transforming an extremely left-skewed data

[Halbana Tarmizi]

Hi all:

I really appreciate this forum. I have been a long time lurker, and now I
need some advices dealing with my data.
I have survey data that I collected for 20 variables. Now, the problem is
three of those variables are extremely left-skewed (most of the responses
are at the highest end). I tried several transformation, i.e., reflect &
inverse, or logarithm, or square root, but couldn't get it normalized.

So, my question is..is there any other transformation that I should try to
bring those data close to normal? I will need them to do confirmatory factor
analysis. Any suggestion is appreciated. Thank you.

Halbana
nebraska - omaha

=====================
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

[Alexander J. Shackman-2]

you need to reflect the data, transform it, then re-reflect it

take care that this exactly preserves the rank order of participants within
a variable pre vs. post transform

see
http://psyphz.psych.wisc.edu/web/pubs/2006/Shackman.AnxietySelectively.Emotion.pdffor
an example

hth, alex shackman


On Dec 6, 2007 10:40 AM, Halbana Tarmizi <[hidden email]> wrote:

> Hi all:
>
> I really appreciate this forum. I have been a long time lurker, and now I
> need some advices dealing with my data.
> I have survey data that I collected for 20 variables. Now, the problem is
> three of those variables are extremely left-skewed (most of the responses
> are at the highest end). I tried several transformation, i.e., reflect &
> inverse, or logarithm, or square root, but couldn't get it normalized.
>
> So, my question is..is there any other transformation that I should try to
> bring those data close to normal? I will need them to do confirmatory
> factor
> analysis. Any suggestion is appreciated. Thank you.
>
> Halbana
> nebraska - omaha
>
> =====================
> 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
>

--
Alexander J. Shackman
Laboratory for Affective Neuroscience
Waisman Laboratory for Brain Imaging & Behavior
University of Wisconsin-Madison
1202 West Johnson Street
Madison, Wisconsin 53706

Telephone: [phone]
FAX: [phone]
EMAIL: [hidden email]
http://psyphz.psych.wisc.edu/~shackman
Calendar {still under construction}:
http://www.google.com/calendar/embed?src=ajshackman%40gmail.com

=====================
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

[Spousta Jan]

Hi,

The first choices in your case are in my opinion these:

X_transf = (x + a) ** b

And

X_transf = exp(x + a)

Where a and b are real constant you should set accordingly to your data (try some settings an look at the transformed distribution; b > 1).

HTH

Jan



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Halbana Tarmizi
Sent: Thursday, December 06, 2007 5:40 PM
To: [hidden email]
Subject: transforming an extremely left-skewed data

Hi all:

I really appreciate this forum. I have been a long time lurker, and now I need some advices dealing with my data.
I have survey data that I collected for 20 variables. Now, the problem is three of those variables are extremely left-skewed (most of the responses are at the highest end). I tried several transformation, i.e., reflect & inverse, or logarithm, or square root, but couldn't get it normalized.

So, my question is..is there any other transformation that I should try to bring those data close to normal? I will need them to do confirmatory factor analysis. Any suggestion is appreciated. Thank you.

Halbana
nebraska - omaha

=====================
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



_____

Tato zpráva a všechny připojené soubory jsou důvěrné a určené výlučně adresátovi(-ům). Jestliže nejste oprávněným adresátem, je zakázáno jakékoliv zveřejňování, zprostředkování nebo jiné použití těchto informací. Jestliže jste tento mail dostali neoprávněně, prosím, uvědomte odesilatele a smažte zprávu i přiložené soubory. Odesilatel nezodpovídá za jakékoliv chyby nebo opomenutí způsobené tímto přenosem.

This message and any attached files are confidential and intended solely for the addressee(s). Any publication, transmission or other use of the information by a person or entity other than the intended addressee is prohibited. If you receive this in error please contact the sender and delete the message as well as all attached documents. The sender does not accept liability for any errors or omissions as a result of the transmission.

-.- --

=====================
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