Figure 4: Tests to assess statistical significance of difference between data sets that are or could be paired