Describing the Observations
摘要
The outcome of every field experiment is a dataset consisting of a set of measures or scores taken on several experimental subjects about one or more properties (e.g., height, weight, concentration, sex, and color). The first task of data analysis is usually to describe such datasets by using the most appropriate statistics. This chapter considers continuous data, counts, and proportions. It introduces the main statistics of central tendency (mean and median), spread (range, interquartile range, deviance, variance, standard deviation, and coefficient of variation), and joint variation (Pearson’s correlation coefficient). The use of R for descriptive statistics is demonstrated, and four types of descriptive graphs (barplots, boxplots, histograms, and scatterplots) are introduced, along with the R coding to produce them.