The online community for software testing & quality assurance professionals
 
 
Calendar   Today's Topics
Sponsors:




Lost Password?

Home
BetaSoft
Blogs
Jobs
Training
News
Links
Downloads



Quality Engineering >> Estimation and Planning

Pages: 1
karty
Member


Reged: 07/07/05
Posts: 33
Test Case Point Estimation
      #682950 - 08/23/11 09:34 PM

Hi

My organisation does not have any test estimation template. So I am preparing a Test Case point estimation template for both manual testing.

Can anybody help me on the below questions


1) In Estimate for Total Test Case Points, 2, 4 and 8 has been mentioned as Simple, Average and Complex respectively. How these numbers are arrived? Is this a predefined numbers or will it change according to project

2) What are the independent Adjustment Factors and what values should it contain?

3) To determine the adjustment factors, Factors are assigned with a Relevance weight ranging from 0 to 10 and Skillset weight of 0 to 3. Are there any specific guidelines in allocating the weight?

4)In ESTIMATE FOR TEST CASE GENERATION, the adjustment = 1 + total adjustment factor. For eg if 0.30 is the adjustment factor, then the
adjustment for Test case genaration is 1+0.30 which is 1.30. Why should 1 be added ? How it is arrived? Is this a constant value or does it change for projects.

5)How the Person Hours Per TCP for test case generation and test execution is calculated?

6) once we arrive at the effort for test case generation and test execution, how to allocate effort for the following phases, Requirements, Test Planning, Test Closure, UAT support and Mangement Activities?


Post Extras: Print Post   Remind Me!   Notify Moderator  
dlai
Junior Member


Reged: 05/02/06
Posts: 1041
Loc: CA, USA
Re: Test Case Point Estimation [Re: karty]
      #683071 - 08/24/11 10:06 AM

Test case estimating before the test is written is very hard, same applies with programming as you don't know what challenges you'll hit. Before you dive in, you sort of might know what challenges you'll hit, but not necessarily the hours it takes to resolve them.

I think you can try using a planning poker approach used in software estimation. You give a number (usually a fibonacci number like 1,2,3,5,8,13,...) and rank things by orders of magnitude. When giving these number scores, don't think about hours, think about complexity relative to other tests. So to tests rated with an 8 are equally complex, but not necessarily take 8 hours.

As your team do the tests, you find out ball park numbers of what the points might equate to.

--------------------
David Lai
Sr. QA / Test Lead
LinkedIn profile


Post Extras: Print Post   Remind Me!   Notify Moderator  
karty
Member


Reged: 07/07/05
Posts: 33
Re: Test Case Point Estimation [Re: dlai]
      #683387 - 08/28/11 09:40 PM

Thanks David...

But currently we have identified TCP as the estimation technique. Can anybody help me on those questions.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Avishek
Junior Member


Reged: 12/30/03
Posts: 2
Loc: Noida
Re: Test Case Point Estimation [Re: karty]
      #683438 - 08/29/11 06:24 AM

My experience tell me...

1) In Estimate for Total Test Case Points, 2, 4 and 8 has been mentioned as Simple, Average and Complex respectively. How these numbers are arrived? Is this a predefined numbers or will it change according to project

.......2, 4 and 8 mostly refer to the steps in a test case. The more complex the test case the more steps it would include. However 2 steps for simple and 4 for average look low to me.
These numbers are actually deduced statistically through observation.

2) What are the independent Adjustment Factors and what values should it contain?

3) To determine the adjustment factors, Factors are assigned with a Relevance weight ranging from 0 to 10 and Skillset weight of 0 to 3. Are there any specific guidelines in allocating the weight?

4)In ESTIMATE FOR TEST CASE GENERATION, the adjustment = 1 + total adjustment factor. For eg if 0.30 is the adjustment factor, then the
adjustment for Test case genaration is 1+0.30 which is 1.30. Why should 1 be added ? How it is arrived? Is this a constant value or does it change for projects.

.... The answer for the above three points is, when estimating we normalize all test cases to simple test cases. The reason for that is the test cases productivity ( test case perday)is calculated based on simple test cases 9 most of the time). In certain cases when the number of test steps in a test case are large and while normalizing the number of simple test case comes to 1.6, then 0.4 adjustment factor is added to make it two simple test cases.

For the last question there is no thumb rule. It is expressed in percentagages. Normally the productivity of test case creation and execution covers everything from requirements management to execution to reporting. So if your productivity figure for test case creation and execution is based on good data and analysis then you dont need to separate out requirements and other efforts. For resource loading you can take 10 %age of total estimation effort as requirements management, planning etc....
Hope I am able to clarify and give some insight, its a vast topic to be explained in detail in this forum


Post Extras: Print Post   Remind Me!   Notify Moderator  
Pages: 1



Extra information
0 registered and 8 anonymous users are browsing this forum.

Moderator:  swt88, AJ, corklad, blueinatl 

Print Topic

Forum Permissions
      You cannot start new topics
      You cannot reply to topics
      HTML is disabled
      UBBCode is enabled

Rating:
Topic views: 15931

Rate this topic

Jump to

Contact Us | Privacy statement SQAForums

Powered by UBB.threads™ 6.5.5