Togaware DATA MINING
Desktop Survival Guide
by Graham Williams
Google

Stacks Versus Lines

As pointed out at http://peltiertech.com/WordPress/2008/08/27/stacked-vs-clustered/a line chart might be better than a stacked bar chart. Depends on situation. Example from Hadley Wickham on r-help 28 Aug 2008. The example from the above URL might be clearer to illustrate the point.



quarter <- as.factor(sample(c("Q1", "Q2", "Q3", "Q4"), 100, replace = TRUE))
year <- as.factor(sample(c(seq(from=2000, to=2008)), 100, replace = TRUE))
category <- as.factor(sample(c(seq(from=1, to=4)), 100, replace = TRUE))
test <- data.frame(quarter, year, category)
tabdf <- as.data.frame(with(test, table(category, quarter, year)))

library(ggplot2)
qplot(year, Freq, data=tabdf, geom="line", colour = category, facets =
 quarter ~ . , group = interaction(category, quarter))
# OR
qplot(year, Freq, data=tabdf, geom="line", colour = quarter, facets =
category ~ . , group = interaction(category, quarter))

# If you _really_ want stacking:

qplot(year, Freq, data=tabdf, geom="area", fill = quarter, facets =
category ~ . , group = interaction(category, quarter))

# OR

qplot(year, Freq, data=tabdf, geom="bar", stat="identity", fill =
quarter, facets =  category ~ .)

# OR finally, like what you originally asked for

qplot(quarter, Freq, data=tabdf, geom="bar", stat="identity", fill =
category, facets = . ~ year)



Copyright © Togaware Pty Ltd
Support further development through the purchase of the PDF version of the book.
The PDF version is a formatted comprehensive draft book (with over 800 pages).
Brought to you by Togaware. This page generated: Sunday, 22 August 2010