ISDS 2000 Chapter : Stat Ch 3
Document Summary
There are 3 major properties of numeric data: Variation, i. e. dispersion or spread (range and standard deviation. Mean- average; value around which observations tend to cluster; balance point of a histogram: population mean ( , sample mean, the mean is susceptible to outliers. Median- the point at which half the data lie above and half the data lie below the median is not susceptible to outliers, as calculations do not utilize all data values. Mode- the data value that occurs most often: can have more than one, the mode is susceptible to outliers. There are 2 measures of variation and shape. Range- largest value minus smallest value: range is susceptible to outliers. Variance- measure of variability that utilizes all data values. Standard deviation- square root of variance: sample standard deviation: s, population standard deviation: , both influenced by outliers, never negative. Measures of shape, z-scores, empirical rule, and determining outliers.