NAME: A Dataset That Is 44% Outliers TYPE: Census SIZE: 43 observations, one variable with case labels DESCRIPTIVE ABSTRACT: The data consist of the numbers of days served in office for the 43 Presidents of the United States as of 4 February 2004. SOURCE: The data were extracted from tables in the 2001 _World Almanac_ . VARIABLE DESCRIPTIONS: Columns Variables 1-14 Last name of President (text, spelled with no embedded spaces) 15-18 days in office (counts) Values are separated by (a variable number of) blanks. There are no missing values. STORY BEHIND THE DATA: The data were gathered to illustrate kinds of outliers. The main pattern in the data is that the majority of presidents served either one or two full terms. However, about 44% of the observations deviate from this pattern, and could be considered outliers. PEDAGOGICAL NOTES: The data illustrate outliers that are not mistakes and not observations that are unusually high or low. The reasons for them are all interesting historically. They illustrate that "outliers" need not be errors but may instead be particularly interesting cases. The data also illustrate that different data displays may differ in their ability to reveal interesting data structure. SUBMITTED BY: Robert W. Hayden 82 River Street Ashland, NH 03217 USA hayden@mail.plymouth.edu