Section 2.3 Identifying Anomalies
A second strategy for problem identification is to look for the exception cases, both good and bad. A little later in the book we will learn about how the core of classic methods of statistical inference is to characterize "the center" the most typical cases that occur and then examine the extreme cases that are far from the center for information that could help us understand an intervention or an unusual combination of circumstances. Identifying unusual cases is a powerful way of understanding how things work, but it is necessary first to define the central or most typical occurrences in order to have an accurate idea of what constitutes an unusual case.
Coming back to our farmer friend, in advance of a thunderstorm late last summer, a powerful wind came through the orchard, tearing the fruit off the trees. Most of the trees lost a small amount of fruit: the dropped apples could be seen near the base of the tree. One small grouping of trees seemed to lose a much larger amount of fruit, however, and the drops were apparently scattered much further from the trees. Is it possible that some strange wind conditions made the situation worse in this one spot? Or is it just a matter of chance that a few trees in the same area all lost a bit more fruit than would be typical.
A systematic count of lost fruit underneath a random sample of trees would help to answer this question. The bulk of the trees would probably have each lost about the same amount, but more importantly, that "typical" group would give us a yardstick against which we could determine what would really count as unusual. When we found an unusual set of cases that was truly beyond the limits of typical, we could rightly focus our attention on these to try to understand the anomaly.
You have attempted of activities on this page.
