Test Approach Estimation Flashcards
Which types of estimation approaches exists?
- Expert-based approach
This approach is to use the experience of the owners of the relevant tasks or experts to derive an estimate. - Metric-based approach
This approach relies on data collected from previous or similar projects.
This approach makes it possible to estimate cost and time required for the project, quite accurately.
What type of data can you use to estimate in a metric-based test estimation approach?
This kind of data might include:
- The number of test conditions.
- The number of test cases written.
- The number of test cases executed.
- The time taken to develop test cases.
- The time taken to run test cases.
- The number of defects found.
- The number of environment outages and how long on average each one lasted.
How can you estimate a whole project with a team of experts or expert area knowledge?
You can either let the task owners individually estimate their own expertise area based on the requirements specification or distribute the specification to experts and then meet and agree on an estimate together.
In agile testing you have deadlines to follow. These needs constant reviewing, this is called test monitoring. So what is the purpose of test monitoring?
It is to recieve constant feedback for timescales to see if your team is on schedule.
What are some commong test metrics?
- How many tests are writting out of the amount you decided on..
- Number of test cases run/not run, and test cases passed/failed
- Defect density, defects found and fixed, failure rate and retest results
- Dates of test milestones.
- Testing costs (cost vs. benefits)
The test metrics that are used, what should they be based / relate to?
The exit criteria. If the exit criteria is to cover the 60 % most important functionalities or have full coverage on all functionality deemed critical, how many of those tests are written / passed is a criteria.
Furthermore the exit criteria is used to determine if a test has been completed or should stop. e.g cost / benefit is not worth it (cost of possible errors in program is lower than finding them before launch now)
What is test control?
Test control are actions that you do when testing is behind - you control.
- You have to make these decisions based on test monitoring information.
What are some Test Control actions?
- Cut down on features
- Delay launch / adjust scope
- Review product risks and make a decision on if something can be left out for first release etc.
What is an incident?
When unforeseen events that impact the project negatively occurs. It can be defects, specification changes, test design problems or misunderstandings.
What is incident managment?
The process of recognizing and investigating as well as taking action on incidents.
How is incident management managed?
There should be appointed an incident manager. Incidents should result in a incident report that has the objectives of providing feedback for developers, ideas for improvements of the test process and quality tracking information for test leaders, like number of incidents, priority and correction status.
What is a risk?
A factor that potentially result in negative consequences for a project, usually expressed as an impact and likelyhood.
- if there were no risk there would be no need for testing.
What is level / risk factor?
impact * probability of a test occuring.
There are two types of risks what are they?
Product and project risks.
Project risks could be Failure of a third party to deliver on time or at all or organisational factors like personal issues, staff shortage etc.
Project risks could be if the software is failure-prone.
So there are risks and they need to be managed, you could use a risk matrix to help you gain an overview over the risks what is that?
It is a matrix that contains all the information about the identified risks, their risk factor and a way to deal or prepare / minimize for a risk.
It helps visualizing the risks and prioritize them.
