General purpose of research designs.

The purposes and criteria for formulating a design of research, conditions for judging causality, and use of research design as a control of variance are discussed. The purpose of a research design is to provide a plan of study that permits accurate assessment of cause and effect relationships between independent and dependent variables. The classic controlled experiment is an ideal example of good research design. Factors that jeopardize the evaluation of the effect of experimental treatment (internal validity) and the generalizations derived from it (external validity) are identified. Sources of variance can be controlled by eliminating a variable, randomization, matching, or including a variable as part of the design. A research project should be so designed that (1) it answers the questions being investigated, (2) extraneous factors are controlled, and (3) the degree of generalization that can be made is valid.