Two data sets, A and B, are summarized in histograms. Data set A's intervals are all 10 units higher than the corresponding intervals in data set B. For example, if data set B has values in [10, 20), data set A has values in [20, 30). Both data sets have the same number of integers and identical frequency distributions.

If all integers in both data sets were replaced by the midpoint of their respective intervals, what would be the difference between the mean of data set A and the mean of data set B?