Greetings,

I am trying to make sense of the design for my Master's thesis. I am looking at the relationship between types of play and anxiety in children. My independent variables are three types of play (continuous, two measured in hours and one on a Likert scale), and my dependant variable is a score on anxiety scale. I also want to look at three potential confounders which are continuous as well. What would be the best way to figure out the 'pure' relationships between my three IVs of interest and DV while accounting for the confounders effect? Will I be able to do that by running a multiple regression with 5 IVs? Or would it be better to divide my confounders into subcategories and look at the relationships between my IVs and DV across the different levels of confounders? Since there are two of them with three levels each, that seems like an overcomplicated design. Please advise which approach to take. Thanks!