 
Junior Member
Typical regression percentage What is a typical % of regression defects? i.e. what is typically considered acceptable? Are there industry standards for this or does this really depend on the complexity of the project etc.
Thanks! 
Moderator
Re: Typical regression percentage It's usually dependent on the project and its complexity, with every project being different there may be a guideline somewhere (that I have never heard of) but no standard.  M
Nothing learns better than experience.
"So as I struggle with this issue I am confronted with the reality that noting is perfect."
 Unknown
Now wasting blog space at QAForums Blogs  The Lookout 
New Member
Re: Typical regression percentage There is no stadards for this Regression.
Usually its purely depen on the complexity of the functionality of the application. The % you identify for the sanity is usually equal to your Regression %. As per my knowledge the ideal % is 40  48 %. 
Moderator
Re: Typical regression percentage [ QUOTE ]
As per my knowledge the ideal % is 40  48 %.
[/ QUOTE ]
???
Perhaps I don't understand the definition of "regression defects" used here.
I think of "regression defects" as old defects found in previouslyfixed areas of code. In that case, the ideal is clearly 0%.
In my shop we very, very rarely find cases where previouslyfixed defects reappear. Not 0%, but very close to it. 
Moderator
Re: Typical regression percentage [ QUOTE ]
The % you identify for the sanity is usually equal to your Regression %. As per my knowledge the ideal % is 40  48 %.
[/ QUOTE ]
If by Regression you are referring to bugs on existing code in a new release I have never seen a standard, or much on this, sanity or otherwise related. I also don't see how you come up with these numbers or why you sanity would be equal. Can you give some background on how you come with this?  M
Nothing learns better than experience.
"So as I struggle with this issue I am confronted with the reality that noting is perfect."
 Unknown
Now wasting blog space at QAForums Blogs  The Lookout 
Member
Re: Typical regression percentage By Regression percentage, I am gathering you want to know the number of defects you are expecting to see by running a regression cycle.
Two questions 
1. What was the need for a regression cycle? Was there a need? Could this decision be incorrect?
2. Which test cases were picked up for the regression cycle? All test cases, failed test cases or selection based on analysis?
These two questions may be able to assist you in determining the effective of finding defects in regression. (In my opinion, regression goal is not find defects, instead it is to ascertain that existing, working functionalities are not impacted due to code change. Code change may be due to defect fixing or new feature)
Posting Permissions
 You may not post new threads
 You may not post replies
 You may not post attachments
 You may not edit your posts

Forum Rules 