madhu

• Jun 29th, 2006

 

Boundary value analysis is a technique where the input values at the boundaries of the input domain are tested.

There is a common assuption that the defects lie at the boundary values of the input domain.so test cases are written such that they cover the values that lie inside the input domain, max value of the input domain and value that is just outside the range.

for example if a system accepts a range of values from 0 to 100 then testing for -1 and 101 is boundary value analysys.

boundary value analysis complements the "Equivalence partitioning".

equivalence partitioning mainly aims on minimizing the test cases expecting that the system acts in same way for a range of inputs which are equally partitioned. 

veeravsk

 Profile Answers by veeravsk 

• Jul 25th, 2006

 

Boundry value Analysis: Boundry value of the given Input of the object. That is max and min.

ex:- If an test boc taking values 4-8. so the boundry conditions are

min=4 - pass

min-1=3 - fail

min+1=5 - pass

max=8 - pass

max+1=9 - fail

max-1=7 - pass

Equalience call partaning:- Methodically reducessing the huge set of possible test cases in much smaller. But equally efficent. That means it depends on the type of the input value.

Ex:- Above Example. That particular text box accept only numeric. it does not allow any other.

a-z - fail

0-9 =pass

Alpaha numeric, special charectors= fail

siri

• May 2nd, 2007

 

I understood the concept of boundary value analysis and example but not equivalance example, can you pls explain me in detail

B Mayur Kumar

• May 25th, 2007

 

We use both of these techniques, to make sure that tester has tested with all possible test inputs, I mean to say, Tester will have minimum of 3 test inputs if he has 3 Equivalence Classes, these three inputs may or may not have included with the valid and invalid boundary values, so it would required to apply boundary Value Analysis. For example: I will take the same months example.-2,-1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14

Valid Class is not underlined set (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)

Invalid Class is Underlined set (-2, -1, 0, 13, 14)

So it is enough to test at least one value from each call, instead of testing with all values in set or classes, because values in a set will have same behavior, that mean if a text box  for month field and which accepts only 1- 12, and if you enter 1 to 12 will have same response.(if requirement say the same)

The will test with only one value from the above sets.

for example 4, -2, 14

Now, I have applied the equivalence class partitioning successfully, but the input data is not the boundary value, so we need to apply the boundary value Analysis, which coveres boundary value to check.

The exact difference I think, both will generates different test data.