Today we discuss software testing effectiveness measures - these are  P-Measure, E-Measure and F-measure. The E-measure is the expected number of failures detected by the set of test cases. The P-Measure is the probability of detecting at least one failure from the set of test cases. Both the P-measure and the E-measure have been used widely.   The F-measure is the number of tests required to detect the first failure.  Also faults are different from failures. A failure is any manifestation of deviation from the correctness of a program execution. One or more faults may be the reasons for the failures. Until a failure occurs, a fault is said to be not known.  A fault is typically found by debugging. A failure is seen by observation. The F-measure is therefore concerning the first fault. It leads to fewer test cases and a reduced cost of overall testing. Even within the same number of test cases, the order of test cases matters.
The detection of more failures does not always help us identify more faults. That said it is still better to reveal as many failures as possible. Intuitively the more failures the testing reveals the more information for debugging faults and the higher the chance of detecting more faults. In this sense, a measure can be to detect as many failures as possible. This is what we attempt with maximum value of E-measure.(courtesy T.Y. Chen and Y.T. Yu)
Failures occur for a subset of the possible input domain of the program. Generally, inputs that produce failures will cluster in regions of  the input domain. Contiguous regions form patterns in the input domain. For two dimensions, there are (a) block pattern, (b) strip pattern and (c) point pattern.
The assumptions made in finding patterns include the following:
(a) the failure region's size, shape and orientation is known
and (b) the region's location in the input domain is not known. (courtesy: cs.umd.edu CMSC737)
courtesy:
#codingexercise
int GetP-Measure(int p)
{
return 1 - (1-p) ^ (1/p);
}
int GetE-Measure(int n, int p)
{
return n*p;
}
int GetF-Measure(int sizeOfInputDomain, int sizeOfFailureRegion)
{
  return sizeOfInputDomain / sizeOfFailureRegion;
}