QA interview questions Must read

  1. What is a test case as opposed to a test script?  In a test case, the scenario is written for a manual test of the system, whereas a test script enables automated testing. However, sometimes these terms are used interchangeably. A test script with published expected results can be used for a test case as well.
  2. What is a benchmark?  This is a marked point in testing the system which identifies efficiency or productivity. The metrics reached at the benchmark determine if requirements have been met or if a particular metric in a new system exceeds the same benchmark determined in the legacy system.
  3. What is QA?  In IT, QA (quality assurance) is the method used to assure the client that a project has met or exceeded requirements. This is achieved by testing the new project, with quantified benchmarks. If a defect is found, it can be fixed before rollout or accepted by the client as a design tradeoff. QA should not only be done at the end of a project but throughout the SDLC to save time and money.
  4. Compare a walkthrough with an inspection.  In a walkthrough, the developers/engineers go through the details of how a project is going to be developed. It is an informal way of sharing information and getting input. In an inspection, there is a formal review of what has been done and whether it meets the requirements.
  5. Compare stress testing, volume testing and load testing.  All three of these are necessary when there is a large number of users expected to be accessing a system at one time. Volume testing check for how much data can be processed and saved. Stress testing is especially important when there are remote users. It determines how much usage a system will take before breaking down. In load testing, the intention is find out approximately how many users can access the system at one time, to be compared to the expected usage.
  6. What are design tradeoffs?  Sometimes, to meet requirements, it is discovered that not all requirements can be met. This may be a question of functionality, cost or execution. If while developing a system, it is discovered that the software is going to cause a slowdown of the system, or there is a bug in the application, the developers need to meet with the clients to determine if a requirement needs to be amended or even dropped. To make this determination, they must determine the severity of the problem, to see if it would actually make an impact on other requirements. Then the problems must be prioritized – does the client feel he can live with the problem? Is cost more important than speed? If a “bug” in a system is accepted by the client or QA, it becomes part of the package and gets termed a “defect.”
  7. Can Web applications be QA’d?  Absolutely. The developer can do unit testing locally (not online) with his/her browser. For Web applications, security is a foremost concern, as are links and database exchanges. There are a huge amount of tools on the market for this type of testing, such as WebQA, eValid and stressIT. These will test links, HTML, functionality and/or performance.
  8. How does one achieve QA?  Quality assurance (QA) should be an attempt to determine and report the quality of a product being created. In IT, this means, at a minimum, meeting the requirements agreed on. This is done by testing, both the software and hardware. In hardware, benchmark programs can be run which will test throughput, efficiency and productivity. In software, this is done by testing at several points during developing, from the original coding to the final application. In the end, it is possible that the project does not meet requirements, This can be a severe discrepancy or a minor one, and its priority then has to be determined by the client. At the end of a project, a ‘lessons learned’ session should be held to determine what can be done to improve QA in subsequent projects.
  9. 9.      What kinds of principles should a QA team have?  The team should endeavor to give feedback as quickly as possible, to save time; if they have an idea of why a defect occurs, they should suggest it to the developer. The team should prioritize their findings so that critical factors are addressed first; bells and whistles should not be as important or more important as functionality. The team should be monitoring and giving feedback all along the phases of the project. Since this work can become tedious, the team must always maintain the pride in their work and recognize its value to the company. The team must be tenacious; watching for problems in a project similar to past projects must not allow them to relax on their monitoring and testing. The team must not only keep an eye on the development process, but also the final product and its effect on users and clients. The team must plan for all phases of the SDLC – requirements, architecture, design, production and rollout.
  10. 10.  How can you apply metrics to a project’s QA?  There are many tools for testing a project’s development and execution, and they will generate numbers. But these numbers need to be analyzed to determine their value in each individual project. In the beginning, QA should determine that the requirements document is precise and complete. In the end the QA should determine the percentage of requirements met. The QA team should evaluate if the system is easy to understand, both from the viewpoint of users and admins. It should also be determined if the system is well modularized, so that parts may be applied to other projects. The team should check to see that documentation is not only present at every phase, but clearly read, for following developers. Of course, the efficiency of the project should be evaluated. On the ‘private’ end, the QA team should evaluate the project team to determine if predictions/proposals are met at the end. This should not only cover costs but time allotments, and recommendations of ways to improve the achievement of a project should be submitted.

Leave a Reply

Your email address will not be published.