Correlation test (software)
In software technology, a correlation test refers to a process in which not only individual parameters of a function are checked for plausibility (for example in terms of data type or value range), but also combinations of these parameters are taken into account.
It is possible that each parameter has a valid value for itself (for example in data type and value range ), but these in combination cause the function to be tested to behave incorrectly, namely when these parameters are correlated by the function .
- example
A rectangular object is to be displayed on the screen. There is a function for this which accepts the dimensions of the rectangle in the parameters X , Y , SX , SY .
- Parameter X specifies the X position of the upper left corner. It must be checked whether X is in the valid display range.
- Parameter SX indicates the X edge length (width of the rectangle). Here it must first be checked whether SX does not exceed the permitted display width.
- In a correlation test, it is now additionally checked whether X + SX is in the valid range of values.