What is a semantic test?
The semantic test, which is also referred to by the acronym SEM, is a test design technique. You use such a technique to design a software test. As a software tester, you have a choice between all kinds of techniques. The design, method and purpose differs for each technique, so it is important to know what they entail. On the agenda: the semantic test (SEM). At the end of this blog you will know what it means, how and when to apply it and what the purpose of the semantic test is!
Techniques for testing software
As a software tester, you can choose from a wide range of test design techniques. There are eleven in total. They have been around for a long time, but are still actively used by software testers today.
A test design technique is a standard way of testing software. The standardized method allows you to test software consistently. We deliberately say consistent, because with a fixed method you can test all software the same. Because you have a standard operating procedure, you can test all types of software the same way. That leads to consistent results. An example of such a test design technique is the semantic test.
What do you test with a semantic test?
With a semantic test you investigate the functionality of input and output data and business rules. For input data, think for example of entering your date of birth on an alcohol sales site. The data you enter is sent to a database. With the semantic test you can examine whether the software is okay. If the software is in order, you will not be able to access the website if your date of birth is less than 18 years before the current date. If an underage visitor does get access to the site, the semantic test, if done properly, will reveal this. Business-rules are restrictions on elements within a database. The fact that an entry field for your year of birth requires you to enter exactly four digits is an example of a business-rule.
The purpose of this test design technique
The purpose of a semantic test is to test the relationship between data. To use the example just given, the semantic test examines the relationship between the date of birth and today. If that should be 18 years, the semantic test tests whether this works well in practice.
The implementation of a semantic test
A semantic test consists of four steps. As a software tester, you want to perform the test as carefully as possible, especially if you specialize in the actual and accurate TMap method. For that reason, the four steps are shown below.
Step 1: Identification
At the first step, you identify the test situations. Often an ALS rule is created at the first step.
Step 2: Logical setup
The second step consists of making the identified rule understandable. You fill in clear variables.
Step 3: Physical setup
Now it's time to turn the 'logical setup' into concrete test cases. If you have come up with '<18' as a comprehensible variable, you can choose '17' as a concrete test case.
Step 4: Establishing the initial situation
To run the test, you need an initial situation. The initial situation from which to test can be the same when running multiple tests.
When do you use a semantic technique?
Chances are you've already deduced the answer to this question from the text above. Still, there is no harm in repeating it. You deploy a semantic test when you want to examine the relationship between input and output data and business rules for functionality. Software that makes full use of such data or that contains business-rules may be suitable to subject to a semantic test.
Do you have the ambition to become a software tester or do you want to know more about it? Then following an online course can help you develop into a professional. The elearning TMAP® Suite Test Engineer and TPI NEXT® Foundation are excellent options for this. Check out our website for offerings and all kinds of information about software testing!