There is no single rule to choose a research method out of very many options. However the broad research goal somehow gives a general indication about the possible research method to be followed for the proposed study. Different research problems have different research goals, which in rum call for different research methods. Choice of research method is guided not merely by the researcher's field of interest or specialization but also by the nature of the problem being studied, the research goals and the level of accuracy wanted. Experience has established the fact that use of one single method in social research is not always enough to respond to the research need rather a combination of methods is more useful to bring desired level of methodological sophistication. There are at least three points that should be borne in mind while choosing an appropriate research
i. Answering the research question
The method must allow the research question to be answered. It is unquestionably important to know and understand thoroughly what questions; the researcher is seeking to answer. A clear statement of the research questions will enable both the level of research and the level of rigor to be needed for the , study. It will also enable a check to be made on the understanding of the nature of the research problem involved.
ii. Current state of knowledge
The present state of art of a particular body of knowledge is also important in determining the research method. If little is known about the variables involved in the research problem then more qualitative and exploratory type of research methods are needed. If, on the other hand, a review of literature shows that a good deal is already known and it is possible to isolate the key variables involved in the process, then explanatory and experimental type of research methods can be opted. However, even where the variables are known in advance, their nature may prevent the use of experimental research methods.
iii. The nature of the variables involved
The choice of a method is governed by the extent to which the variables can be manipulated and measured in a controlled manner. In the natural sciences, it is often possible to make the subject of the research to do what the researcher wants to do. On the other hand, in social sciences, this is not always possible. Broadly speaking, a choice must be made between an experimental or non-experimental approach. Moreover, the particular design, or strategy, to be followed will be selected from a variety of available models or constructs to suit the situation. For the purposes of evaluating the alternative research methods, the following questions should be considered:
a. What is the universe to which the researcher wishes to generalize the results?
b. What level of accuracy is expected out of the method?
c. and how much bias would be tolerated ?
d. What is the maximum sampling error that would be allowed?
e. If a particular method is to be employed, will it be within the time and financial limitations set for the study?
i. Answering the research question
The method must allow the research question to be answered. It is unquestionably important to know and understand thoroughly what questions; the researcher is seeking to answer. A clear statement of the research questions will enable both the level of research and the level of rigor to be needed for the , study. It will also enable a check to be made on the understanding of the nature of the research problem involved.
ii. Current state of knowledge
The present state of art of a particular body of knowledge is also important in determining the research method. If little is known about the variables involved in the research problem then more qualitative and exploratory type of research methods are needed. If, on the other hand, a review of literature shows that a good deal is already known and it is possible to isolate the key variables involved in the process, then explanatory and experimental type of research methods can be opted. However, even where the variables are known in advance, their nature may prevent the use of experimental research methods.
iii. The nature of the variables involved
The choice of a method is governed by the extent to which the variables can be manipulated and measured in a controlled manner. In the natural sciences, it is often possible to make the subject of the research to do what the researcher wants to do. On the other hand, in social sciences, this is not always possible. Broadly speaking, a choice must be made between an experimental or non-experimental approach. Moreover, the particular design, or strategy, to be followed will be selected from a variety of available models or constructs to suit the situation. For the purposes of evaluating the alternative research methods, the following questions should be considered:
a. What is the universe to which the researcher wishes to generalize the results?
b. What level of accuracy is expected out of the method?
c. and how much bias would be tolerated ?
d. What is the maximum sampling error that would be allowed?
e. If a particular method is to be employed, will it be within the time and financial limitations set for the study?