Sociological research is often a complex process. Even when conducted correctly sociological research many render ambiguous finding and more questions then answers. But it is often not conducted correctly and its finding can be misused. Because of all this it is important to have understanding of sociological research methods. My goal is to give a general idea of how sociologist obtain and evaluate information about human societies. We will also examine how such information is used in constructing and testing theoretical ideas. I do not expect to make you a competent research with only a few pages devoted to research methods.
A basic research design
A basic goal of research in any discipline is to isolate cause and effect. Although the scientists may simply want to know the properties or characteristic of a substance say subatomic particles like quarks this also means learning how the substance behave under varying condition. In short we are still referring to cause and effect. In the case of quarks perhaps the building blocks of all matter their existence can be discovered only by splitting atoms in a superconducting collider. The process involves smashing particles at 20 trillion electron volts to determinate if quarks are thrown off in the process. Again we are referring to cause smashing and effect quarks set free Usually however we know the effect riots, earthquakes and are looking for the cause.
There is a basic research design that serves as the model or ideas for many science. This ideal scientific method con be called the classical experimental design. In its simplest form a classical experiment involves isolating a substance in a laboratory. Once isolated the characteristics or behavior of the substance can be studied by allowing interaction with another substance is the effect. If the substance is isolated with only the one additional substance of stimulus allowed then we can be confident this additional substance or stimulus is the cause of any observed effect. Let us me stress this point the scientist attempts to achieve total control of all possible causes by allowing only cone possible cause to be released at a time. By doing this the scientist can be certain of the cause for any observed effect.
Let consider an example. A biological wants to know why a particular type of plant grows larger in one part of the country than In another there are many possible reasons perhaps soil quality temperature moisture characteristic air quality and the lost could go on. As suggested in figure the plants can be brought into a laboratory setting and each possible cause of larger growth brought in one at a time. More realistically perhaps the scientist would also check for combination of possible cause’s particular types of soil with a particular quantity of water in common terminology the size of the plants is the dependent variable the thing we went to explain or the effect. The possible causes are independent variables. In general a variable is any substance or condition that can have a varied state. With our example of plants growth we have short medium and tall or any measure you want plants size is a variable in this case the dependent variable with our independent variables we can have varying quantities of sunlight water or carbon dioxide in the air in some cases of course we might say there is no variance of a condition. To use the old saying there is no such thing as a little pregnant you either are or you are not. But in the case of such nominal variables we can still refer to variance at only two levels it exists or it doesn’t exist.
Perhaps however the scientist is not employed to provide specific answer to specific questions for example how to make plants grow larger perhaps the scientist wants to understand plant growth in general molecular structure or genetic structure. With this example we can refer to basic research. The goal is to construct general theories of say genetic structure. When an accumulation of empirical observation through time theories of genetic structure have been developed.The scientist’s job is to test the theories through research. When working from theory we can refer to deductive research Rather than inductive research. Using another example we observe that a theoretical physicist wants to know if the theory that quarks are the building blocks of all matter is accurate. This physicist is doing basic research rather than applied research. As with applied research. However the classical experimental design can still be involved. As indicated in figure 2.2 the process of scientific research can be viewed as a cycle. In principle we may begin with either observation or theories in reality however the scientist begin with some mixture of the two. If the scientist is working more from firm theory to be tested the deductive side is involved. If the scientist is working primarily with observations and empirical generalization and is attempting to formulate theory the deductive side is involved.
Beginning with the deductive side of figure 2.2 we not that specific hypothesis must be drawn from general theories. Hypothesis can be described as specific predictions that can be tested. A very simple example might be when interacts with result will be.
The cycle of scientific research theory testing

The next step in the research cycle operartionalization By operartionalization we refer to concrete definition and indicators of the variables suggested in hypothesis. At time operartionalization is rather straightforward. In the example above cell growth can be easily defined and indicated. In sociological research however operartionalization is often every example. A sociological may want to explain the causes of revolution do we mean only cases where rebels actually take power and produce massive social change there are many more cases of the latter and it can make a big difference in research on the causes of revolution. Measurement most simply means a method of indicating amounts, degrees or condition of a variable. Measurement is sometime simple to accomplish plant height in centimeters molecules of co2 per cubic inch and so on. Some time it is quit difficult measuring the weight of unseen quarks. In the social sciences measurement is seldom, easy because we are dealing with complex attitude complex behavior unseen social forces and so on.
Finally from observation made in the research process we may from empirical generalization. These can be simply defined as generalization from the empirical observations. For example with a specific observation a scientist finds the when x and y combine the result is z. assuming the research was conducted properly with out unknown independent variable affecting the process the empirical observable generalization is whenever x and y combine the result is always z. the empirical generalization is then compared with the existing theory for used to produced one. If it is totally consistent with the theory it is supported. Most likely this is only one part of the theory and thus other parts require testing. And also most likely the findings or empirical generalization are only partially consistent with the theory or totally contradictory. In this case the theory must be amended and further research conducted.

Author's Bio: 

Written by Murad Ali. He is a graduate from Karkorum International University Gilgit, Pakistan.