GDAtools (version 1.3)

pem: Computes the local and global Percentages of Maximum Deviation from Independance (PEM)

Description

Computes the local and global Percentages of Maximum Deviation from Independance (PEM) of a contingency table.

Usage

pem(x)

Arguments

x

Contingency table. Accepted formats are matrices and 'table' objects.

Value

Returns a list:

peml

Matrix with local percentages of maximum deviation from independance

pemg

Numeric value, i.e. the global percentage of maximum deviation from independance

Details

The Percentage of Maximum Deviation from Independance (PEM) is an association measure for contingency tables (see Cibois, 1993). It is an alternative to khi2, Cramer coefficient, etc.

References

Cibois P., 1993, Le PEM, pourcentage de l'ecart maximum : un indice de liaison entre modalites d'un tableau de contingence, Bulletin de methodologie sociologique, n40, p.43-63. http://cibois.pagesperso-orange.fr/bms93.pdf

See Also

table, chisq.test, assocstats

Examples

## Computes the PEM for the contingency table
## of jazz and age variables
## from the 'Music' example data set
data(Music)
x <- table(Music$Jazz,Music$Age)
pem(x)