I’m designing an experiment for one of my psychology courses and ended up with a design with one independent variable, one dependent variable and one moderator. Which analysis method, which I also will include in the proposal, should I use?
I can't provide a detailed answer given that I don't know all the details about what you want to do. A detailed answer would also be solving your class problem for you. But I would consider regression analysis. You can include both the IV and the moderator as causal and interacting factors in your model. There are other approaches, but this is what seems the most direct to me. To identify the best type of regression for your topic, you will have to consider what type of data you have, if there are any dependencies in your data, and the assumptions of each type of regression. I hope that helps get you started in the right direction. Best of luck.
Christopher A Varnon First of all thank you for answering, it’s greatly appreciated. Secondly, as I read more about regression analysis a problem came up which is, in the design there are two IVs which is fine but one is binary and the other one is continuous and the dependent variable is continuous, also. I’m guessing I should use linear regression based on what I’ve read, but my question is whether that would create a problem for the binary variable or not? Or if I should totally use another method?
Also to note, I don’t expect multicollinearity between the IVs and there shouldn’t be a confounding variable.
It sounds like are going in the right direction. Linear regression is great for a continuous DV, and it can handle both continuous and categorical IVs. It will take a little practice to interpret your regression equations, but I find this approach very useful and I use it frequently in my work.