Letters in Mean Comparisons: What They Do and Don’t Mean
- 1 March 2018
- journal article
- research article
- Published by Wiley in Agronomy Journal
- Vol. 110 (2) , 431-434
Abstract
Core Ideas: Letter displays allow efficient reporting of pairwise treatment comparisons. It is important to correctly convey the meaning of letters in captions to tables and graphs displaying treatment means. The meaning of a letter display can and should be stated in a single sentence without ambiguity. Letter displays are often used to report results of all pairwise comparisons among treatment means in comparative experiments. In captions to tables and charts using such letter displays, it is crucial to explain properly what the letters mean. In this paper I explain what the letters mean and how this meaning can be succinctly conveyed in a single sentence without ambiguity. This is contrasted to counter‐examples commonly found in publications.Keywords
This publication has 10 references indexed in Scilit:
- A tutorial on the statistical analysis of factorial experiments with qualitative and quantitative treatment factor levelsJournal of Agronomy and Crop Science, 2018
- If not multiple comparisons, then what?Annals of Applied Biology, 2017
- Multiple Comparison Procedures—Cutting the Gordian KnotAgronomy Journal, 2015
- Statistical Methods in Biology2014
- Algorithms for compact letter displays: Comparison and evaluationComputational Statistics & Data Analysis, 2007
- An Algorithm for a Letter-Based Representation of All-Pairwise ComparisonsJournal of Computational and Graphical Statistics, 2004
- Multiple Comparisons1996
- Statistics notes: Absence of evidence is not evidence of absenceBMJ, 1995
- The Philosophy of Multiple ComparisonsStatistical Science, 1991
- Biometry1982