Graphically

Statistics begins with skills and principles for examining data. In this section, we will learn how to graphically display univariate data. The following principles apply:

• To interpret data, you must first know their context, i.e., their origin, manner of collection, etc.
• Always examine the data. Specifically look for
  1. Regular overall pattern in data.
  2. Important deviations from that pattern.

NOTE: Observations which do not follow the regular pattern (sometimes called outliers) should not be removed from the data unless there is good reason to do so. A ``good'' reason could be that the outlier is a ``typo'' or that the equipment used to measure the observation failed. In other words, the removal of extreme observations should be based on technical or scientific knowledge rather than mere habit. Sometimes, in fact, we are interested in finding outliers. For instance, universities like to give scholarships to students who have unusually high scores on standardized tests such as the ACT or SAT.