ABSTRACT
This chapter provides a review of data types and statistical methods relevant to performance indicator development and analysis. The two broad categories of statistics are inferential and descriptive. Inferential statistics are used to infer population parameters based on sample data, while descriptive statistics are used to organize and data in a convenient, useable form. Much of descriptive statistics deals with obtaining as much information as possible from small samples of the population data. Inferential statistics, like descriptive statistics, are used to infer qualities in a population based on random samples taken from the population. Linear regression, Bayes' theorem, and contingency tables are three inferential statistics methods useful in Performance Management. Correlation is useful in determining whether there is overlap in performance indicators—suggesting one of the indicators can be dropped without affecting decision making. Indicators are often "adjusted" or subject to a statistical process for reducing, removing, or clarifying the influences of confounding factors.
