General

In an experiment, the researcher changes a variable of the object under investigation in a controlled manner in an experimental set-up or research design. This change of a variable serves to obtain knowledge.

Experiments are divided into field experiments and laboratory experiments. In addition, there are so-called quasi-experiments, which I will not discuss here.

In the laboratory experiment the environment variables can be defined firmly.

The definition of the environment variables in field experiments is basically not problem-free because of the influence of the external unknown.

 

Laboratory Experiment

Field Experiment

  Laboratory Experiment Grüne Wiese mit blauem Himmel

Advantages

  • Controlled environment variables
  • Reduction of complexity
  • Can be controlled to a high degree
  • High internal validity
  • Realistic
  • Test persons often do not know that they are participating in an experiment and therefore act normally.
  • High external validity

Disadvantages

  • Not completely close to reality
  • Low external Validity

 

  • Environment variables that are difficult to control
  • Low internal validity

Validity check

Internal

External

Characteristics

Artificial or human-made experimental arrangement

In a natural environment, independent variables are changed under the best possible protection against other influences

 

Internal vs. External Validity

For internal validity, changes in the endogenous variable are only dependent on changes in the independent variable. The effects due to third factors can be largely excluded.

Regarding external validity, external factors are of importance. The results are generally valid and can be transferred to the population.

The greater the control over external influencing variables, the higher the internal validity. The higher the internal validity, the less realistic experiments are. Experiments represent a compromise between internal and external validity.

 

Creation of Experiment Design

Various procedures have been established for the design of experiments. Two types of design have been helpful to me—that of Lawson and that of Quenouille.

Design of experiments in five steps (according to Lawson):

  1. Definition of the object to be examined
  2. Selection of the experimental design
  3. Definition of the data collection procedure as well as the model to analyze the data
  4. Data analysis and interpretation of results
  5. Summary of the results

According to Quenouille, the following considerations are essential for the design of experiments:

  1. the need for reproducibility, i.e., to be repeatable under identical conditions
  2. the need for randomization, i.e., the ability to randomly assign a treatment
  3. Selection of the experiment design for an expectation loyalty of the treatments and the best possible accuracy

The experimental arrangement is essential to get valid results in the context of the design of experiments.

 

Experimental Arrangement

There are two types of experimental arrangement:

  • Within Subject Design
  • Between Subject Design.

Using the Within Subject Design, the participant gets all treatments. Instead, the Between Subject Design assigns participants to specific treatments. As a result, one group gets treatment A, and another group gets treatment B, for example.

 


Core literature

  • Runco, M.A.; Sakamoto, S.O. (1999): 4 Experimental Studies of Creativity. In: Handbook of Creativity. Eds.: Sternberg, R.J. Cambridge University Press, Cambridge 1999, pp. 62-92.
  • Shadish, W.R.; Cook, T.D.; Campbell, D.T. (2002): Experimental and Quasi-Experimental Designs for Generalized Causal Inference, Wadsworth Cengage learning 2002.
We use cookies

We use cookies on our website. Some of them are essential for the operation of the site, while others help us to improve this site and the user experience (tracking cookies). You can decide for yourself whether you want to allow cookies or not. Please note that if you reject them, you may not be able to use all the functionalities of the site.