When planning analyses for a study it is important to be clear about what type of data you’ll have.  Once you know if the outcome measure is continuous, categorical, or time-to-event, you can choose the appropriate methods.  Understanding your data is very important, so do not skip the step of looking at descriptive statistics first, including looking at distributions and graphs whenever possible.  Next, you can start to look at associations between variables (bivariable) to get a sense of how variables relate to one another.  This step can and should also use graphs and tables to visualize data whenever helpful.  Once these relationships are understood, modeling techniques can be used.  Models allow for both unadjusted and adjusted estimates to be calculated and can include more than one covariate.  Modeling can be used to evaluate potential confounding along with effect modification.