Experimental research design is the designing of entire research process based on two sets of variables- independent and dependent variables. In this design, independent variables are manipulated through several treatments and effects of those treatments are analysed. By this, effect of independent variable on dependent variable is obtained. The experimental research design enables gathering of relevant datas that helps us to make better decisions. An efficient experimental research facilitates smooth functioning of entire research process. That is, from writing the hypothesis to the final analysis. The three basic principles of experimental research design are:-
- Principle of Replication
- Principle of Randomisation
- Principle of Local Control
Principle of Replication
According to this principle, experiment should be repeated more than once to eliminate the main effects and interactions. Thus, statistical accuracy of experiment and precision of study are increased.
Principle of Randomisation
According to this, we can eliminate the effects of extraneous variables by proper designing of experiment. When we conduct a research, this principle provides protection against the effects caused by extraneous factors. The extraneous variables refers to independent variables that are not related to purpose of study.
Principle of Local Control
According to this principle, extraneous factors are varied intentionally and it's effects are measured. Also, these effects can be eliminated from experimental errors. Here, the research field is divided into several homogeneous parts(blocks) and these blocks are again divided into a number of parts which is equal to the number of treatments. Then, treatments are randomly provided to the parts of block.
Types of Experimental Research Design
The main classification of Experimental Research Design is as follows:-
- Informal Experimental Research Design
- Formal Experimental Research Design
Informal Experimental Research Design
It refers to the design in which least analytical procedures are performed. The subclasses of this design are:-
- Before and after without control design
- After only control design
- Before and after with control design
Formal Experimental Research Design
It refers to design in which more analytical procedures are performed. The subclasses of this design are:-
- Completely randomized design
- Randomised block design
- Latin square design
- Factorial design
Before and after without control design
One single test group or area is selected and treatments are introduced. The dependent variable is evaluated before and after the treatment. The effects of treatments are obtained by finding difference between those levels. The limitation of this type of design is that extraneous variations built up by passage of time.
Save your time!
We can take care of your essay
- Proper editing and formatting
- Free revision, title page, and bibliography
- Flexible prices and money-back guarantee
After only control design
Two groups or areas which include test and control area are selected and treatments are introduced in test area only. The dependent variables on both area is evaluated and their value difference is obtained. This design is much more superior to before and after without control design. Also, this design is very much helpful in situations where pretesting is not possible.
Before and after with control design
Two groups or areas are selected and treatments are introduced into both areas. Before and after the treatment, dependent variables on those areas are evaluated. The degree of change in dependent variables on each groups are noted. Their value difference between each areas indicates the effects of treatments. This design is much more superior compared to previous designs.
Completely randomized design
This design is centred around two basic principles- Principle of replication and Principle of randomisation. Here, to the randomly assigned subjects, treatments are provided. The various forms of completely randomized design are:-
- a. Two group simple randomized design
- b. Random replication design
Two group simple randomized design
From a well defined population, a sample is selected randomly. Those selected sample is assigned to both control and experimental groups. Two different types of treatments, that is, usual and special treatments are introduced and the amount of gains from each group is noted. By comparing the amount of gains, researcher decides whether to accept or reject the hypothesis. This design is simple and it randomises the differences among the sample items. The main limitation of this design is that it cannot control extraneous variables.
Random replication design
The main limitation of previous design is the uncontrolling of extraneous variables. Here in this design, it is rectified by providing a number of repetitions for each treatment. This design not only provides replication, but also, it randomises any individual differences among those conducting treatments.
Randomised block design
The randomised block design is an extended version of completely randomised design. This design is based on all three principles of experimental research design. That is, Principle of replication, Principle of randomisation and Principle of local control. As in previous design, subjects are divided into groups (blocks) and treatments are provided. The number of treatments will be equal to number of subjects in each block. When there are large number of experimental comparison groups, this design is used inorder to bring homogeneity among selected different groups. Also, it reduces variability among the treatment groups.
Latin square design
This type of experimental research design is widely used in the field of agricultural research. Here, two blocking factors are represented in rows and columns. It is designed in such a way that no treatment occurs more than once in any one row or any one column. This design is based on the assumption that there is no interaction between blocking factors and treatments. The main reason why this design is limited to a particular field is that the number of rows, columns and treatments are always equal.
Factorial design
Here, two or more independent variables are manipulated to measure their effects on dependent variables. This design facilitates smooth testing of several hypothesis at same time. There are two types of factorial design. They are Simple factorial desig and Complex factorial design.
Simple factorial design
It is also known as two factor factorial design. Here, two independent variables are manipulated to measure their effects on dependent variables. Hence, there are two treatments of experimental variable and two levels of control variable. Here also, the subjects are assigned at random and treated in the same way as in randomised block design. By this design, the main effects of treatments and main effects of levels are obtained. Also, it is possible to evaluate the interaction between treatments and levels.
Complex factorial design
It is also known as multifactor factorial design. Here, more than two independent variables are manipulated to measure their effects on dependent variables. Consider there are three independent variables. So, one experimental variable ( independent variable) having two treatments and two control variables and each one is having two levels. Therefore, a total of eight cells will be there. Thr interactions between all three variables are obtained by this design.
Stuck on your essay?
