Definition
An experiment is a "controlled manipulation of events, designed to produce observations that confirm or disconfirm one or more rival theories or hypotheses. To experiment is to put questions to nature, and the experimental method is contrasted with the passive acceptance of whatever observations happen along" (Answers.com, 20101).
An experiment has two important elements in their definition: "manipulation" of the variables of interest, otherwise known as treatment or intervention, and a "controlled" environment, where it is possible to eliminate the influence of any other variables than those being manipulated. Both elements are important in order to establish a causal relationship between the intervention and consequent changes occuring during or after the intervention.
In social sciences, random allocation of subjects to experimental conditions (a randomized experiment) is a common method of controlling for the possible influence of non-experimental variables. The randomization method itself does not actually "controls" the influence of those variables but allows for their influence to be randomly distributed in all experimental groups. Such random distribution is later controlled statistically.
Randomized experiment
R = randomized allocation to groups; X = treatment; - = no treatment; O = observation or measurement (called "pretest" when it occurs before treatment, and "posttest" when it occurs after treatment). The group with treatment is called the "experimental group"; the group without treatment is called the "control group".
Posttest control group design
| Experimental group | R X O |
| Control group | R - O |
Pretest - Posttest control group design
| Experimental group | R O X O |
| Control group | R O - O |
or
| Experimental group | R O X O |
| Control group | R - - O |
Solomon four group design
| Experimental group 1 | R O X O |
| Control group 1 | R O - O |
| Experimental group 2 | R X - O |
| Control group 2 | R - - O |
Want to know more?
- *http://www.socialresearchmethods.net/tutorial/Abrahams/SRM%20main.html**
- A good introduction to research design.
