China
Africa
Explore chapters and articles related to this topic
Exploratory Data Analysis
Published in Daryl S. Paulson, Applied Statistical Designs for the Researcher, 2003
An easy way to make a boxplot display by hand is to begin with a skeletal boxplot. A skeletal boxplot illustrates the “five-number summary;” i.e., it shows only the maximum and minimum data values (the extremes), the hinges,* and the median. It is constructed by drawing a box between the hinges (4.733 and 8.198) and depicting the median (6.405) as a solid line through the box (Fig. 4). To finish the skeletal boxplot, draw dashed lines from the hinges to the extremes (2.051 and 11.701). These dashed lines are called “whiskers” (Fig. 5). (Boxplots are also known as “box-and-whiskers” plots.)
A generalized BLUE approach for combining location and scale information in a meta-analysis
Published in Journal of Applied Statistics, 2022
Xin Yang, Alan D. Hutson, Dongliang Wang
In this section, we briefly review estimators of the sample mean and standard deviation based on sample quantiles derived summaries for a single study, with an emphasis on the existing optimal ones under different scenario that are not simulation-based. A more detailed literature review on all existing estimators can be found in Wan et al. [23], Luo et al. [17], and Weir et al. [24]. To be consistent, we use the same notations as those in Hozo et al. [14], Wan et al. [23], and Luo et al. [17]. Let n be the sample size of the study and denote the five number summary as a is the minimum value, m is the median, b is the maximum value. In clinical studies, the five number summary may not be fully reported. Three most common scenarios in the literature are
Obsessional slowness in obsessive-compulsive disorder: identifying characteristics and comorbidities in a clinical sample
Published in International Journal of Psychiatry in Clinical Practice, 2023
Erin Crowe, Maria C. Rosário, Ygor A. Ferrão, Lucy Albertella, Euripedes C. Miguel, Leonardo F. Fontenelle
Removal of missing data and “past” responses on the OS variable resulted in a final sample of 667 participants for statistical analysis. Histograms and statistically significant Shapiro–Wilk values indicated that the assumption of normality was violated for each of our continuous predictors, therefore Mann–Whitney U-tests were performed and z-values (corrected for ties) were reported. Five-number summary statistics were presented instead of means and standard deviations. No outliers were detected upon inspection of studentized residuals and expected cell frequency assumptions were not violated.