Testing and Test Data
It is important to test algorithms to check how they perform under a range of conditions.
This includes testing any validation you have created to ensure it performs as expected.
When creating a testing plan, the test data that you use shouldn’t be random values, but rather values that fulfil the following test criteria.
Normal data is test data that is typical (expected) and should be accepted by the system.
Extreme data is test data at the upper or lower limits of expectations that should be accepted by the system.
A pair of values at each end of a range:
- The data at the upper or lower limits of expectations that should be accepted
- The immediate values before or beyond the limits of expectations that should be rejected
Abnormal data (erroneous data)
Abnormal data is test data that falls outside of what is acceptable and should be rejected by the system.
Example of normal data, extreme data, boundary data and abnormal data:
A system has validation to ensure that only integers between 1 and 10 are entered as an input.
The test data for this could be: