Understanding the Experimental Research Design(Part II)
- 1. By Dr. Shaloo Saini Assistant Professor, MKCE (CT Group of Institutions, Jalandhar, Punjab)
- 2. Experimental Designs Pre- Experimental True- Experimental Quasi- Experimental Statistical Designs
- 3. These Research Designs follow the procedure of Experimental designs but do not include a control group. Here no comparison is made with a non treatment group. The Pre Experimental Research Design can be categorised as: One Shot Case Study One Group Pre Test Post Test study Static Group Design
- 4. This design is also called After Only Design. In this design there is a single group which is exposed to the treatment and the effect of treatment on the dependent variable is measured. This design is least reliable because it lacks random assignment, control groups and multiple observations.
- 5. This design is better than one shot case study case method because it includes a pre-test score. In this design the Dependent Variable is measured before giving any experimental treatment, then the experimental treatment is given to the subjects, thereafter the Dependent Variable is again measured. The effect of experimental treatment is measured by subtracting the pre-treatment scores from the post treatment scores. The significant difference in the pre-test and post test scores determine the effectiveness of experimental treatment.
- 6. This design is better than the previous two designs because it overcomes the drawback of absence of control group. In this design, two groups are chosen. One is experimental group that receives the treatment. Another is control group which does not receive the treatment. This design is also called ‘After Only with Control Design’ because it measures the scores of the two groups after the treatment only. The drawback of this method is that the is no pre-testing and the effect of extraneous variables also persists.
- 7. True Experimental Research Design follows the Principle of Randomization and hence they are superior to other types of Experimental Research Designs. The True Experimental Research Design can be categorised as: Pre-Test -Post-Test with Control Post-Test only with Control Solomon Four Group Design
- 8. This Experimental Research Design measure the effect of a variable on the carefully controlled sample. The test units are randomly assigned to the experimental group and the control group. The Pre test scores of test units on the Dependent Variable are recorded for both the groups. Then the test units are exposed to the Treatment and thereafter the test scores on the Dependent Variable are again recorded. The Treatment effect has been significant or not can be determined by subtracting the change in Dependent variable scores in the control area from the change in Dependent Variable scores in the experimental area. The provision of control group and randomization makes this design reliable.
- 9. In this Experimental Research Design only the post treatment scores are recorded and considered to measure the effect of Treatment given. The provision of control group and randomization makes this design reliable. Each group is randomly chosen and the treatment is administered to experimental group. Thereafter post treatment scores are recorded to determine the significant difference between the Experimental Group and Control Group. This design is easy to administer, economical and simple but it lacks pre test measure. Therefore it is difficult to determine weather the difference in the scores is due to effectiveness of the treatment or it is inherited difference which existed even before the treatment was introduced.
- 10. This design is the extension of Pre Test Post Test control group design. This design requires four groups. These groups have four different experiences: 1. Pre Test-Treatment-Post Test 2. Pre Test- No Treatment-Post Test 3. Treatment-Post Test 4. No Treatment-Post Test The effectiveness of the treatment can be evaluated by statistical comparisons between group 1 and 2 and between group 3 and 4
- 11. Quasi Experimental designs are quicker to use and less expensive, but suffer from drawbacks that they do not involve randomization. Therefore this design lacks in experimental control. The Quasi Experimental Research Designs can be categorised as: Time Series Design Multiple Time Series Design
- 12. In this experimental design the pre testing and post testing of subjects is done at different intervals of time. This design determines the long term effect of treatment therefore measurement is done over a period of time. For example the effect of advertisements on the consumer’s buying behaviour can be recorded at repeatedly for a period of time.
- 13. The basic difference between time series design and multiple time series design is that along with experimental group a control group is also introduced. In this experimental design the pre testing and post testing of subjects in experimental group and control group is done at different intervals of time.
- 14. Statistical Research Designs helps to measure and eliminate the effect of Extraneous Variables. In statistical Designs a ‘Blocking Factor’ is introduced. It is the Extraneous Variable which the researcher is able to isolate and eliminate in the effect. The Statistical Research Designs can be categorised as: Completely Randomised Randomised Block Design Latin Square Design Factorial Design
- 15. This experimental research design is based on Principles of Replications and Principles of Randomization. It is applied when the Dependent Variable is influenced by a single Independent Variable. This design can be represented using two designs: A. Two Group Simple Randomized Design B. Random Replication Design
- 16. In this design two groups are formed experimental group and control group. The test units are randomly allotted to both the groups. These two groups are subjected to different treatments of the Independent Variable. For example to test the effectiveness of some new teaching method. Students can be randomly allotted to control group and experimental group. The experimental group will be taught with new teaching method and control group will be taught with traditional method.
- 17. In this design the test units are randomly assigned to the experimental group and the control group and the treatment is repeated many times moreover the person who are giving the treatment are also randomly selected. This design not only controls the differential effect of extraneous variables but also randomizes the individual differences among the people who conduct the study
- 18. In this design the principal of Local control is applied using the blocks. The subjects are homogeneous within each block with respect to selected variable. This variable is often the extraneous variable which influence the dependent variable even though it does not explicitly form a part of the study. For Example to study the effect of rise in fair(IV) on the passenger traffic (DV), The residential Areas may act as Extraneous Variable that may influence the passenger traffic. This can be controlled by developing blocks of residential areas like Urban Area, Rural Area, Suburban Areas. The Busses operating within each block are randomly assigned to the treatment. This design can be used to study the effect of fair rise on passenger traffic and the interaction between Residential area and fair rise by applying ANOVA.
- 19. Latin Square Design can be used to control two Extraneous variables. Taking again the example of rise in fair and passenger traffic. Two extraneous Variable Residential areas and Type of Bus can be controlled by Latin Square. The rows contains Types of buses- Luxary, Delux and Regular. The column contains the types of residential area blocks- Urban, Rural, Suburban. The column reveals the effect of residential areas on the passenger traffic and the rows reveal the effect of type of bus on the passenger traffic. After the experiment is carried out and the net increase in passengers under each treatment calculated the average effect of treatment can also be worked out.
- 20. Factorial Designs are used where the effect of more than one Independent Variable on the Dependent Variable is to be measured. Unlike any other statistical designs, the Factorial Designs helps to study the interaction between the variables. Factorial Designs are of Two Types: 1. Simple Factorial Design: A simple factorial design studies the effect of two factors on the Dependent Variable. 2. Complex Factorial Design: Complex Factorial Designs are used when more than two factors are to be studied simultaneously on the Dependent Variable.
- 21. There are four Types of Experimental Research Designs which are, 1. Pre Experimental: These Research Designs follow the procedure of experimental designs but do not include a control group. Here no comparison is made with a non treatment group. 2.True Experimental: These Research Design follows the Principle of Randomization and hence they are superior to other types of Experimental Research Designs. 3. Quasi Experimental: These Research Designs are quicker to use and less expensive, but suffer from drawbacks that they do not involve randomization. Therefore this design lacks in experimental control. 4. Statistical Research Designs: These Research Designs helps to measure and eliminate the effect of Extraneous Variables. In statistical Designs a ‘Blocking Factor’ is introduced. It is the Extraneous Variable which the researcher is able to isolate and eliminate in the effect.
- 22. Bajpai N. (2015). Research design formulations. In Business research methods (pp. 49). Nodia: Pearson Education. Cooper, D. R., Schindler, P. S., & Sharma, J. K. (2012). Measurement scales. In Business Research Methods (8th ed., pp. 341). New Delhi: Mc. Graw Hill Education (India). Gupta S.K. & Rangi P.(2017). Research Designs In Research Methodology (4th ed., pp3.5).Punjab: Kalyani Publishers(India). Kothari C. R. & Garg G.(2019). Research Designs. In Research Methodology Methods and Techniques( 4th ed., pp.29-38) New Delhi: New Age International Publishers(India).
- 23. This Power Point Presentation has been made while referring to the research books written by eminent, renowned and expert authors as mentioned in the references section. The purpose of this Presentation is to help the research students in developing an insight about the Experimental Research Design(Part- II). I hope the students will find this presentation useful for them. All the Best Dr. Shaloo Saini