Correlations or Regression?


Would someone please be able to advise me on whether I should use correlations or a regression. I want to compare each IV with my DV.
  • My DV is continuous.
  • Some of my IVs are continuous, i.e., interval? (if a correlation, a Pearsons?)
  • Some of my IVs are each measured on one scale each, i.e., ordinal? (if a correlation, a Spearman's rank order?)
My issue is that a regression does not allow me to see the direction of the correlation? But, it would allow me to compare if the IVs influence each other, and then influence the DV? Is there a way to compare the influence of IVs on each other, and then that influence on the DV, through simple correlations?

Thank you so much in advance if you can offer any help!


Less is more. Stay pure. Stay poor.
Regression is probably the best course since it adjusts for the other covariates. Yes, it does show you the direction of that association in whether the model coefficients are positive or negative.