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

MysticM
Posts: 1
Joined: Sat Nov 28, 2015 8:48 pm

Hi, I'm new to this forum. I have little statistical knowledge and a conference deadline Monday. I need to analyze a simple data-set. I have variable A (with a condition same for all cases), a moderator variable B and the outcome variable C. I'm trying to see if variable A condition and variable C conditions are present, to what extent variable B mediates variable C? I've tried to write a simple syntax to compute two variables: New_Variable_1 (variable A if Moderator Variable B >0, & Variable 3) and New_Variable_2 (variable A if Moderator Variable B =0 & variable 3). So then I could compare them to variable 3. But when I run descriptives for new variables it gives me N of cases same for both new variables, 1 and 2 (( What am I doing wrong?? And is it even the right way to approach this problem?
I can also share more details, or a syntax file if anyone is willing to help...

Thank you!!
P.S. I'll be taking the statistical analysis class next quarter...
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

a few questions:
I have variable A (with a condition same for all cases),
do you mean that every single case in your dataset is in the same condition? if that is the case, you cannot use variable A as a predictor, as there is no variability you cannot say anything about it's effects.
Imagine you want to know the relation between gender and height: if you only have females you cannot draw conclusions about that.
a moderator variable B and the outcome variable C. I'm trying to see if variable A condition and variable C conditions are present, to what extent variable B mediates variable C?
what do you mean by "are present"?
is B a moderator or a mediator?
I've tried to write a simple syntax to compute two variables: New_Variable_1 (variable A if Moderator Variable B >0, & Variable 3) and New_Variable_2 (variable A if Moderator Variable B =0 & variable 3). So then I could compare them to variable 3.
this is unclear.

What would help is information on your variables like this:
Var A: condition A, nominal, 2 levels
Var B: continuous variable, interval level between -2 and + 20
Var c: etc...

hypothesis: Var A is related to var C, this relation is moderated by Var B.

### Who is online

Users browsing this forum: No registered users and 2 guests