The biggest advantage of a true experiment is that it is the only experimental design that allows researchers to determine the cause and effect relationship between two factors. The factor that a researcher modifies is known as the independent variable and the effect he seeks to observe is called the dependent variable. For example, if a researcher studies the occurrence of cancer in people who smoke, existence of the disease is the dependent variable and smoking is the independent variable. Performing a true experiment helps the scientist study the nature of the relationship between the dependent and independent variable.
In a true experiment, the researcher uses two groups: an experimental group and a control group. The purpose of the control group is to account for extraneous factors other than the independent variable that may have an influence on the experimental results. In such experimental designs, researchers assign subjects randomly to either the experimental or control group. Continuing with the cancer research example, the scientist assigns both male and female subjects of different age groups randomly to either the experimental or control group. Randomization removes the influence of these extraneous factors such as sex and age on the occurrence of cancer and helps the researcher focus on the relationship between smoking and cancer.
Researchers collect data during the experiment and subject this to statistical analysis to determine the significance of their findings. Whether in the physical sciences or the social sciences, true experiment results are considered reliable because they lend themselves to such statistical study.
In a true experiment, a researcher manipulates only one independent variable at a time and tests its influence on the dependent variable. This helps him arrive at a simple yes or no answer. In the cancer example, the researcher can state his results in terms of whether smoking causes cancer or not. However, it is also possible to test more than one dependent variable at the same time by using higher level statistical methods of analysis. Thus, true experiments permit the researcher with flexibility in designing his study.