RES 341 800 words response

# RES 341 800 words response - which at worst can just slip...

This preview shows page 1. Sign up to view the full content.

Based on your skew value and histogram, discuss the best measures of central tendency and dispersion of your data. Justify your selections. There are three measures of central tendency used in statistics: the mean, the median, and the mode. Each have their special domains of application, and also areas where one should avoid them. The MEAN describes data well when all the data "bunch up" near the middle of one's histogram (frequency distribution), and is relatively symmetric (both sides of the graph look roughly the same). In this case, the mean is a good measure of "where the center of the data" is. When there are outliers -- extreme values -- in the data, the mean does not do well, since even one outlier can affect it greatly. For example, the mean income in the small town where Bill Gates lives is very high, just because of him. In such cases, we use the MEDIAN, which is just the middle of a set of values, once we order them from lowest to highest. Adding in an extreme value does little to affect the median,
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: which at worst can just slip to the next closest value in the middle of the data. For this reason, it's often called a "robust" statistic. When our data are very skewed, we also like to use the median, because it is not affected much by the small proportion of the data with very high or very low values. This is a reason medians are often used for INCOME distributions, especially in countries where wealth is concentrated. Finally, we can also use the MODE. Modes are used in situations like voting, where the candidate with the most votes wins. Or, if we are a clothing manufacturer and can only make ONE size, we would make the mode: the one size the most people wore. Modes aren't used as much in statistics, but are sometimes useful is the data are VERY skewed, or if there seem to be two or more "humps" in the data, indicating a "multi-modal" distribution....
View Full Document

## This document was uploaded on 05/10/2011.

Ask a homework question - tutors are online