Six Sigma is a process-oriented tool used in a variety of organizational settings. The goal of Six Sigma is to eliminate process flaws in order to increase efficiency. Six Sigma is rooted heavily in statistics. The two fundamental risks of statistical analysis are alpha risk and beta risk, both of which involve incorrect hypotheses.
The term Six Sigma refers to a statistical measurement. Sigma is a Greek letter used to represent a standard deviation, which is a statistical measurement of the distance of a data point from the mean. Six Sigma refers to the fact that the goal of a Six Sigma process is that the process specifications include everything withing six standard deviations from the mean. Statistically, this means that the goal of the process is to have no more than 3.4 defects out of one million opportunities.
Alpha risk is the risk that someone will incorrectly conclude that there is an error when there really is no error. For example, a quality inspector might decide that there is a defect in a manufactured product when there really is no defect. This might lead the inspector to believe that the entire process must be changed in order to meet Six Sigma requirements.
Beta risk is the opposite of alpha risk. Beta risk is the risk of concluding there is no error when there actually is an error. For example, a quality inspector could conclude that there is no defect in a product when, in fact, there is.
Alpha and beta risk can be illustrated in a decision matrix. Using a two-by-two matrix, the vertical axis asks whether a particular hypothesis is true, and the horizontal axis asks whether the hypothesis was assumed to be true. If the hypothesis was correct and assumed to be correct, or if the hypothesis was false and assumed to be false, there was no error. However, if the hypothesis was true and assumed to be false, an alpha error has been made; and if the hypothesis is false and assumed to be true, a beta error has been made.