Hanging Dot Plots and identifying outliers in samples from asymmetrical populations with Hubert and Vandervieren's robust outlier rule. (26th December 2012)

• Record Type: Journal Article
• Authors: Bradstreet, Thomas E.; Gillespie, Amy; Zhang, Shuping; Zhu, Xingshu
• Is Part Of: Wiley interdisciplinary reviews. Volume 5:Number 1(2013)
• Page Start: 62
• Page End: 67
• Publication Date: 2012-12-26
• Subjects: hanging dot plot -- Hubert and Vandervieren's robust outlier rule -- SAS macro -- descriptive statistics -- confidence interval; Mathematical statistics -- Data processing -- Periodicals; Science -- Data processing -- Periodicals; Social sciences -- Data processing -- Periodicals; Mathematical statistics -- Periodicals; 519.50285
• DOI: 10.1002/wics.1241
• Languages: English
• ISSNs: 1939-5108

Abstract

Abstract: Visualizing individual data, empirical distributions, and summary statistics for small to moderate sized samples (≤ 500) is critical in early phases of drug research. We developed a customized SAS ® macro, %HangingDotPlot, which displays dot plots with parametric or distribution‐free descriptive statistics and confidence intervals simultaneously on the same page for multiple groups of data. Potential statistical outliers are identified with a fence rule adapted from an adjusted box plot for skewed distributions, which incorporates the medcouple, a robust measure of skewness. WIREs Comput Stat 2013, 5:62–67. doi: 10.1002/wics.1241 This article is categorized under: Statistical and Graphical Methods of Data Analysis > Statistical Graphics and Visualization