Research Validity

The ability to draw conclusions from the results of a research study will depend on the study's validity. Validity can be characterized somewhat as falling into one of two broad categories, internal and external validity. Internal validity impacts the degree to which a researcher or consumer of research can conclude that there is a causal link between the factors under study. The purpose of random assignment, manipulation of the independent variable, and control of extraneous variables is to maximize internal validity. Common threats to internal validity that are managed with these strategies include: history, maturation, measurement error, bias in participant selection, experimental mortality/attrition, experimenter bias, expectation, and sensitization caused by pretesting. External validity basically determines the degree to which a research study's findings extend outside of the study setting and to what groups or segments of the population. Many aspects of internal and external validity can be managed through careful design of the research methods.