Descriptive statistics

Descriptive statistics are a set of technologies (ie, formulas and procedures) used for describing mathematical properties of a database (eg, factor analysis of all variables in a dataset) or part thereof (eg, central tendency of particular variables).

As a dataset is always specific to a particular group, descriptive statistics are also bound to such particular group. This means that descriptive statistics can always say something about such group, independently of whether the data is usable further or not. That is, these statistics always 'describe' properties of the particular group under research (eg, of a sample). They are often used for assessing the quality of the data in the dataset and, thus, for ascertaining if it is appropriate to continue further into using inferential statistics².

Illustration 1 offers a snapshot of typical descriptive statistics.

Illustration 1. Descriptive statistics
Univariate
Type of analysis	nominal	ordinal	interval
Central tendency	mode	median	mean
Dispersion	modal dispersion	SQR, SPR	standard deviation
~	variation ratio	IQR, IPR	variance
Graphs	pictograph	stem-and-leaf	histogram
~	bar graph	line graph	error bar
~	pie chart	box-and-whisker
Asymmetry	---	rank skewness	skewness
Peakness	---	---	kurtosis
Location	---	percentile	z-scores
Bivariate
Type of analysis	nominal	ordinal	interval
Correlation	---	rho	r
Multivariate
Type of analysis	nominal	ordinal	interval
Grouping	---	cluster analysis	principal component analysis
~	---	---	factor analysis

References

1. full reference in the following format AUTHOR (date work).Title. Reference location, date publication. ISBN/ISSN.

+++ Notes +++

2. When descriptive statistics are used on a whole population it is unnecessary to go any further into inferential statistics, as there is no need to infer mathematical properties of a population whose mathematical properties are already known.

Want to know more?

Wiki of Science - Inferential statistics: This page provides access to other statistic technologies used for making inferences from a sample to a population.