I was going through a sample data set from my school textbook and using minitab i got 60.98 for the mean. Median was 61 and mode is 62. For some otehr data sets as well the mean, median mode are away by about a fraction or one whole number. The distribution or variance is large comparitively for example they deal with numbers from 35 to higher 70s. When i make the histograms some of them appear kind of bell shaped but there's a bit of a skew here and there .

What i wanted to know is do i have to use Empirical rule ONLY when mean , median mode are exactly the same ( without rounding off or anything )? and when i see a perfect bell shaped curve?