information on Test Confidence ratio of Project
If you're making a (final) test summary report about the progress and quality of a project it's good to mention two things. The usual statistics about the progress and defects and a (qualitative) judgement about the quality of the project. The latter isn't possible to describe in statistics or figures. I think a general feeling about the quality of the project is as important as statistics, because with statistics one can proove everything and nothing.
If you really want to use some kind of test confidence ratio I think the best option is to use the defect detection percentage. This is used by taking the number of defects in the current test level and divide it by the number of defects in the current and all subsequent test levels.
For example: there are four test levels with a various number of defects
UT: 33 defects
ST: 75 defects
SIT: 98 defects
UAT: 18 defects
If one calculates it for the UT the formula is the following 33 / (33 + 75 + 98 +18) = 14,7%, for ST it's 75 / (75 + 98 + 18) = 39,2%, etc.
Take into account that if this is calculated for one project it says nothing. You need to have the history of several project to fully take advantages of this formula. This way the percentages of a test level can be compared with the average of the other projects and this way you have a fairly objective way of measuring the quality of a project.
The obvious disadvantage of this ratio is that it can only be calculated at the end of a project and not in the middle of a project.
If you really want to use some kind of test confidence ratio I think the best option is to use the defect detection percentage. This is used by taking the number of defects in the current test level and divide it by the number of defects in the current and all subsequent test levels.
For example: there are four test levels with a various number of defects
UT: 33 defects
ST: 75 defects
SIT: 98 defects
UAT: 18 defects
If one calculates it for the UT the formula is the following 33 / (33 + 75 + 98 +18) = 14,7%, for ST it's 75 / (75 + 98 + 18) = 39,2%, etc.
Take into account that if this is calculated for one project it says nothing. You need to have the history of several project to fully take advantages of this formula. This way the percentages of a test level can be compared with the average of the other projects and this way you have a fairly objective way of measuring the quality of a project.
The obvious disadvantage of this ratio is that it can only be calculated at the end of a project and not in the middle of a project.
Comments
Post a Comment