Five Threats to Internal Validity in Experimental Designs

Five threats to internal validity in experimental designs

This paper will identify and discuss a least five threats to the internal validity in experimental design. In this paper, we will discuss ways of controlling each of those threats that the researcher identified.

Before we can discuss threats to internal validity in experimental designs, we must first understand just what internal validity is. Internal validity is the type of validity that only relates to the exact accomplishment of the behavior or plans that the researcher evaluated. Internal validity permits a researcher to maintain that it is, in fact, the behavior or plans that caused a change in the group that was treated ("Internal validity," 2010).

History threats can create a threat to internal validity when the group under consideration experiences an occasion unconnected to the action for which an impact on their presentation of the posttest. For example, if the researcher is evaluating a drug deterrence program, and at some point of the study a accepted personality dies because of a drug overdose, the outcome that is observed may have been grounds for the groups response to the famous person death rather than the program that was executed (White, 2010).

Maturation threats effects that the research contributors are growing older, more knowledgeable, intelligent, and more skilled. Between the pretest and the posttest, and this may cause the experimental effect rather than the behavior that they are studying. For example if you give a group of fifth graders a particular mathematics test and monitor their mathematic abilities get better between the pretest and the posttest, this difference may be caused by the actuality that the children have grown older and are merely better able to understand math conceptions (White, 2010).

Testing can be a threat to internal validity while the fact that a group has taken a pretest or a number of pretests causes an examined effect on the posttest. A group may do better on a posttest not for the reason that the behavior that was applied, but because the pretest prepared the group members to do better. For example, if fifth graders were given a mathematics pretest before they are given special mathematics program. The fifth graders might remember some of the problems and correct answers. The fifth graders are given the test; again, they may answer more of the questions correctly because of the fifth graders knowledge with the test rather than their math skills (White, 2010).

Instrumentation threats is when the experimental effect is caused by a difference in the why the pretest and posttest are considered, rather than the impact of the treatment that was applied. This can occur when the dimension depends on observers who become experienced over time, or if the observers change from the pretest to the posttest. This can also occur if the dimension test

...