catseyes (version 0.2.5)

catseyes: catseyes

Description

The catseyes() function is used to plot catseye interval(s) onto a an existing basic R plot background. Catseye plots illustrate the normal distribution of the mean (picture a normal bell curve reflected over its base and rotated 90 degrees), with a shaded confidence interval; they are an intuitive way of illustrating and comparing normally distributed estimates, and are arguably a superior alternative to standard confidence intervals, since they show the full distribution rather than fixed quantile bounds. The catseyes() function requires pre-calculated means and standard errors (or standard deviations), provided as numeric vectors; this allows the flexibility of obtaining this information from a variety of sources, such as direct calculation or prediction from a model -- see examples below. NOTE: The drawn vertical range of the outline spans 99.8% of the distribution of the mean.

Usage

catseyes(
  x,
  ymean,
  yse,
  dx = 0.1,
  conf = 0.95,
  se.only = TRUE,
  col = "black",
  shade = rgb(0.05, 0.05, 0.05, 0.2),
  lwd = 1,
  plot.mean.line = FALSE,
  fTransform = NULL
)

Arguments

x

numeric horizontal position(s); if factor, will be converted to integer in factor level order

ymean

numeric mean(s)

yse

numeric standard error(s); may use standard deviation(s) for population level plots

dx

specifies the width (in x direction) of the catseye interval(s)

conf

specifies the confidence of the confidence interval (conf=.95 for alpha=.05)

se.only

boolean, if TRUE (default) will shade only +/- 1 standard error about the mean, overriding conf, otherwise if FALSE will shade the confidence interval (per conf) about the mean

col

specifies the color of the outline of the catseye, as well as the interval point & line, if shown

shade

specifies the color of the shaded confidence region

lwd

sets the line width of the interval and outline

plot.mean.line

boolean, draws a horizontal line at the position of the mean if TRUE

fTransform

Optional function to transform catseye plot from normal distribution (as with analyzing log-tranformed data, see example under catseyesplot)

References

Cumming, G. (2014). The new statistics: Why and how. Psychological Science, 27, 7-29. <doi:10.1177/0956797613504966> pmid:24220629 http://www.psychologicalscience.org/index.php/publications/observer/2014/march-14/theres-life-beyond-05.html

Examples

# NOT RUN {
#Show catseye plots for 4 groups with means of c(-3,2,-1,6)
#    and standard errors of c(1,2,4,3)
plot(NULL,xlim=c(.5,4.5),ylim=c(-10,10),xlab="",ylab="",main="4 Groups",xaxt="n")
axis(1,at=1:4,labels = c("Group1","Group2","Group3","Group4"))
catseyes(1:4,ymean=c(-3,2,-1,6),yse=c(1,2,4,3))
#Optionally, add points and lines (usually lines only when joining time sequence)
lines(1:4,c(-3,2,-1,6),type="b")
# }