abul

 Profile Answers by abul 

• Jan 5th, 2007

 

Boundary value analysis : A process for selecting test cases/data by identifying theboundaries that separate valid and invalid conditions. Tests are constructed totest the inside and outside edges of these boundaries, in addition to theactual boundary points. Experience suggests that these tests have a higherpayoff than valid and invalid values chosen at random.

Boundary value testing: Testing the program’s response to all extreme inputvalues. Testing above the upper limits and below the lower limits.

Equivalence partitioning :A test technique that utilizes a subset of data that isrepresentative of a larger class. This is done in place of undertakingexhaustive testing of each value of the larger class of data. For example, abusiness rule that indicates that a program should edit salaries within a givenrange ($10,000 - $15,000) might have 3 equivalence classes to test: - Less than$10,000 (invalid) - Between $10,000 and $15,000 (valid) - Greater than $15,000(invalid)
Boundary value analysis

A process for selecting test cases/data by identifying theboundaries that separate valid and invalid conditions. Tests are constructed totest the inside and outside edges of these boundaries, in addition to theactual boundary points. Experience suggests that these tests have a higherpayoff than valid and invalid values chosen at random.

Boundary value testing:Testing the program’s response to all extreme inputvalues. Testing above the upper limits and below the lower limits.

Equivalence partitioning :A test technique that utilizes a subset of data that isrepresentative of a larger class. This is done in place of undertakingexhaustive testing of each value of the larger class of data. For example, abusiness rule that indicates that a program should edit salaries within a givenrange ($10,000 - $15,000) might have 3 equivalence classes to test: - Less than$10,000 (invalid) - Between $10,000 and $15,000 (valid) - Greater than $15,000(invalid)





A group of tests forms an equivalence class if they all testthe same thing, if they will all likely find the same bugs and if one test willnot find a bug it is likely none of them will. Also the tests are in the sameequivalence class if the tests involve the same input variables, they result insimilar operations in the program, they affect the same output variables andnon of them force the program to do error handling or all of them do.

Equivalence testing : If you expect the same result from two different tests theyare considered to be equivalent. An example of an equivalence test would be atest that the expected results are Y or N. The Y and y are one equivalencetest and the N, n and all other results that are not N or n are equivalent to Nor n test.

Error guessing: Test data selection technique for picking values that seem likely tocause defects. This technique is based upon the theory that test cases and testdata can be developed based on the intuition and experience of the tester.

Some people have a natural intuition for test data or test casegeneration. While this ability cannot be completely described nor formalized,people's ability to create test data that has a high probability to catcherrors should not be overlooked. Guessing carries no guarantee for success, butneither does it carry any penalty

State transition : Testing to see if the program can switch correctly from one stateanother. Tests sequence order requirements and program interruption errors.

Cause Effect graph :A tool used to derive test cases from specifications. A graph whichrelates causes (or input conditions) to effects is generated. The informationin the graph is converted into a decision table where the columns are thecause-effect combinations. Unique rows represent test cases.

Race Conditions : Errors caused by events happening in the wrong order. If A wasexpected before B, but B was executed first, a race condition error would beproduced. Many “irreproducible” bugs are cause by this condition.