Research Designs

Research design plays the determinant role in the study of different problems and conducting experiments. In actuality, the modern science is practically invalid without the research design. In this respect, social science is particularly dependant on the effective research design because social studies need a high degree of reliability of the results of the research, validity and objectivity of the study. All these purposes can hardly be met without the application of an effective research design. At the same time, there is no ideal or universal research design in the modern science. Instead, today, scientists, including social scientists, apply various research designs which can vary consistently and they applied in different context and either research design has its own strengths and weaknesses that limits and, simultaneously, specifies its applicability. In other words, some research designs are more effective in certain context, while others are more effective in a different context that makes these research designs effective in different contexts which define the specialization of research designs, i.e. their application in particular studies and circumstances, when they provide the maximum effectiveness. Nevertheless, what proves to be beyond a doubt is the necessity of the use of research design because it glues the research and makes it scientifically valuable and solid work.
In order to understand the importance of the research design, it is necessary to briefly dwell upon the essence of the research procedure. As a rule, a research involves a variety of methods and strategies which scientists use to conduct the research. At the same time, a research includes sample groups, measures, treatments or programs, and naturally methods of assignment. However, all these constituent elements of a research are not interlinked with each other. Instead, they represent a relatively chaotic selection of tools and elements which may be incorporated into a research. For instance, a social scientist can choose irrelevant sample groups which cannot provide the possibility to verify the reliability of the research because they are chosen spontaneously, without considerations on the research design. For instance, a social scientist can choose a sample group comprised of males of the same age, while the age and gender may affect consistently the effects of the research and responses of the sample group to experiments conducted by the scientist. Or else, the choice of such a sample group may be accompanied by the choice of methods of the research which are ineffective in relation to this group. It proves beyond a doubt that if a social scientist researches gender related issues and chooses a male or female sample group only than whatever methods he or she uses he or she will unable to achieve the goals of the research and study objectively gender-related issues. This example proves the fact that the research design is essential for the organization and conducting of a reliable and effective research. In fact, without the research design, the research is disorganized and disintegrated. As a result, its effectiveness and reliability decreases dramatically.
Instead, the research design contributes to the proper organization of the research and agglutination of all its constituent elements. Moreover, the research design can contribute to the selection and proper organization of elements of the research which are the most effective to achieve purpose of a specific research. In such a way, the research design proves to be the power which defines the effective organization of the research.
In such a context, benefits of the application of different types of research design are obvious. In this respect, it should be said that the weaknesses and strengths of various research designs define their applicability in different fields. Basically, it is possible to distinguish three major types of research design: randomized or true experiment, quasi-experiment and non-experiment. All these research designs imply the application of methods and strategies as well as other elements of a research which are the most effective for them. Moreover, they can be clearly distinguished into two categories: on the one hand, there are research designs that involve experiments, while, on the other hand, there are research designs that do not involve experiments or are non-experimental.
Specialists (Bangert-Drowns, 1993) often refer to experimental research designs as the most rigorous of all research designs, or as the “gold standard” against which all other designs are judged. Such a view of experimental research designs is, to a significant extent, justified because if it is possible to implement an experimental research design well then the experiment is probably the strongest design with respect to internal validity (Anderson, 1979). At this point, it is worth mentioning the fact that the internal validity is at the center of all causal or cause-effect inferences. Therefore, when a researcher wants to determine whether some program or treatment causes some outcome or outcomes to occur, then he or she is interested in having strong internal validity. To put it in simple words, the research design may be briefly worded in the following manner: if the program is given, then the outcome occurs. In fact, such an obvious simplicity is probably the major strength of experimental research design which it is possible to refer the randomized or true experiment.
On the other hand, there is quasi-experiment research design which looks like an experiment research design but lacks the key ingredient – the randomized assignment. Specialists (Borg, 1989) argue that such research designs give the experimental purists a quasi feeling. With respect to internal validity, they often appear to be inferior to randomized experiments, but, nevertheless, taken as a group, these experiments are easily more frequently implemented than their randomized cousins (Campbell, 1988).
In this respect, it is important to explain that the randomized or true experiment research design implies that the design uses randomized assignment to groups. As for the quasi-experiment design, this research design does not involve the use of randomized assignments to groups. Instead, it uses multiple groups or multiple waves of measurement. In actuality, these two types of design are quite different and have their own strengths and weaknesses. For instance, the true experiment research provides the possibility to use the sample group and to make it pass through a series of assignment which can differ consistently. In such a way, it will be possible to study in details the sample group and distinguish basic responses of the group to different assignments. As a result, a detailed analysis of the sample group may be conducted and its reactions in different contexts may be researched.
Basically, this research may be used effectively in social studies when the reactions of people on different environment are needed to study. For instance, researchers can use this research design to study the impact of various stressors on the relationship between people within the sample group. The subjects can be exposed to various stressors and after that researchers can study how their relationship within the group changes and what is the reaction of the subjects on the stressors.
However, this research design has a substantial drawback because the entire research, the experiment is focused on one sample group selected by the researcher. Hence, it is impossible to study possible effects of the experiment on different people. For instance, the impact of stress factors is different on different people, but various groups of people may develop different relationships and, therefore, the responses to the impact of stressors will be different. Consequently, their relationships will change in a different way and the use of one sample group only may be a substantial limitation for the entire research, its validity and reliability.
In stark contrast, if multiple groups or waves of measurement are used this problem may be eliminated since researchers can analyze the outcomes and results of the research using multiple groups. In such a context, if responses of different groups do not coincide the researcher can reveal basic principles and points which are common for all the subjects of the research in all of the groups.
On the other hand, the quasi-experiment research design, which involves multiple groups or waves of measurement, has its own limitation compared to the true-experiment design, since it does not use randomized assignments and, therefore, the entire research may be focused on the specific issue and it can fail to take into consideration the impact of other significant issues. For instance, returning to the example of the research of the impact of stressors on the relationship of a sample group, the quasi-experiment research design is likely to be focused on the impact of a specific type of stressors on multiple groups rather than on the use of various stressors on one sample group. As a result, the impact of one type of stressors can be researched in details only.
In such a way, it is obvious that true experiment and quasi-experiment research designs vary dramatically. At the same time, it is worth mentioning the fact that the quasi-experiment research designs can acquire quite radical forms and form non-experimental research designs which do not imply the use of multiple groups or multiple waves of measurements. However, such a type of research designs is not very popular or widely spread because specialists (Campbell, 1988) consider this type of research designs unreliable, while its internal validity is likely to be lower compared to other types of research design.
Thus, taking into consideration, all above mentioned, it is possible to conclude that the type of a research design produces a profound impact on the essence and, therefore, outcomes of the research. At the same time, a research design is an essential condition of conducting a reliable and adequate study because it is due to the research design all elements of the study glue together and form a solid and well-organized study. The proper organization of the study contributes to the reliable outcomes of the study. However it is very important to choose the proper type of research designs because their effect may vary consistently and the use of improper research design can undermine the reliability of the entire study, while its outcomes would be scientifically uncertain and inadequate.

References:
A cozy world of trivial pursuits? (1996, June 28) The Times Educational Supplement. 4174, pp. 14-15.
Anderson, J. A. (1979, Nov. 10-13). Research as argument: the experimental form. Paper presented at the annual meeting of the Speech Communication Association, San Antonio, TX.
Anderson, Linda M. (1979). Classroom-based experimental studies of teaching effectiveness in elementary schools. (Technical Report UTR&D-R- 4102). Austin: Research and Development Center for Teacher Education, University of Texas.
Asher, J. W. (1976). Educational research and evaluation methods. Boston: Little, Brown.
Babbie, Earl R. (1979). The Practice of Social Research. Belmont, CA: Wadsworth.
Bangert-Drowns, R.L. (1993). The word processor as instructional tool: a meta-analysis of word processing in writing instruction. Review of Educational Research, 63 (1), 69-93.
Beach, R. (1993). The effects of between-draft teacher evaluation versus student self-evaluation on high school students' revising of rough drafts. Research in the Teaching of English, 13, 111-119.
Beishuizen, J. & Moonen, J. (1992). Research in technology enriched schools: a case for cooperation between teachers and researchers. (ERIC Technical Report ED351006).
Borg, W. P. (1989). Educational Research: an Introduction. (5th ed.). New York: Longman.
Campbell, D. T., & Stanley, J. C. (1963). Experimental and quasi-experimental designs for research. Boston: Houghton Mifflin.
Campbell, D.T. (1988). Methodology and epistemology for social science: selected papers. ed. E. S. Overman. Chicago: University of Chicago Press.