plotClarkeGrid: Plot a Clarke Error Grid

In ega: Error Grid Analysis

Description

The function uses ggplot to draw the Clarke error grid lines according to the criteria described in the original publication by Clarke et. al. (see reference below). If zones have not already been assigned via the zones parameter, the function getClarkeZones is called first. The values in referenceVals and testVals are then superimposed as a scatter plot. Some basic plot parameters can be specified as arguments, but the return value can also be stored and modified further before plotting (see examples and vignette).

Usage

plotClarkeGrid(referenceVals, testVals, title = "Clarke Error Grid",
  xlab = "", ylab = "", linesize = 0.5, linetype = "solid",
  linecolor = "black", linealpha = 0.6, pointsize = 2, pointalpha = 1,
  zones = NA, unit = "gram")

Arguments

referenceVals	A vector of glucose values obtained via the reference method.
testVals	A vector of glucose values obtained via a non-reference method (e.g. a new meter). The values in this vector are paired with those in referenceVals, so the length should be the same.
title	The main plot title. Defaults to "Clarke Error Grid".
xlab	The x-axis label. Defaults to "Reference Glucose Concentration (mg/dL)".
ylab	The y-axis label. Defaults to "Test Glucose Concentration (mg/dL)".
linesize	The size to be used when drawing the zone lines. The acceptable values are the same as for geom_segment. The default is 0.5.
linetype	The type of line to be used when drawing the zone lines. The acceptable values are the same as for geom_segment. The default is "solid".
linecolor	The color of the zone lines. The acceptable values are the same as for geom_segment. The default is "black".
linealpha	The alpha (transparency) level to be used when drawing the zone lines. The acceptable values are the same as for geom_segment. The default is 0.6.
pointsize	The size to be used when plotting the glucose data points. The acceptable values are the same as for geom_point. The default is 2.
pointalpha	The alpha (transparency) level to be used when plotting the glucose data points. The acceptable values are the same as for geom_point. The default is 1.
zones	An optional character vector specifying the Clarke zones for each paired value. If this is not supplied, getClarkeZones will be called to generate zone labels.
unit	A string specifying the units of measurement. This should be either "gram" (the default) for mg/dl or "mol" for mmol/l.

Value

A ggplot object is returned. If the return value is not assigned, a plot is drawn.

References

Clarke, W. L., D. Cox, L. A. Gonder-Frederick, W. Carter, and S. L. Pohl. "Evaluating Clinical Accuracy of Systems for Self-Monitoring of Blood Glucose." Diabetes Care 10, no. 5 (September 1, 1987): 622-28.

See Also

getClarkeZones ggplot

Examples

library(ggplot2)

# default
plotClarkeGrid(glucose_data$ref, glucose_data$test)

# with options
plotClarkeGrid(glucose_data$ref, glucose_data$test,
              pointsize=1.5,
              pointalpha=0.6,
              linetype="dashed")

# store return value and modify
ceg <- plotClarkeGrid(glucose_data$ref, glucose_data$test)

ceg + theme_gray() +
   theme(plot.title = element_text(size = rel(2), colour = "blue"))

Example output

ega documentation built on May 1, 2019, 8:01 p.m.