I think that the first and most essential step in data analysis is to look at your data. It is always a good idea to first subject data to what I like to call the "interoccular trauma test" - By interoccular trauma I mean look at the data and see if something jumps out at you as strange. It is also extremely useful to look at descriptive statistics such as the minimum and maximum value, the mean and standard deviation for variables of interest. Graphs provide another means by which to evaluate our data. All of these strategies should be used prior to hypothesis testing.