Research Methods & Analysis

Answer : Sampling is an important method used in the formulation of social research. As the rate of social change has increased, the importance of sampling as a technique of social research is also increased.

For practical and cost effective reasons, it is often impossible to collect information about entire population of people of things in which social researchers are interested.

In these cases, a sample of the total is selected for study. The main criteria are to ensure that a sample provides faithful representation of the totality from which it is selected, and to know as precisely as possible the probability that a sample is reliable in this way.

Randomization meets these criteria, because it protects against bias in the selection process and also provides a basis for applying statistical distribution theory. It allows an estimate to be made of the probability that conclusion drawn from the sample are correct.

Sample is a collection of observations for which one has data with which he/she is going to work. Almost any set of observation for which one has data constitutes a sample. In other words, sampling means a method in which a portion of the Universe is selected with a view to drawing conclusions about the Universe or population.

Population is not necessarily synonymous with a population of a community or a State. Population in the manner is used in sampling statistics is constituted of all the individuals, things, events, documents or observation, belonging to a designated category characterizing specific attributes which a particular study should principally cover. A.L. Bowley has done a pioneering work on sampling statistics in the realm of social sciences. The importance of sampling lies in the vastness and unorganized social structure.

The foremost importance of sampling is time saving. As societies are dynamic and particularly rapid changes are taking place in modem societies, so sampling enables to use short-cut methods or else the measurement would be out of date before the study on the population is completed. Sampling helps to study limited population in its totality. As social reality is vast and unorganized, sampling limits the boundaries of social research and save money and manpower.

As sampling is done on small group it is possible to repeat the research on social group again and improve the validity of tabulation and analysis.

Simple random sampling is in a sense, the basic theme of all scientific sampling. It is the primary probability sampling design. Indeed, all other methods of scientific sampling are variations of the simple random sampling A simple random sample is selected by a process that not only gives to each element in the population an equal chance of being included in the sample but also makes the selection of every possible combination of cases in the desired size, equally likely.

Suppose, for example, that one wants to have a simple random sample of two cases from a population of six cases – A, B, C, D, E and F. There will be the following possible combinations of cases, each having two elements from this population viz. AB, AC, AD, AE, AF, BC, BD, BE, BF, CD, CE, CF, EF, DE, EF i.e. in all 15 combinations. If we write each combination on equal cards and mix then ask even a blind man to pick, each of the cards will be afforded the same chance of being selected in the sample.

In the stratified random sampling, the population is first divided into a number of strata. Such strata may be based on a single criterion (e.g. educational level) or a combination of two or more criteria (e.g. age and sex). In stratified random sampling, a simple random sample is taken from each of the strata and such sub samples are brought together to form the total sample.

In general , stratification of the Universe for the purpose of sampling contributes to the efficiency of sampling if it establishes classes, i.e. if it candivide the population into classes of members or elements that are internally comparatively homogenous and relative to one another, heterogeneous, with respect to the characteristics being studied.

It is quite possible that the process of breaking the population down into strata that are internally homogenous and relatively heterogeneous in respect of certain relevant characteristics is prohibitively costly. In such a situation, the researcher may choose to select a large simple random sample to make up high cost by increasing total size of the sample and avoiding hazards attendant upon stratification and not to apply stratified random sampling.

Answer : Once the problem to be answered in the course of research is finally instituted, the researcher may, if feasible, proceed to formulate tentative solutions or answers to it. These proposed solutions or explanations are called hypothesis, which the researcher is obliged to test on the basis of facts already known or which can be made known. If such hypothesis are not formulated even implicitly, the researcher cannot effectively go ahead with the investigation of his problem, because, in the absence of direction which hypothesis typically provide, the researcher would not know what facts to look for and what relation or order to search for among them. Lundberg aptly remarks:

“The only difference between gathering data without a hypothesis and gathering them with one is that in the latter case, we deliberately recognize the limitation of our senses and attempt to reduce their fallibility by limiting our field of investigation so as to prevent greater concentration of attention on particular aspects which past experience lead us to believe are insignificant for our purpose. Hypotheses may be developed from a variety of sources.

We examine some major ones. Personal life-stories are a factor in determines the kinds of a Parson’s perception and conception and this factor may in turn direct him to certain hypothesis quite readily. For example, Thorstein Veblen, being an outsider studied capitalism in England objectively and without any knowledge of it. Analogies are often a fountainhead of valuable hypotheses. Students of sociology and political science in the course of their studies would have come across analogies wherein society & state are compared to biological organism.

Hypothesis may also rest on the finding of other studies. The researcher on the basis of the findings of other studies may hypothesize that similar relationship between specified variables will hold good in the present study too.

Answer : Objectivity and value neutrality in social research can be understood from the viewpoint of methods and procedures of positivistic and interpretative discipline in social science. August Comte argued that the application of natural science methodology to the study of man would produce a “positive science of society” that represents objectivity in social research. It would show that behaviour in the social world is governed by laws in the same way as bahaviour in the natural world. But on the other, many sociologists reject the view that natural science methodology is appropriate for the study of human behaviour. As a result the natural scientist is not required to explore and take account of the internal consciousness of matter in order to explain its behaviour. The way they understand i.e. interpretative discipline are very much related to subjectivity in social Research. But the fact is that the advocator of objectivity simply observe its activity from the outside. They are able to explain its behaviour merely as a reaction to all external stimulus. In this sense, the data of the natural sciences are draws from direct observation of the behaviour of matter, behaviour which can be qualified according to objective criteria. The same procedures are possible in the observation of human behaviour.

The moral responsibility of the sociologists is the centre of value Judgement in sociology. In research, a scientist does not judge whether it is good or bad for chemicals to combine in a certain way or for the birth rate to go up or down. A researcher is seeking the truth whatever it might be. It is unethical to fake the results of an experiment or other research, so that one can report the kinds of results one wants to find. Sometimes the researcher’s desire unconsciously affects the results of the study. The approach and perspective behind value-neutrality is related in Weberian understanding of the notion of ‘Verstehern’. Weber has given the emphasis that in social research complete value-free observation can’t be possible. Therefore, to study any social phenomena or problems, it needs to observe from the value-neutral basis. It tries to maintain the objectivity in social research.

Answer : Problem of objectivity in social research: As compared with natural science, there is much of a problem of obtaining objectivity in social research. The problem is primarily because in social research we deal with human beings who live in society and have certain social interests. They participate in social movements and accept certain values and ways of life. They have certain social and emotional attachments which can’t be detached from real life. As such their behaviour, attitude and approach can influence social life.

Infact, there are many difficulties in obtaining objectivity in social research. Personal views, concepts and beliefs of the investigators are bound to creep in the study. An investigator is also influenced by the social, political and economic conditions, situations and environments around him. The problem of objectivity may be outlined in many points: (i) Emotional effects (ii) Complexity of subject matter (iii) Lack of Unifformity (iv) Problem of general knowledge (v) Nature of subject matter (vii) Social philosophy and its effects (viii) Problem of hurried research (ix) Problem of morality (x) Problem of external pressures (xi) The notion of ethnocentrism (xii) Prejudices and bias (xiii) Problem of prevalent notions (xiv) Customs (xv) Vested interests

Value neutrality in social research: A debate in social research continues in terms of value-free, value-neutral and value-laden. Weber considers that in social research, value-free approach can not be possible as positivist pointed out in terms of observation, comparison and experimentation. Weber was a firm advocate of separating the rationality of social inquiry from the attempt to make the world more rational. He argued that the personal evaluation of the sociologist must be separated from the analysis of society that he/she conducted. The main points in this debate of Weber’s understanding can be briefly state as:

(i) The sociologist in his/her study of society is principally concerned with the analysis and understanding of values, as these are the crucial elements of any society. However, the sociologist must not let his/her own values come in the way of a clear understanding of the subject matter.

(ii) The sociologist, as a human being is fundamentally involved in evaluating, or passing certain value judgement. In so far as he is personally involved he himself can hardly avoid having or living by values. The values by which sociology develops, however, are the values by which knowledge and science develop as the commitment to dispassionate inquiry. In this inquiry, the sociologist’s own experience of valuing or disvaluing is itself a data, giving insight into the meaning and relevance of that which he explores.

(iii) The development of a value-free social science in this sense is necessary for creating a body of reliable and assured knowledge. As to whether such knowledge becomes subsequently cause for action is not within the domain of merely sociology as a discipline. Knowledge can guide action only when the discipline itself becomes reliable.

Having discussed the problem of objectivity and value-neutrality in Social Research, we have to examine the limitations associated with the tools of measurement in Social Science Researches. One method can be to find out whether the results still remain consistent even if the measures are consistent and the individuals and the situation remains the same, but test carried out under independent situations. In order to eliminate the chances of errors, it is always better, if there are repeated observations and measurements of the same individuals or situations. But repeated experiments can’t be carried out on human beings because characteristics of the study will go on changing and also any such attempt is likely to create fatigue and annoyance. It is therefore desirable that in such a situation the experiments might be carried out on the sample of the universe rather than universe as a whole.

Answer : In social research we aim at opening new vistas of knowledge and wisdom. Social research helps in finding out social reality, simplifying the mysterious within the common place in social life. It is with the help of social research that social planning can become possible.

It helps in finding similarities and differences of various cultures the methods in which different organizations control different cultures and influence social behaviour and also the extent of different problems which face the society. When facts about cultures and social behaviour are known it becomes possible to have greater power of control over social phenomenon.

It helps in promoting better understanding and social cohesion and brings to surface underlying unity out of diversity of human societies and vice-versa.

A good social research will also help in pointing out the structural changes needed by the society and under the existing situations how far is it possible to introduce such changes and with what speed.

Appropriate welfare reforms can be introduced in the society only when social research helps in finding out their needs and necessities. Even state can introduce legislative measures with the help of the findings of social researchers.

Social research helps the society in making predictions and goals for the future. It becomes possible because they have necessary data at their disposal.

Every social research helps in improving old techniques and innovating the new ones. Exiting tools of research are modified.

Answer : To many sociologists, an objective science of society remains the goal of Sociology. Their view of objectivity is summarized: “objectivity means that the conclusion arrived at as the result of inquiry and investigation are independent of the race, colour, creed, occupation, nationality, religion, moral preference and political predisposition of the investigators.” If his research is truly objective , it is independent of any subjective element, the personal desires, that he may have. An increasing number of sociologists now argue that pursuit of an objective value-free Sociology is pursuit of an illusion. Sociologist cannot shed his taken for granted assumptions, prejudices and commitments. He cannot escape from his past experiences which will colour everything he sees. With respect to methodology, it means that the entire research process will be influenced by these factors. They will shape the sociologists assumption about the nature of social reality. Some of the arguments are being advanced for Sociology not being objective : (1) sociological judgement are subjective, being coloured by actors’ own experiences. (2) All propositions are limited in their meaning to particular language context (3) All sociological theories are produced by, and limited to particular social group. Such a doctrine is often taken to be an outcome of the Sociology of knowledge which treats all knowledge as a function of social location. (4) All observation are necessarily theory —laden. (5) In that all members of the society have different values, sociologists will unconsciously but necessarily, have their arguments influenced by their values. Some sociologists argue that objectivity in some or all of the above sense is not necessarily desirable. It is argued that, the sociologists, for example, should be critical and espouse particular values.

Answer : Questionnire is one of the important way of data collection. It asks open-ended and closed ended question from the repondent to know about the issues. But sometimes respondent gives such a answer to the question that it would be hard to say it is objective or true. For example, in a Gallup Poll Survey in 1939, 88% of the US population described themselves as middle class. The survey was repeated again and the ' lower class was replaced by working class. Now, the 51% people described themselves as working class. Thus there is the problem of language. The standard or literary languages are not properly understood by the illiterate population. Questionnaire, due to this reason should be applied to the uniform group. Both, for the literate and illiterate, separate questionnire should be prepared. Secondly, for the various religious, cultural, racial and stratified group, separate questionnire according to their characteristics should be prepared, if not so, the result may be negative. In Indian context also, there is the sharp rural-urban distinction based on caste, language, poverty, unemployment, stratification etc. Thus, it is difficult to get exact data by applying uniform sampling.

Answer : The reliability of sample taken from a population can be assessed by the spread of sampling distribution, measured by standared deviation of this distribution, called the standared errors. Suppose we have a population of N items, of which a certain proportion posses a given car. Out of this population a number of different sample of size can be drawn, this number is very large, when it is small in relation to N. In each of these samples there will be certain proportion of persons who are car owners and the frequency distribution of these proportions is called sampling distribution of ‘P, and the standard deviation of the distribution is called standard error of ‘P, I may be shown that for large values of n. Say n, this distribution is tend towards the normal of Gaussian form and that the standared error of P will be thus in 96 percent of all samples the proportion of car owner will lie in the range.

The larger the value of the simple size n, the smaller the standared error. A little study of the formula will show that the sample size n is much more important in determining the size of the standared error, than the sampling fraction n/N. which shows the proportion of the sample to the total population. At first sight this leads to the surprising result that a sample of 2500 individuals has much the same reliability. Whether it comes from O population of 100,000 individuals or one million. It is alway the aim of the efficient sampling method to give maxium reliability at minimum cost, but it would be clear that any sampling design must achieve a campromise.