TRANSFORMING DATA

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

STICKY

TRANSFORMING DATA

Postby STICKY » Mon Sep 03, 2007 2:12 pm

please can someone help?
I have been advised to try transformations of my data using log(x+c) or
log(max(x,c)) for some constant c.
I am using SPSS 12 and can't find this in the list of functions.
Many thanks
Smash
Moderator
Posts: 233
Joined: Tue Aug 07, 2007 11:48 am

Postby Smash » Mon Sep 03, 2007 9:53 pm

???definitly there are some functions in compute variables
ln(x) or lg10(x)
sticky
Posts: 3
Joined: Mon Sep 03, 2007 2:50 pm

Postby sticky » Tue Sep 04, 2007 5:41 am

Thank you for responding.
There are no functions under (x) and I am familiar with LG10 but am no Statistician!!
Not sure if log (x + c) is the same as LG10 and have no idea what log(max(x,c)) for some constant c actually means. But hoping that you may.
I have continuous data that is positively skewed but with some zero and near zero values which I am advised need careful handling - hence the transformation prescription. Hope that makes sense?
Smash
Moderator
Posts: 233
Joined: Tue Aug 07, 2007 11:48 am

Postby Smash » Tue Sep 04, 2007 9:22 am

Ok,
Use c=0.1 as a constant, so you would have computes as:
compute x2=ln(x+0.1).
or
compute x2=max(ln(x,0.1)).
or the same with lg10.
sticky
Posts: 3
Joined: Mon Sep 03, 2007 2:50 pm

Postby sticky » Tue Sep 04, 2007 5:17 pm

Thanks for that, i'll give it a try :D :D

Who is online

Users browsing this forum: No registered users and 1 guest

cron