Quality Window: Relate Screen
The Relate Screen is the third of four main charting screens in Quality Window. It is made up of three elements: the Control Chart, Scatter Plot, and Statistics Box.
This screen has a very unique capability. Once the variable in the top window is selected Quality Window scans all remaining variables in that application and performs an auto-correlation. This means that the variables displayed in the second window are ranked by the absolute correlation values determined by QW Viewer. A correlation value is a statistical determination of the strength of a relationship between two data populations. If a correlation value is highlighted in the Relate View it indicates that there strong statistical evidence that two variables are related in some manner.
Control Chart
The Relate Screen features two Control charts, each charting a separate data variable over a color coded overlay of your specification and control limits. Average is automatically calculated, and limit calculations and x-axis labels are easily customized from their defaults. The Control Chart comes with 3 different standard deviation formulas, Population, Lowest and Moving Range.
With one glance these two Control Charts allow you to visually and statistically determine if there is a relationship between the two variables.
Scatter Plot
The Scatter Plot chart graphically shows the relationship between the two variables by taking the values of each data point pair and plotting them on an x y chart. Positive relationships are shown by a diagonal grouping moving from bottom left to top right, whereas negative relationships create a diagonal grouping moving from top left to bottom right. Weak relationships create a true "scatter" effect, with no discernible pattern. The limits for each variable are also displayed with colour coded bars.
Statistics Box
The Relate Screen includes a Statistics box that shows selected standard statistical calculations, and displays the results for each variable side by side for easy comparison. The number of displayed statistics can be set by the users. Additional information is available on the Statistics page.
|
