An experiment is a scientific device for obtaining an answer to the question in experimental research, “If this is done, under carefully controlled conditions, what will happen?” An experiment is a situation in which the relationship between cause and effect is determined. The underlying logic is: If two situations are alike in every respect and if one element is added to one but not to the other, the resultant difference may be considered to be the effect of the element added. Similarly, suppose two situations are alike in every respect and one element is removed from one situation but not from the other. In that case, any observable difference may be attributed to the element removed.
There are five key steps in designing an experiment:
- Consider your variables and how they are related
- Write a specific, testable hypothesis
- Design experimental treatments to manipulate your independent variable
- Assign subjects to groups, either between-subjects or within-subjects
- Plan how you will measure your dependent variable
Experimental Validity
In most experimental research carried out outside the laboratory, several extraneous variables tend to confound the result in such a way that, it is difficult to evaluate their influence on the result. When such extraneous variables affect the results it becomes difficult to conclude that the independent variable causes changes observed in the dependent variable. The extraneous variables in effect introduce rival hypotheses which could account for the observed changes not attributable to the experimental variable under consideration. Although these extraneous variables cannot be eliminated a researcher should be able to identify and anticipate them and take possible measures to minimize their influence through appropriate research design and careful execution. So that the conclusions arrived at through experimental research are accepted as true, the experiment must have internal and external validity.
Read: The need and importance of research in education
Internal Validity
Internal validity is the basic minimum limit necessary to make the results of the experiment interpretable. Internal validity questions whether the experimental treatment makes a difference in the dependent variable. An adequate answer to this question requires adequate internal validity. Before it can be answered, the researcher must be confident that extraneous variables have not produced an effect that is being mistaken as an effect of the experimental treatment. According to Best, an experiment has internal validity to the extent that the factors that have been manipulated (independent variables) have a genuine effect on the observed consequences (dependent variables) in the experiment setting.
External Validity
The value of experimental research lies in its ability to provide a basis for generalizing from a sample to the total population. The conclusion of a research study that spaced drill improved spelling more quickly among ten students of class IV in Pir Sohawa than regular drilling has very little practical value unless the research enables us to generalize to all primary school children.
External validity concerns the power of the experiment to generalize variable relationships Lu a wider population of interest and to non-experimental settings. Achieving external validity is difficult but sound experimental design, appropriate statistical analysis and many replications with varied population samples in a variety of settings may minimize the problem. (Best: 153)
Experimental validity is an ideal to aspire to, for it is unlikely that it can ever be completely achieved. Internal validity is very difficult. to achieve in the non-laboratory setting of the behavioral experiment where there are so many extraneous variables to attempt to control. When experimental controls are tightened to achieve internal validity, the more artificial, less realistic situation may prevail, reducing the external validity or generalizability of the experiment.
Some comparison is inevitable so that a reasonable balance may be established between control and generalizability between internal and external validity. To gain maximum experimental validity, some efforts are made to improve the experimental designs, which are rarely, if ever, perfect. According to William Wersma, both internal and external validity are important, and through experimental design, we attempt to embrace adequate validity. Since enhancing one type of validity may tend to jeopardize the other, we often attempt to secure an adequate compromise. The compromise is essentially that of attaining sufficient control to make the result interpretable while maintaining enough realism so that the results will generalize adequately to the intended situations.
Read our blog: PakSci Mission
