Summary statistics: Difference between revisions
Content deleted Content added
Simon_J_Kissane (talk) No edit summary |
Simon_J_Kissane (talk) No edit summary |
||
| Line 9: | Line 9: | ||
Statisticians commonly try to describe the observations in |
Statisticians commonly try to describe the observations in |
||
#a measure of [[central tendency]] like the [[arithmetic mean]] |
#a measure of [[central tendency]] like the [[arithmetic mean]] |
||
#a measure of [[statistical dispersion]] like the [[standard deviation]] |
#a measure of [[statistical dispersion]] like the [[standard deviation]] |
||
#a measure of the shape of the distribution like [[skewness] or [[kurtosis]] |
|||
| Line 19: | Line 21: | ||
The [[Gini coefficent]] was originally developed to measure income inequality, but can be used for other purposes as well. |
|||
Revision as of 02:36, 23 September 2001
The easiest way to approach this subject is to focus on what we have and what we want to achieve:
- We have a set of observations which we want to summarize.
- We want to communicate as much as possible as simply as possible.
Statisticians commonly try to describe the observations in
- a measure of central tendency like the arithmetic mean
- a measure of statistical dispersion like the standard deviation
- a measure of the shape of the distribution like [[skewness] or kurtosis
There are other alternatives, of course. The median and mode are both measures of central tendency. To describe the statistical dispersion, we can use the statistical range, the interquartile range, or the absolute deviation.
The Gini coefficent was originally developed to measure income inequality, but can be used for other purposes as well.
back to statistical theory -- summarizing statistical data