These methods can be used to describe (or quantify) features of a quantity, process, or system. For example, one can quantitatively describe the shape of a distribution by its mean [mean, median, mode, skewness (the degree to which a distribution deviates from a symmetric bell-shaped distribution; positive numbers (e.g., 0.2) represent a left-tilted or peaked distribution, while negative numbers (e.g., -0.2) represent a right-tilted or peaked distribution), kurtosis (the measure of the degree to which a distribution is peaked or flat), and interquartile range]. One can use statistical distributions to make predictions about the behavior of quantities (e.g., the next occurrence of an outlier at a specified location), process (e.g., the time until the next lightning strike), and systems (e.g., the likelihood of catching a cold).
The information created by this process is critical and can be used for a variety of purposes: to roughly estimate the size of a population, to predict the likelihood of an event occurring, to select the appropriate sample size for an experiment, to describe the behavior of a product or service, to predict a system's behavior in response to input parameters, and to perform statistical hypothesis testing such as a t-test or chi-squared test, among others. To accomplish these goals it is necessary to understand the statistical process used to generate and analyze data. While statistical data analysis is a rigorous discipline, modern statistical programming tools and the widespread availability of open-source programs such as R , Python , and Octave have made statistical computation easier and much more accessible. d2c66b5586