Spine Plots and Spinograms
Spine plots are a special cases of mosaic plots, and can be seen as a generalization of stacked (or highlighted) bar plots. Analogously, spinograms are an extension of histograms.
# S3 method for default spineplot(x, y = NULL, breaks = NULL, tol.ylab = 0.05, off = NULL, ylevels = NULL, col = NULL, main = "", xlab = NULL, ylab = NULL, xaxlabels = NULL, yaxlabels = NULL, xlim = NULL, ylim = c(0, 1), axes = TRUE, …)
# S3 method for formula spineplot(formula, data = NULL, breaks = NULL, tol.ylab = 0.05, off = NULL, ylevels = NULL, col = NULL, main = "", xlab = NULL, ylab = NULL, xaxlabels = NULL, yaxlabels = NULL, xlim = NULL, ylim = c(0, 1), axes = TRUE, …, subset = NULL)
- an object, the default method expects either a single variable (interpreted to be the explanatory variable) or a 2-way table. See details.
"factor"interpreted to be the dependent variable
y ~ xwith a single dependent
"factor"and a single explanatory variable.
- an optional data frame.
- if the explanatory variable is numeric, this controls how
it is discretized.
breaksis passed to
histand can be a list of arguments.
- convenience tolerance parameter for y-axis annotation. If the distance between two labels drops under this threshold, they are plotted equidistantly.
- vertical offset between the bars (in per cent). It is fixed to
0for spinograms and defaults to
2for spine plots.
- a character or numeric vector specifying in which order the levels of the dependent variable should be plotted.
- a vector of fill colors of the same length as
levels(y). The default is to call
- main, xlab, ylab
- character strings for annotation
- xaxlabels, yaxlabels
- character vectors for annotation of x and y axis.
levels(x), respectively for the spine plot. For
xaxlabelsin the spinogram, the breaks are used.
- xlim, ylim
- the range of x and y values with sensible defaults.
- logical. If
FALSEall axes (including those giving level names) are suppressed.
- additional arguments passed to
- an optional vector specifying a subset of observations to be used for plotting.
spineplot creates either a spinogram or a spine plot. It can
be called via
spineplot(x, y) or
spineplot(y ~ x) where
y is interpreted to be the dependent variable (and has to be
x the explanatory variable.
x can be
either categorical (then a spine plot is created) or numerical (then a
spinogram is plotted). Additionally,
spineplot can also be
called with only a single argument which then has to be a 2-way table,
interpreted to correspond to
table(x, y). Both, spine plots and spinograms, are essentially mosaic plots with
special formatting of spacing and shading. Conceptually, they plot
\(P(y | x)\) against \(P(x)\). For the spine plot (where both
\(x\) and \(y\) are categorical), both quantities are approximated
by the corresponding empirical relative frequencies. For the
spinogram (where \(x\) is numerical), \(x\) is first discretized
breaks argument) and then
empirical relative frequencies are taken. Thus, spine plots can also be seen as a generalization of stacked bar
plots where not the heights but the widths of the bars corresponds to
the relative frequencies of
x. The heights of the bars then
correspond to the conditional relative frequencies of
x group. Analogously, spinograms extend stacked
The table visualized is returned invisibly.
Friendly, M. (1994), Mosaic displays for multi-way contingency tables. Journal of the American Statistical Association, 89, 190--200. Hartigan, J.A., and Kleiner, B. (1984), A mosaic of television ratings. The American Statistician, 38, 32--35. Hofmann, H., Theus, M. (2005), Interactive graphics for visualizing conditional distributions, Unpublished Manuscript. Hummel, J. (1996), Linked bar charts: Analysing categorical data graphically. Computational Statistics, 11, 23--33.
## treatment and improvement of patients with rheumatoid arthritis treatment <- factor(rep(c(1, 2), c(43, 41)), levels = c(1, 2), labels = c("placebo", "treated")) improved <- factor(rep(c(1, 2, 3, 1, 2, 3), c(29, 7, 7, 13, 7, 21)), levels = c(1, 2, 3), labels = c("none", "some", "marked")) ## (dependence on a categorical variable) (spineplot(improved ~ treatment)) ## applications and admissions by department at UC Berkeley ## (two-way tables) (spineplot(margin.table(UCBAdmissions, c(3, 2)), main = "Applications at UCB")) (spineplot(margin.table(UCBAdmissions, c(3, 1)), main = "Admissions at UCB")) ## NASA space shuttle o-ring failures fail <- factor(c(2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 2, 1, 2, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1), levels = c(1, 2), labels = c("no", "yes")) temperature <- c(53, 57, 58, 63, 66, 67, 67, 67, 68, 69, 70, 70, 70, 70, 72, 73, 75, 75, 76, 76, 78, 79, 81) ## (dependence on a numerical variable) (spineplot(fail ~ temperature)) (spineplot(fail ~ temperature, breaks = 3)) (spineplot(fail ~ temperature, breaks = quantile(temperature))) ## highlighting for failures spineplot(fail ~ temperature, ylevels = 2:1)
[Example files for LinkedIn Learning course:](https://linkedin-learning.pxf.io/rweekly_spineplot) ```r # Description: Example file for spineplot # main idea: creating spineplots # width of bars = frequency of X # height of bars = frequency of y # Y must be a factor and is the dependent variable spineplot(ChickWeight$weight, ChickWeight$Diet) # spineplot(x,y) # interesting observations # Height of bars indicates obs per diet. Diet 1 has more obs # Width of bars indicates obs per weight. More chicks are weighed between 50 and 100 # or... spineplot(Diet ~ weight, data = ChickWeight) # spineplot(y ~ x) # bells and whistles spineplot(Diet ~ weight, data = ChickWeight, breaks = fivenum(ChickWeight$weight), col = c(5:8), xlab = "Chicken Weight", ylab = "Chicken Diet") # The above is actually a spinogram - like a histogram # example of a true spine plot. Both x and y must be factors spineplot(Diet ~ factor(weight), data = ChickWeight, col = c(5:8)) spineplot(factor(weight) ~ Diet, data = ChickWeight, col = c(1:nlevels(factor(ChickWeight$weight)))) # subset of data # use the 1st half of data. Would make sense to have a more sophisticated selection spineplot(Diet ~ factor(weight), data = ChickWeight, subset = c(1:(578/2)), drop.unused.levels = TRUE, col = c(4:7) ) ```