The add vantage of the null hypothesis testing is when used to determine errors in tolerances the null hypothesis will always land you on the side of safety. This is often used when the importance of quantity or quality is extremely important. The case of filling medicine in to a capsule would require a very close tolerance, the null hypothesise could set an undefined error, however this is much safer than a wide tolerance which could result in a dangerous outcome.
A one foot liquid head of water will exert a force of 1.04167 psi. 27.7 inches of water = 1 psig... Mean and dirty 1foot=1/2psig
Usually they would be observations with very low probabilities of occurrence.
Tally You can extend the idea of tally further to what Statisticians refer to as a Stem & Leaf Plot, which is a simple type of histogram. With a Stem & Leaf plot your tallies are placed in 'bins' according to their value, allowing you to observe the distribution of your data. Let's say you have a data set: 0, 7, 10, 13, 18, 19, 21, 29, 55, 57, 59, 59, 60 The Stem & Leaf Plot would look like this: 0|07 1|0389 2|19 3| 4| 5|5799 6|0 i.e. '0' becomes '00' and '7' becomes '07' for the purposes of this Stem & Leaf Plot which needs two digits to illustrate its shape in this way.