My analyses have a dichotomous independent variable, one continuous mediator, and one continuous dependent variable. I have not come across any guidance regarding recommended sample size for this type of model. I'd appreciate any advice.
I don't think the sample size is a problem unless you use SEM analysis for mediation, also because it is mostly used with boottrapping of a thousand samples or more.
Thank you, Cristian Ramos-Vera. Do you have a reference for this information?
Some journals require a justification of sample size for publication, and there are a few well cited articles regarding sample size in mediation (e.g., Fritz & MacKinnon, 2007). However, I cannot find any articles advising the same for mediation with a categorical antecedent; if you are aware of any then do let me know.