hi everyone
I'm in a fix, could someone please help me here?

how do I explain in linear regression the phrase "when the other variables are held constant" or "when they are controlled for", what is the simplest way to explain it? The audience is people who do not know what regression is.

Thanks very much
love satchi

I'd just use an example - so e.g. if you want to measure the effects of eating chocolate on mood, when you've already taken into account the effect of gender.

hi Sneaks
yes that's it!!! I completely forgot about giving examples! That will be easier then.
thanks a lot
love satchi

can another example be given here?

erm,

if you want to see whether liking justin bieber predicts idiocy, after controlling for age?

good one Sneaks
:-) :-) :-)
love satchi

But how do you do that - how do you actually hold the other variables constant? I don't understand the process...

Can anyone please tell me?

you do it using hierarchical regression. In SPSS, you do it by going through the normal regression process, but you use different 'blocks'. In block 1 you would include things you wanted to control for (e.g. gender). and in block 2 anything you wanted to test after the effects of gender. You can access the different 'blocks' by just clicking a button in the regression menu.

Shown about 1.20 in this vid http://www.youtube.com/watch?v=Zms3teHKUL0

does that mean the variable being tested for; example idiocy is tested after gender is tested in liking justin? im still unclear to explain this controlling factor in simple terms.. im actually new to spss.. would appreciate some help.. thank you very muchy :)

yeah, so if you want to predict whether liking justin bieber predicts idiocy, but you want to see this relationship, after you have controlled for the effect of age on idiocy.