Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Commentary

Commentary CONTRIBUTIONS Error Bars: Are They the King’s Clothes? Graphs are an integral part of most ecological publications and presumably should be integrated with the rest of the text. However, the use of error bars in Ecology, and other ecological journals, usually serves no purpose, or duplicates information that is given in the text. In this paper, I will show that scientific communication would be facilitated if editors discouraged authors from using error bars, and promoted the use of scatterplots (dispersion graphs) or their categorical equivalents. Ecological papers are usually some mix of data reporting, statistical or inferential data manipulation, and biological interpretation. I am advocating that graphs in empirical publications should be used primarily for data reporting. Tukey (1972) discussed many types of graphs that can be used for data evaluation. However, despite the continual “discovery” that inferential tests are overused (e.g., Nunnally 1960, Deming 1975, Guttman 1985, Yoccoz 1991, Johnson 1999), graphical evaluations have generally not replaced standard statistical tests. I reviewed a recent issue of Ecology (80[5], 1999). Error bars represented one standard error of the mean (1 SE) on 27 graphs, 2 SEs on 7 graphs, standard deviations (S) on 4 graphs, and 95% confidence http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Bulletin of the Ecological Society of America Ecological Society of America

Commentary

Bulletin of the Ecological Society of America , Volume 81 (2) – Apr 1, 2000
Download PDF
10 pages

Loading next page...
 
/lp/ecological-society-of-america/commentary-OK4V0g2yju

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Ecological Society of America
Copyright
Copyright © 2000 by the Ecological Society of America
Subject
Articles
ISSN
0012-9623
DOI
10.1890/0012-9623%282000%29081%5B147:C%5D2.0.CO%3B2
Publisher site
See Article on Publisher Site

Abstract

CONTRIBUTIONS Error Bars: Are They the King’s Clothes? Graphs are an integral part of most ecological publications and presumably should be integrated with the rest of the text. However, the use of error bars in Ecology, and other ecological journals, usually serves no purpose, or duplicates information that is given in the text. In this paper, I will show that scientific communication would be facilitated if editors discouraged authors from using error bars, and promoted the use of scatterplots (dispersion graphs) or their categorical equivalents. Ecological papers are usually some mix of data reporting, statistical or inferential data manipulation, and biological interpretation. I am advocating that graphs in empirical publications should be used primarily for data reporting. Tukey (1972) discussed many types of graphs that can be used for data evaluation. However, despite the continual “discovery” that inferential tests are overused (e.g., Nunnally 1960, Deming 1975, Guttman 1985, Yoccoz 1991, Johnson 1999), graphical evaluations have generally not replaced standard statistical tests. I reviewed a recent issue of Ecology (80[5], 1999). Error bars represented one standard error of the mean (1 SE) on 27 graphs, 2 SEs on 7 graphs, standard deviations (S) on 4 graphs, and 95% confidence

Journal

Bulletin of the Ecological Society of America – Ecological Society of America

Published: Apr 1, 2000

There are no references for this article.