What is it?
In experimental research, one or more variables are manipulated to see if this produces a change in one or more variables. For example, while testing a communication one could show to two groups of consumers an identical message in which only a piece of the information differs between the two, and then can assess if this influences consumers risk/benefit perception. An example would be testing the impact of different numerical formats (e.g., percentage vs. frequency) for presenting risks/benefits on consumers attitudes towards a specific food (e.g., red meat).
The first type of variables (e.g., different numerical formats) is called independent variables, while the second is called dependent variables (e.g., consumers attitudes). If other uncontrolled or unexpected factors (e.g., consumers' mood) do not interfere with this, it is assumed that the change in the independent variables has led to the change observed in the dependent variables. In other words, a causal relationship between the two variables is inferred. Thus, experimental research is very useful to determine cause-effect relationships. When using this type of research it is important to know the causal hypotheses, the independent and dependent variables, and the strengths and weaknesses of experimental designs.
When to use it?
Experimental research is particularly suited:
- When the main research goal is to identify cause and effect relationships.
- When the researcher aims to achieve control over environmental or extraneous factors (e.g., gender) that might interfere with the effect of the independent variables over the dependent variables.
- When the independent variables can be manipulated.
- When cases or participants can be randomly assigned to different experimental conditions.
- To test the impact of a program, intervention or communication in changing consumers' attitudes, perceptions and behaviour.
The use of experimental research provides a number of advantages relative to other types of research, namely:
- The confidence with which a causal relationship can be inferred. This derives from the greater degree of control that can be exercised in experimental research.
- The replication of the effects and their causes is facilitated given the standardization of procedures.
- Possibility of creating or simulating situations that might rarely occur. This can be achieved by planning and specifying the exact conditions under which the experiment is going to occur. This ensures that the participants’ performance is primarily influenced by the independent variables introduced and manipulated by the experimenter. By controlling variables the researcher has the opportunity to create a context that may be difficult to find in natural situations.
Disadvantages and limitations
Experimental research has some disadvantages and limitations relative to other types of research, namely:
- Lower generalization of findings collected under the created or simulated conditions to their naturally occurring settings. The conditions created and the high level of control may make the situation artificial and unnatural, which might difficult their prediction under real-life situations. This can be reduced by producing conditions closest to the natural environment.
- Choosing the appropriate experiment design can be difficult, thus demanding an experienced experimenter for this.
- Important independent variables might be difficult to manipulate and complex behaviours might be difficult to measure (e.g., interpersonal communication).
- Confound variables can be difficult to identify and control. These are variables that tend to change in co-occurrence with changes in the independent variable, and it might thus be difficult to assess if the effects occurred due to the independent variables or to these confounding variables.