Data Analysis Tools
Tools for analyzing and interpreting data so that areas to improve become apparent.
What type of data do I have?
Variables charts (measurement data)
(Learn more)Consists of measurements of a characteristic 
, such as length, weight, density, time, or pressure.
| Control charts | Is your process stable and in control? |
| X-bar & range | Use this if your data has a subgroup size of 2-10 observations. |
| X-bar & sigma | Use this if your data has a subgroup size of 11 or more observations. |
| X-MR | Use this if your data has a subgroup size of 1 observation. |
| Median | Use this to analyze measurement data when you want to plot all observations. |
| Run chart | Use this to see trends and patterns if there is not enough data for a control chart. |
| Histogram | Use this to determine if your data has a normal distribution . |
| Capability analysis | Use this to determine if your process is capable of producing output within specification limits. |
Attributes (counts data)
(Learn more)Consists of defects per item (nonconformities ) or the number of defective items (nonconforming ). For example, the number of non-working parts in sample or the number of blemishes counted on an individual part.
| Control charts | Is your process stable and in control? |
| np-chart | Use this if your data is a count of nonconforming units and the subgroups are all the same size. |
| p-chart | Use this if your data is a count of nonconformities and the subgroups are all the same size. |
| c-chart | Use this if your data is a count of nonconformities and the subgroups are all the same size. |
| u-chart | Use this if your data is a count of nonconformities and the subgroup size varies. |
| Capability analysis | Use this to determine capability for attributes data. |
Pareto (counts in categories)
(Learn more)Consists of a count of items or occurrences, such as the number of defective items, the number of scratches on a door panel, or how often a specific problem occurs.
| Pareto diagram | Use this to analyze counts that are in categories. |
Rare event
(Learn more)Use this when other control charts are not effective to determine if your process is stable.
| g-chart | Use this if your count data occurs infrequently. It is used by counting the number of events between rarely-occurring error or a nonconforming incident. |
| t-chart | Use this if your error or non-nonconforming incident occurs infrequently. Each point on the chart represents an amount of time that has passed since the prior nonconforming incident occurred. |
Interpreting quality charts
Control charts (both variables and attributes)
- Analyze for special cause variation
- Any point lying outside the control limits
- 7 or more points in a row above or below the center line
- 7 or more points in one direction
- Any nonrandom pattern
- Declare the system stable or unstable
- When do you recalculate control limits?
- What do the chart pairs mean?
(variables control charts only)
Step-by-step interpretation
Answer "yes" or "no" to a series of questions about your control charts.
Histogram
Follow these steps to interpret histograms.
- Study the shape.
- Calculate descriptive statistics.
- Compare the histogram to the normal distribution.
