Everybody knows about average.
But they really do not know its traps. Try to answer to the following question:
Which one of the following two car-dealers is the better one?
(Dealer A) “More than half of the customer-satisfaction scores were over 90 points (out of 100)”
(Dealer B) “Average of the customer-satisfaction score was 55 points (out of 100)”
I bet many of you think “Dealer A” is better than “Dealer B”. But it is not necessarily the case.
Look at the chart showing the score disctribution.
It illustrates the scores are spreat into extremely low and extremely high.
BUT the average is 55 points while , at the same time, over half of the scores are over 90 points.
Dealer A and Dealer B are the same. AVERAGE hides data disctibution, which sometimes causes confusion.