Provide a brief statistical argument for each of the following:

A, If I were to find a couple of outliers in a sample of size 100, I would consider the reasons why the outliers exist, but they would not be a major factor in the study outcome. The reasoning behind that is that my population is quite large and the data sample is relatively large. I would however calculate the Inter Quartile Range, just to confirm that the outliers have been correctly identified.

B, Assuming we want to examine a sample of the population of at least 100 elements, we might want to calculate the average monthly expenditure that this sample incurs. After data collection we notice that we have collected a couple of outliers or even a couple of values that were twice as big as the next highest value.

We should first of all make sure that the values are correct by verifying that incorrect values have not been reported, if the values are correct we should evaluate the situation, in a case where the number of elements is so limited, such high values could substantially change the value of our average and standard deviation.

We have two possibilities to make our case study reliable, the first (the simplest) is to eliminate the outliers thus keeping all the statistics true and not altered. The second is to modify the outliers by bringing them closer to the maximum value or to the minimum value in order to keep the number of elements unchanged by slightly modifying our statistical results, it will be necessary to evaluate case by case to decide which method to use in my specific case I would focus more on changing the outliers in order to make the statistics as truthful as possible.