I conclude that eventhough it is always better to implement fourth reuse regardless of the project given that 
all the real measurements are smaller for the higher levels of reuse. However, different reuse levels may 
achieve higher level of value reduction and therefore when normalized with respect to the initial value, different
reuse strategies rank higher than the others. This conclusion can be easily observed by comparing normalized
values vs. non-normalized values. In the non-normalized case (not dividing the final value by the initial and
multiplying by 100) reuse strategies always rank the same, first reuse is the worst and fourth is the best, but
in the case of normalized values (final/initial*100) strategy ranks vary.