You are here: Vanderbilt Biostatistics Wiki>Main Web>DynamitePlots (revision 8)EditAttach

Dynamite Plots

  • The height of a bar indicates the mean, and the vertical line on top of it represents the standard deviation or standard error.
  • When sample size is small (e.g., n < 15), a dynamite plot should be replaced by a dot plot in which every data point is represented.
  • When sample size is large, a box plot should be used.
  • Dynamite plots have a high ink:information ratio and cause an optical illusion in which the reader adds some of the error bar to the height of the main bar when trying to judge the heights of the main bars.
  • Dynamite plots hide the raw data and typically only show one-sided confidence intervals. They usually assume the confidence interval is symmetric.

An example:


Means and standard deviations (or standard errors?) of Groups A and B look identical. But what does that mean? Group C and D look identical, too.
The same data are shown in a dot plot below.

An article with nice dot plots

(appeared in nature medicine):
Potti, A. et al. "Genomic signatures to guide the use of chemotherapeutics", Nature Medicine (2006).

Departmental Policy

At a department meeting on 18Oct06, statisticians in the department voted in favor of the policy regarding dynamite plots:

Dynamite plots often hide important information. This is particularly true of small or skewed data sets. Researchers are highly discouraged from using them, and department members have the option to decline participation in papers in which the lead author requires the use of these plots.

R code

>> dot plot
Edit | Attach | Print version | History: r11 | r9 < r8 < r7 < r6 | Backlinks | View wiki text | Edit WikiText | More topic actions...
Topic revision: r8 - 17 Dec 2007, TatsukiKoyama
 

This site is powered by FoswikiCopyright © 2013-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback