Topic: Mutation testing

Message

Mutation testing

Mutation testing is verifying the testing procedure itself. It may also
be used to predict how many issues are yet to be found within the
system. To conduct Mutation testing one needs to purposely seed some
issues into the product (beware to remember each one). The amount of
found seeded defects indicates the level of testing coverage.

Verifying testing thoroughness

Suppose you have seeded 10 issues into different parts of system
and then provided it to the next testing stage. After completing
testing stage you have found 100 issues, 8 of which are from the seeded
sample. This means that defect removal efficiency of this stage is
about 80%.

DRE = DTF / DTS = 8 / 10 = 0.8

DRE - Defects removal efficiency

DS - Defects seeded

DSF - Seeded defects found

Determining the total amount of defect (including not found)

With example above, knowing that you have found 8 of 10 seeded
issues enables you to suppose that the total amount is proportional to
the amount of found issue at about the same ratio. Building a simple
proportion will give you a formula for determining the total amount of
defects:

DT = DTF * DS / DSF = 100 * 10 / 8 = 125

DT - Total defects (predicted)

This means you've got about 25 issues remaining not found.

Determining areas of weak coverage

Another possible application of this method is indicating the areas
when you would probably require more testing. Just look where, what
part of product, the not-found-seeded-issues are and add tests to those
areas.
_________________

Edited by moderator - 04May2007 at 11:10pm

Author	Message
varsha Newbie Joined: 02Apr2007 Online Status: Offline Posts: 1	Topic: Mutation testing Posted: 03Apr2007 at 12:01am
	Mutation testing Mutation testing is verifying the testing procedure itself. It may also be used to predict how many issues are yet to be found within the system. To conduct Mutation testing one needs to purposely seed some issues into the product (beware to remember each one). The amount of found seeded defects indicates the level of testing coverage. Verifying testing thoroughness Suppose you have seeded 10 issues into different parts of system and then provided it to the next testing stage. After completing testing stage you have found 100 issues, 8 of which are from the seeded sample. This means that defect removal efficiency of this stage is about 80%. DRE = DTF / DTS = 8 / 10 = 0.8 DRE - Defects removal efficiency DS - Defects seeded DSF - Seeded defects found Determining the total amount of defect (including not found) With example above, knowing that you have found 8 of 10 seeded issues enables you to suppose that the total amount is proportional to the amount of found issue at about the same ratio. Building a simple proportion will give you a formula for determining the total amount of defects: DT = DTF * DS / DSF = 100 * 10 / 8 = 125 DT - Total defects (predicted) This means you've got about 25 issues remaining not found. Determining areas of weak coverage Another possible application of this method is indicating the areas when you would probably require more testing. Just look where, what part of product, the not-found-seeded-issues are and add tests to those areas. _________________ Edited by moderator - 04May2007 at 11:10pm Post Resume: Click here to Upload your Resume & Apply for Jobs

	IP Logged

naruboinac Newbie Joined: 11May2007 Online Status: Offline Posts: 1	Posted: 11May2007 at 6:40pm
	What is an Automation Framework 2. While automating an application what are the reusable components u would consider 3. what is an object,method and class and property.Give sp examples 4. what are addins 5. why are addins used 6. what is a patch 7. why would u go for automation 8. write a script for comparing the name in report against the name u inputed 9. tell me a scenario for which u have prepared an autromation script 10. what estimations u would consider while automating an application 11. Also tell how did u approach in preparing script for scenario in q 9 12. what is obkject repository 13. extension for object repository 14. extenmsion for qtp script 15. how do u handle exceptions programmatically and thru qtp 16. what is difference between Test Object and RuntimeObject and GETTO property and GETRO property 17. how do u set a value for an object 18. what is synchronization and wait 19. how do u report result 20. what metrics u would consider so as to decide which are the manual test cases u can automate 21. what are the different types of licences 22. what is the lastest version u are usin 23. how do u report results in QTP 24. what is Object identification 25. Define mandatory,assisitiv e,base filter properties 26. what is smart identiifartion 27. what is Test Guard Technology 28. what is an identifier 29. what are actions,reuable and external actions 30. what are types of object repository 31. how would u use VB functions in QTP 32. what are rgular expressions 33. what is reporter .report 34. what re different checkpoints 35. can u wriote customized checkpoinst if so how 36 What is an Object spy 37. Did u use any test management tool to integerate QTPi with? 38. Can QTP support linux 39. What are broswers,applicatio n and OS QTP supports 40. What are minimum environment considerations fer QTP
	IP Logged


One Stop Testing Forum : Software Testing @ OneStopTesting : Test Plans @ OneStopTesting