. Ware]

Marc,

The assumption of normality pertains to the "conditioned" dependent
variable, or the residuals... not to the dependent variable itself... What
do your residuals look like?

wbw

__________________________________________________________________________
William B. Ware, Professor                         Educational Psychology,
CB# 3500                                       Measurement, and Evaluation
University of North Carolina                         PHONE  [phone]
Chapel Hill, NC      27599-3500                      FAX:   [phone]
Office:  118 Peabody Hall                            EMAIL: [hidden email]
Adjunct Professor                                    School of Social Work
__________________________________________________________________________


On Mon, 31 Jul 2006, Marc wrote:

> Hello everybody.
>
> I made the K-S test on the dependent variable and saw that it is not
> normally distributed (Asymp. Sig. = 0,000).
>
> Is there a data transformation or some magical incantation that exists
> somewhere to transform this data into a normally distributed set ?
>
> Bye,
>
> Marc.
>