OrdinalLogisticBiplot (version 0.4)

plotOrdinalVariable: This function plots in the reduced space an ordered categorical variable.

Description

Graphical representation of a polytomous ordered variable in the reduced space, according to Vicente-Villardon & Hernandez-Sanchez(2014) methodology. It can be choosen some parameters related to the way in which the variable is plotted.

Usage

plotOrdinalVariable(ordinalfVar,nameVariable,estimRows,planex = 1,planey = 2, xi=-3.5,xu=3.5,yi=-3.5,yu=3.5,margin=0,CexVar=0.7,ColorVar="blue", PchVar=0.7,addToPlot=FALSE,showIIC = TRUE,iicxi=-2.5,iicxu=2.5, tol = 1e-04, maxiter = 100, penalization = 0.1)

Arguments

ordinalfVar

The ordinal variable. It must be an ordered factor.

nameVariable

Name of the variable that the user wants to represent.

estimRows

Matrix with the estimated coordinates for the individuals in the reduced dimension.

planex

Dimension for X axis.

planey

Dimension for Y axis.

xi

Minimum value on the x-axis.

xu

Maximum value on the x-axis.

yi

Minimum value on the y-axis.

yu

Maximum value on the y-axis.

margin

This value stablishes the space between the plotted items and the border of the window.

CexVar

Size of the category points. It can be an array with the cex information for each variable.

ColorVar

Color for the variables. It can be an array with the color information for each variable.

PchVar

Symbol for the variables. It could be an array with the pch information for each variable.

addToPlot

Boolean parameter to decide if the user wants to add the ordinal variable representation to an existing plot.

showIIC

Boolean parameter to decide if the user wants to see the item information curves for each variable. Default value is FALSE.

iicxi

Lower limit for the X-axis when plotting item information curves.

iicxu

Upper limit for the X-axis when plotting item information curves.

tol

Tolerance for the iterations.

maxiter

Maximum number of iterations.

penalization

Penalization used to avoid singularities.

References

Vicente-Villardon, J. L. & Hernandez, J. C. & (2014) Logistic Biplots for ordinal data with an application to job satisfaction of doctorate degree holders in Spain. Preprint available at arXiv.

Examples

data(LevelSatPhd)
olbo = OrdinalLogisticBiplot(LevelSatPhd,sFormula=NULL,
  numFactors=2,method="EM")
ordinalfVar = factor(LevelSatPhd[,1],ordered=TRUE)
levels(ordinalfVar) = c("VS","SS","SD","VD")
estimRows = olbo$RowCoords
nameVariable = "Salary"   
plotOrdinalVariable(ordinalfVar,nameVariable,estimRows,planex = 1,
  planey = 2,xi=-1.5,xu=1.5,yi=-1.5,yu=1.5,
  margin=0.2,CexVar=0.7,showIIC = TRUE)