NS FL2 P/S Q26

Post Reply
Amber9
Posts: 29
Joined: Thu Jul 04, 2019 11:08 am

NS FL2 P/S Q26

Post by Amber9 » Sun Jul 14, 2019 11:53 am

Hi,

I'm not sure if I understand the differences between confounding, mediating and moderating variables. Confounding and mediating variables affect both independent and dependent variables whereas the moderating variables affect only the independent variable?

Thanks!
NS_Tutor_Mathias
Posts: 259
Joined: Sat Mar 30, 2019 8:39 pm

Re: NS FL2 P/S Q26

Post by NS_Tutor_Mathias » Tue Jul 16, 2019 1:35 pm

Mediating variables are part of the causal chain. They are essentially how your IV->DV relationship actually works. For example, smoking supposedly correlates with a lower lifetime healthcare cost. It does so through the mediating variable of mortality.

Code: Select all

IV -> Mediating Variable -> DV
Moderating variables change the strength of the relationship. For example, long jump performance might correlate with body height. But body fat percentage or BMI might be moderating variables - how much height predicts long jump performance depends also on how heavy the jumper is.

Code: Select all

IV -------> DV
      ^
      |
      |
moderating variable
And confounding variables influence both the IV and DV, but not really the relationship between the two. For example, exercise levels are inversely correlated with weight gain. However, if a study of these two factors were to not control for age, it would be heavily flawed as age influences both exercise levels and baseline weight gain.

Code: Select all

IV -------> DV
^	    ^
|	    |
|	    |
confounding variable
Post Reply