Friday, September 20, 2013

Outliers

It's less crowded on the edges of the bell-shaped curve.
Welcome to the fringe.

   Statisticians are fond of scoring everything in numerical gradients, finding the average and the median in every set of results. When a set of scores can offer the average -- also known as the mean -- these results can then be displayed in a manner that shows how far each score deviates from the mean. The resulting graph then takes the shape of a bell.
   A simple calculation of the range of deviations can bring another average, or a standardized deviation from the mean. Within the bell curve, it turns out that 95 percent of all the scores display within two standard deviations on either side of the mean, which is shown as a vertical line down from the top of the bell.
   Medicine makers and physicians rely on this to assure patients that a drug works. They seldom note that for some -- the outliers, those who are not within the Great Middle of the bell -- the drug either doesn't work, or it works too well, with potentially serious side effects.
   Moreover, the medical profession isn't the only one to use the bell curve in analyzing results and effectiveness or in describing events.
   In any instance where number scores can be established -- whether in classroom tests or human behavior -- the bell curve shows up. Teachers use it in grading tests. Sociologists use it in judging behavior.
   Society has a strong urge to pressure members to "get with the program" and to "be part of the team." Some, however, the 2.5 percent on the fringes on either side of the Great Middle -- are quite comfortable in knowing they are not "average."

No comments:

Post a Comment