## About this lesson

The normal distribution charts the type of variability in a process parameter that is being measured when the only cause for variation is natural random physical effects. It's the desired distribution when improving a process since it delivers a predictable level of process performance.

## Exercise files

Download this lessonâ€™s related exercise files.

Normal Distribution.docx203.4 KB Normal Distribution - Solution.docx

203.7 KB

## Quick reference

### Normal Distribution

The normal distribution charts the type of variability in a process parameter that is being measured when the only cause for variation is natural random physical effects. It's the desired distribution when improving a process since it delivers a predictable level of process performance.

### When to use

The normal distribution is used in the analysis of data during the Analyze phase. It will also be used to verify that the only variation in the solution developed in the Improve phase is random variation. Furthermore, it will be used to track process performance during the control phase as part of statistical process control.

### Instructions

The normal distribution, or Gaussian distribution, is the best representative of random variation in a process input or output. This distribution is often referred to as the bell-shaped curve. Its characteristics are that the distribution is symmetric, with a peaked center and the upper and lower tails approaching zero.

The curve can be described with a standard deviation scale. Setting the “0 value” for the scale at the center of the curve, the standard deviations can be used to show the percentage of the data values within the span of different standard deviations.

The statistical techniques associated with statistical process control and hypothesis testing will rely heavily on the use of normal curves. Both of those topics are covered in more depth in other GoSkills programs.

### Hints & tips

- Your data will probably not be a perfect normal curve. However, the hypothesis testing course will show how to determine if a normal curve is good approximation.

Lesson notes are only available for subscribers.