Please bear with me, I'm hopeless at statistics and so very lost right now.
I'm wanting to run a multiple regression with 9 IVs and 1 DV. I want to control for the effect of age (not one of my IVs). What model would I use?
Second question, the age variable is highly skewed towards younger participants, as well as having positive kurtosis. It's not an IV, I just want to control for it, so should I do a data transformation on it to bring it closer to normality?
I want to control for the effect of age (not one of my IVs). What model would I use?
There's nothing wrong with trying not just one model, but many models as that can help you get a better understanding of what is happening in your data.
Just wondering if you know how age is affecting your data? If I didn't know, one thing I'd try is stratify by age. So make 4 groups of age (if you have enough observations, else you can make just 2 or 3 age groups), and think of each group as its own study. So you'd run your model 4 times, getting beta each time. Then take a look at what's happening to your betas, to see how age affects your data.
By stratifying, you avoid using age as an IV; although I'm not sure why you wouldn't want to use age as an IV?
You can do this without transforming the age variable. I don't think I've ever log-transformed the age variable....I guess you could say though that making age groups is a kind of transformation. Anyway, like I said at the start, you can try an age transformation too and run your models again, as this might help you to understand your data. I'd do no transformation first, for sure.