Experimental design techniques

experimental design techniques

1. Experimental design techniques

2. The design of fundamental relationship under controlled situation is called experimental design. Controlling is nothing but holding one variable constant while other variables are free to vary within the experiment. Independent variables are manipulated in order to measure their effect on dependent variable.

3. An experiment is an observation under controlled conditions. It is a design in which some of the variables being studied are manipulated to control the conditions.

4. Types Pre –experimental designs Quasi-experimental design Observational or true experimental design

5. Pre- experimental designs It is the design in which the basic experimental steps are followed without assigning any control group. The researcher studies a single group with no any comparison within this group. It is a preparatory to true experimental design. It is the simplest form of research design.

6. Quasi experimental design The study subject are not randomly assigned to the groups. Manipulation of an independent variable. It differs from experimental research because there is no control, random selection and no active manipulation. It is performed when a control group cannot be created or random selection cannot be performed. It is used in certain medical and paramedical studies.

7. True experimental design It employs statistical analysis to support or reject a hypothesis. It has control group. It is the most accurate type of experimental research it supports or disproves the hypothesis using statistical analysis It is the only design where it cause and effect the relationship between the variables.

8. True experimental design satisfy the following There must be a control group which wont change An experimental group which will experience the changed variables. There must be a variables that can be manipulated by researcher. There must be a random distribution It can be further categorized as

9. Post test only control group design

10. After-only design Experimental and control group are similar. Uncontrolled factors affect both the groups casually. The experimental group is exposed to assumed variable but the control group remains unexposed. After experimentation both the groups are compared there is some effect produced in the experimental group

11. The control group assumed to e exactly same but it is not true.

12. Before after desi Dependent variables is measured both before and after the exposure of groups to experiment. It use several group. It has one or more control group. Major weakness is a group is influenced by the external factors . It is more reliable than after- only d

13. Ex-post design Researcher depend on historical background t, because sometimes it is not possible to divide the population into two clear and similar group. It is necessary to study the entire historical background. The past is studied through the present.

14. Panel study designs Method of study of a particular subject over time by using different kinds of data. The researcher secure direct evidences of time dependent relationship among variables. It involves repeated observations on the same subject at different period of time is known as time-series study. It is continuous , deep and reliable design It has limitation such as loss of panel member absenteeism and rigid attitude of members.

15. Post -test only design It has two randomly assigned groups. Experimental group and control group. The treatment is applied to experimental group. The post-test is conducted on both the group to measure the effects of the treatment. When pretesting is not possible or not required this design is commonly used.

16. Pre test post test only design In this design experimental and control group are formed They are pretested for independent variable. During experiment only experimental group receives the treatment. Finally both the groups are post tested to investigate the effects of the independent variable on the dependent one.

17. Solomon four group design Four groups randomly formed. Two experimental and two control group. One pre tested and one un pretested group receives the treatment. All four group receives the post test. Post test result demonstrate the effect of the dependent variable comparing to the effect of the independent variable on the dependent variable. It is used to eliminate potential source of error.

18. Factorial design In this design two or more independent variable are simultaneously manipulated to observe their effects on the dependent variable. It is useful to the researcher to test two or more hypothesis in a single experiment. The independent variables in this design are called as factors.

19. Subdivision of factors is called a levels. It helps the researcher to investigate the individual effect of each treatment on the dependent variables as well as their joint effect.

20. Randomised block design Subject population is grouped into relatively homogenous subgroups [ blocks] within which the experiment is replicated It is used when there is inherent difference between subjects and possible differences in experimental conditions. If there are a large number of experimental group the randomised block design is used because this design make the group homogenous.

21. It is used when need to reduce the noise or variance in data. The noise may be attributed to the differences between the blocks . This design helps to detect and examine actual effects of interest much more accurately.

22. Cross over design Different orders of treatment is randomly assigned to the subjects. It is also known as repeat measures design. The groups for comparison need to have an equal distribution as well as there should be high level of similarity among the subjects. The subject in this design serve as their own control

23. Limitation is subjects exposed to first treatment may effect their responses to second treatment. Experimental research is aimed to draw conclusions by controlling the external factors in order to provide definite conclusions about the effect of the independent variables on the dependent variables. The purpose of experimental research design is to test the hypotheses of a fundamental relationship between variables.