#Testing There are 8 marks available for this section.
##Introduction In this section you must test your system and document the evidence of that testing. To do this you will need to:
- Provide your test plan
- Provide your test data
- Provide annotated samples of tests runs for normal, erroneous and extreme data that are cross referenced with the test plan
- Provide an evaluation of your testing phase
The following sections will help you to produce a framework for your Testing.
##Test Plan You produced an outline and detailed testing plan for your system during the design phase of the coursework. Both of these should be used as a basis for the test plan section here.
Refer to pages 261-262 of Bond A2 for examples of testing plans.
###Individual tests Ensure that your individual tests have appropriate names and descriptions - they should be unique and reasonably descriptive. At a minimum you should have between 5-7 series of tests and for each series at least 4-6 individual tests.
###Highlighting changes During the implementation phase you may have changed some aspects of your design therefore it may be necessary to modify your testing plan. If so, make the changes but be sure to explain and justify any variations from the original plan.
You should clearly justify any additions or removals from your plan and ensure that they are easily identifiable from the original plan. Keep in mind that colour coding tends not to look great when the document is printed in black and white at the end of the course.
##Test Data You produced example normal, erroneous and extreme test data for your system during the design phase of the coursework. Your test data formed part of your detailed test plan.
Again, you may find you need to modify your test data to fit any changes made at the implementation stage - this is fine but make sure you explain and justify these changes.
For each data set you should also describe the type of data set it is and explain why it is a good choice for the given test.
###Test data types You should balance your test data so that you get a good coverage of normal, erroneous and extreme tests. If you feel that one type is not appropriate then you should include a detailed justification of your reasoning.
##Annotated Samples You should provide annotated screenshots showing your testing in action.When provided screenshots you should always show the system before the test, after the test and at any relevant stages in-between. In addition, ensure that there are clear headings for each screenshot.
Use text-boxes and arrows to highlight important features of the test and be sure to cross-reference the test with your testing plan. For example, you should have an evidence column in your testing plan table which you can use to reference evidence headings and page numbers e.g. Figures 2a-c, pg 45-46.
###Sampling. Choose 2-3 tests of each type of test to provide evidence for - you do not need to evidence every single test.
###Referencing Cross-referencing in Latex is easy - make sure you check the examples.
##Evaluation Explain how you approached the testing, what major problems you discovered and what you did about them. Do a full evaluation of your testing and say what its strengths and weaknesses are.
Explain how your testing proves that your system is reliable and robust.
###Approach to testing Here you need to summarise what you did to test the program. Make sure that you explain and justify the techniques used.
###Problems with testing In this section you should highlight those tests that have failed, referring to any necessary evidence.
###Strengths and Weaknesses of Approach Be sure to focus on the strengths and weaknesses of your approach to testing not of the functionality of the system.
###Reliability and Robustness You need to explain what is meant by each term and indicate, with evidence, whether the system is reliable and robust.
##Mark Scheme
| 0-2 marks | 3-4 marks | 5-6 marks | 7-8 marks |
|---|---|---|---|
| No test plan | Test plan with at least typical and erroneous data | A nearly complete test plan showing expected results | A well designed test plan showing expected results |
| Limited evidence of test output | supported by annotated test output | supported by selected samples of carefully annotated and cross-referenced test output | supported by selected samples of carefully annotated and cross-referenced hard copy showing test runs that prove the reliability and robustness of the candidate’s system |
| Only tested using typical data | Incomplete testing, e.g. omitting a boundary test | All significant aspects thoroughly tested using typical, boundary and erroneous data |