Graph Testing

May 31, 2020

---

Graph is one of the most widely used structures for supposition. It is a well-defined structure which makes the testing or case studies easier. It is considered as an effective way to determine the systematic test selection of a system. This model is used to represent the desired information to be captured or collected. Next, the information to be shared is also presented and easily understood by the testers.

For graph based testing firstly, the tester is asked to collect the data for graph model and then cover all the elements for a particular graph. In this process of testing, the tester is first assigned by the responsibility of creating a graph followed by other steps. These are discussed next.

Steps for graph based testing include:

1. Build the graph model
2. Identify the test/major requirements
3. Select the path to cover those requirements
4. Select the data to be entered

As we know software application is made up of some objects. These objects are identified and graph is prepared.

It was earlier used for hardware testing but now it is introduced to the software testing. Various testing methods in the software testing includes: static vs. dynamic testing and the box approach includes white box testing, black box testing, specification based testing, visual testing and grey box testing. Cause and effect testing comes under the graph based testing that forms a part of black box testing. In this type of testing, one can choose the test case that relates to cause and effect.

Here, cause represents the input condition that gives information about the internal change in the system whereas effect refers to the output conditions or the transformation of the system.

This graph is helpful in certain circumstances:

• To determine the present problem so that decision can be taken fast.
• To correlate the factors affecting the system.
• To recognise the main cause of a problem.

Advantages of using cause effect graph are as follows:

• It highlights the area from which the data is taken and can be taken for additional studies.
• It helps in motivating the team.
• The data is arranged in such manner that even a non-technical person can also read it.
• It helps in detecting the reason of differences occurred in a process.
• It helps the team to decide the root reason of a problem.
• It is very appropriate for large systems.

Disadvantages of using cause effect graph are as follows:

• Difficult to design.
• It is difficult to choose the important input in limited time.
• There can be chances where path of drawing the graphs is not clear.
• There are chances of repetition of data already entered in the graph.

It comes under the category of black box testing in software testing. Mostly cause effect graphs are helpful in testing as they are easy to understand and displays the parameters of the advantages and disadvantages of a programme. This graph usually shows the nodes which represent causes on the left side and the nodes representing effects on the right side. The intermediate nodes present between the effect and cause combines input using the commonly used logical operators AND/ OR. Constrains can also be added to the causes and effect sides. These are represented by edges tagged with the constrain symbol using a dotted line.

In the today's world of development everything is turning into visual. No one wants to read the piles of boring text and then analyse or test the software. The new feature of graph based testing has made cause and effect graphs interesting. Due to this technique it has become possible for the non-technical person to understand the concept. This process is time saving as only the important texts are studied under the graph.