Example, with R
The simplest dotplots are known as rugplots. (For clarity, we have labeled points with their
values.)
You can draw a rugplot using a pencil and ruler, or with R
Note, because they overlap less, dashes or crosses are often preferred to dots.
With R: rug(y) would ADD a rugplot to an existing graph.
Definition and Use
- Dotplots are used to display frequency distributions of individual values.
- There is no pooling of values into class intervals.
- Despite their name, dotplots may employ a variety of symbols other than simple dots.
Tips and Notes
- Rugplots can be useful for small sets of values, but may be an unsuitable way to display distributions that have non-unique (tied) values. Histogram-type dotplots (also known as dot histograms) display tied values one above the other, as here.
- Dot histograms provide rather more information than standard histograms, because actual values are plotted - not class intervals.
- Since the systematic ordering of points one above the other can introduce arbitrary and spurious patterns, some authors prefer jittered dotplots - where each values is plotted against a uniform random number.
- Dotcharts are sometimes called dotplots, but they are very different. Dotcharts are bivariate plots used instead of bargraphs for displaying frequencies of nominal variables. Do not confuse dotcharts with dot histograms.
Test yourself
With
R the first piece of code (below) gives the dotplot shown above.
Dotcharts are bivariate plots: one variable is numeric (in this case percentages), the other variable is non-numeric (nominal, or categorical). Whilst these categories COULD be ranked according to various criteria, their order here is wholly arbitrary.
Useful references
- Chambers, J.M. et al. (1983). Graphical methods for data analysis. Wadsworth International Group/Duxbury Press, Belmont & Boston.
- Describes one dimensional scatter plots (= rugplots), one dimensional scatter plots with stacking (= dot histograms) and jittered dot plots (equivalent to rank scatterplots) in Chapter 2. Recommended!
- Jacoby, W.G. (2006). The dot plot: A graphical display for labeled quantitative values. The Political Methodologist 14(1): 6-14. Full text
- Describes the use of dotcharts (Cleveland dot plots)
- Kabacoff, R.I. (2012). Quick-R: Dot Plots.
Full text
- Only covers dot charts, also known as Cleveland dot plots.
- Wilkinson, L. (1999). Dot plots. The American Statistician 53 (3), 276-281. Abstract
- Points out that 'Wilkinson dot plots' have been used for more than 100 years to show distributions in detail. Recommended!
- Wikipedia: Dot Plot (statistics).
Full text
- Uses the term Wilkinson dot plots for dot histograms, and Cleveland dot plots for dot charts. Rug plots are not mentioned, although one is used in the section on kernel density estimation.
See Also