ADVERTISEMENTS:
This article throws light upon the top seven methods used for conducting social research. The methods are: 1. Observation Method 2. Interview Method 3. The Questionnaire Method 4. Projective Techniques 5. Scaling Technique 6. Sampling Method 7. Case Study Method.
1. Observation Method:
We are almost constantly engaged in observation of things, objects, processes and even thought-ways of people. It is our basic method of obtaining information about the world around us. All observation, however, is not scientific observation.
Observation becomes a scientific tool for the researcher to the extent that it serves a formulated research purpose, is planned systematically, is related to more general theoretical proposition, is recorded systematically and is subjected to check and controls on validity and reliability.
ADVERTISEMENTS:
This is not to say, however, that observations of great scientific import cannot occasionally be made just by stumbling across the unexpected or chance coincidences.
The history of science is replete with evidences of valuable discoveries founded on unsystematic, haphazard and casual observations that were unrelated to some pre-defined and established research purpose, e.g., the discovery of radium and penicillin.
Many types of data sought by the social scientists can be obtained through direct observation. Direct observation of behaviour, of course, is not the only method by which the scientist can obtain data. Interviews, questionnaires, records, etc., can under certain situations, replace and supplement observations by the scientist.
“But indeed”, as John Dollard puts it, “the primary research instrument would seem to be the observing human intelligence trying to make sense out of human experience….”
ADVERTISEMENTS:
Compared to the other methods of data collection, observation has some distinct advantages:
1. One great asset of the observational technique is that it is possible to record behaviour as it occurs. Many other research techniques depend entirely on people’s retrospective or anticipatory reports of their own behaviour. But these reports are by and large given when the respondent is somewhat removed from stress and strains that influence his behaviour in the ordinary routine.
However, at this time, the respondent is influenced by other pressures peculiar to the research situation. The observational techniques yield data that pertain directly to typical behavioural situations. A researcher would prefer observational methods, should he have reasons to believe that distortions in ‘recall’ or reminiscence are likely to occur.
2. Most items of our behaviour are so much a part of our habit that they escape an aware detection and resist translation into words. Anthropologists observing foreign cultures have noted that many facts worth recording are so much taken for granted by the native people that they do not think they are worthy of reporting.
3. Studies may deal with subjects who are not capable of giving verbal reports of their behaviour or feelings for the plain reason that they cannot speak, e.g., infants or animals. Such studies necessarily depend on observation.
4. Observation is independent of people’s willingness to report. Many a time, a researcher meets with resistance from persons being studied. People may not have the time or they may be unwilling to be interviewed or tested.
Although observation cannot always overcome such resistance, it must be conceded that relatively speaking, it is less demanding of active co-operation on the part of the subjects and hence less exacting for the subjects.
We assume that observations would yield data which are comparable on the inter- observer basis and hence can afford well grounded generalizations.
But we can no longer talk about the ‘observer’ in abstract, because we have experience of only one observer, namely, the person we can talk about and observe my body even my thoughts, but whenever I regard myself as an object in this way, there is sense in which I am still the subject who observes.
ADVERTISEMENTS:
If in any scientific activity an observer confronts what is observed, sometimes the observation is reflexive, and the same person plays both roles.
The only experiences of observation to which I have direct access are my own, and I cannot tell if the experiences of others are like mine, even if they are described to me, for the experiences themselves are private and ‘Sui Generis’, whereas the description is public and makes use of the categories of language.
But this very thing may go against the observational method not so much for methodological, as for ethical reasons. For example, in the participant observation, insights are often gained at the expense of morality (e.g., where there is deception of subjects).
Observational techniques are, of course, not without their own brand of limitations. The major ones are as follows:
(1) It is often impossible to anticipate the occurrence of an event precisely enough to be able to be present to observe it. Even the observation of regular daily occurrences sometimes becomes difficult because of the possibility that unforeseen factors may interfere with the observational task.
(2) The practical possibility of applying observational techniques is limited by the duration of events. For example, life histories cannot be generated this way. Besides, some occurrences that people may not be willing and able to report are rarely accessible to direct observation (for example, private behaviour).
(3) It is often held that observational data cannot be quantified. This, however, is a gross misconception. It should be remembered that the anthropologists were pioneers in the use of observational techniques and did not feel the need for quantifying their observation.
This is no reason to suppose that observational data are typically un-amenable to quantification. Social researchers will do well to bear in mind that observational data like other data are not incapable of being quantified.
Observation may serve a variety of research purposes. It may be used to explore the given area of subject-matter or to gain insight into the research problem and provide a basis for development of hypotheses. It may also be used to gather supplementary material that would help interpretation of findings obtained by other techniques.
Lastly, observation may also be used as the primary method of data-collection in descriptive studies as also, in the experimental studies designed for testing causal hypotheses.
2. Interview Method:
The interview method is quite effective in giving information about all these aspects. G. W. Allport in his classic statement, sums this up beautifully. He says, “If you want to know how people feel, what they experience and what they remember, what their emotions and motives are like, and the reasons for acting as they do — why not ask them?”
The interview approach simply stated, involves a person designated the interviewer asking questions (mostly) in a face-to-face contact (generally) to the other person or persons, designated the interviewee/s who give answers (mostly) to these questions.
It does not mean, as the term ‘mostly’ in the brackets suggests, that all the time it is interviewer who asks questions. On rare occasions, the interviewee may also ask certain questions and the interviewer responds to these. In any case, it is certain that the interviewer initiates the interaction (interview) and the interviewee is at the receiving end.
In so far as it affords a “portrait of human personality”, i.e., information about the social background which governs one’s scheme of life, one’s inner strivings, tensions, wishes and the changes in one’s behavioural relations, the interview has been a widely used method in empirical studies.
Thomas and Znaniecki used this method to get at some aspects of their subject, ‘the Polish Peasant’, concerning the patterns of adjustment of immigrants to new social situations.
Some of the most sophisticated interview techniques were used in their study entitled, “The Authoritarian Personality” by Adorno and associates. Stouffer and associates used the interview method extensively in their celebrated study entitled “The American Soldier.”
Interview may be said to be one of the most commonly used techniques (usually employed with other techniques of data collection but also on occasions, singly) of data collection in studies of human behaviour. The reassessment of the qualitative interviews has helped the interview method attain tremendous importance in contemporary research in the field of social and behavioural science.
3. The Questionnaire Method:
A questionnaire consists of a number of questions printed (or typed) in a definite order on a form (or set of forms).
The form/s are usually mailed to the respondents who are expected to read and understand the questions and reply to them in writing in the relevant spaces provided for the questions on the said form/s. Ideally, the respondent has to answer the questions on his own, i.e., totally unaided. A schedule also has a reference to proforma containing a set of questions.
The researcher/interviewer puts to respondents the questions from the proforma in the order as these are listed and records the replies. In certain situations, the schedules may be handed over to the respondents and the interviewer may get these filled in his presence, offering necessary explanations with reference to the questions if and when necessary.
The signal advantage of the questionnaire method is that it offers great facility in collecting data from large, diverse and widely-scattered groups of people. The distinctive characteristic of the questionnaire has been aptly summarized by Johan as ‘Written- verbal stimulus’ and ‘Written-verbal response.’ It is used in gathering objective, quantitative data as well as for securing information of a qualitative nature.
In some studies, the questionnaire is the sole research tool utilized but it is more often used in conjunction with other methods of investigation. In the questionnaire (as also in the interview) technique, great reliance is placed on questions which he is exposed to as also for data on his behaviour.
The subjects’ reports may not be taken at face value; these may be interpreted on the basis of other available knowledge about them (subjects) or in terms of some psychological principles. Needless to say, the questionnaire (also interview) approach can normally help obtain only materials that the respondent is willing and able to report.
It is well worth noting that persons are not only reluctant to openly report their feelings, plans, fears and so on; they may in point of fact, be unable to do so. We may not be aware of many of our beliefs and hence may not be able to report them.
Nevertheless, each one of us has a unique opportunity to observe himself and to that extent one is in a position to and often will communicate this knowledge about himself to others.
But such reporting or communication, especially one that diagnoses and explains why one’s behaviour was what it was, requires qualities of penetration much beyond the reach of average persons. It is given to only a few to be able to engage in self-diagnosis.
The capacity to peep into depths of one’s personality is conspicuous by its absence among the people at large. It is precisely this that works to the detriment of the efficacy of the questionnaire method. Despite the limitations of self-report, it is often possible and useful to get people’s own accounts of their feelings, attitudes, etc., by means of questionnaires.
4. Projective Techniques:
A projective test involves the presentation of a stimulus situation devised or selected, because it will mean to the subject not what the experimenter has arbitrarily decided it should mean, but rather whatever it must mean to the ‘personality’ who gives it his personal, private, idiosyncratic meaning and organization.
The assumption underlying the projective test is that the individual’s organization of the relatively un-structure stimulus situations is indicative of the basic trends in his perception of the world and his response to it.
Projective techniques were first devised by psychologists and psychiatrists concerned with the diagnosis and treatment of patients afflicted by emotional disorders. Such tests attempt to give a comprehensive, picture of the individual’s personality structure, his emotional needs or his conflicts and complexes.
The use of such tests, however, requires intensive specialized training. In so far as certain tests have often been employed with advantage in the investigation of certain kinds of problems in sociology, social psychology and anthropology.
In projective tests, the individual’s responses to the stimulus situation (a photograph or a symmetrical but meaningless ink-blot design) are not taken at their face value. The stimuli may arouse in the subjects many different kinds of reactions. There are thus, no ‘right’ or ‘wrong’ answers. The emphasis is on his perception or the meaning he gives to it and the way to which he organizes it or manipulates it (perception).
The nature of the stimuli and the way in which they are presented do not clearly indicate the purpose of test or the way in which the response is to be interpreted. The individual is not asked to asked to talk about himself directly.
The ostensible subject-matter or stimulus may be a photograph, a picture, an ink-blot, etc. However, the responses to these stimuli are interpreted as indicating the individual’s own view of the world, his personality structure, his needs, tensions and anxieties, etc.
The are interpreted in terms of some pre-established psychological conceptualization of what the person’s responses to the stimulus (test situation) mean.
Rorschach Test:
One of the more frequently used projective techniques is the Rorschach test. This test consists of ten cards having prints of ink-blots. The design is symmetrical but meaningless (see the picture below):
The subject is asked “What might this be?” The subject’s responses, e.g.,” This could be two women gossiping” or “Reminds me of the human lungs” or “A butterfly”,etc., are interpreted on the basis of some pre-established psychological framework .
Obviously, the task of interpreting as to what a particular response means in terms of the subject’s personality characteristic, is indeed, a very difficult, tricky and specialized one. Interpretations may not always be the same for different scientists working on the same response. There is also the problem of validity.
Thematic Apperception Test (T.A.T.):
This is another often used projective test. The test consists of a series of pictures about which the subject is asked to tell stories. Some of these pictures deal with the ordinary day-to-day events while others may represent unusual situations.
The stories that subject tell constitute the basis for the investigator to draw certain inferences about their personality, tensions, adaptive aspects of behaviors and expressive aspects, etc. such inferences rest on the assumption that what the respondent perceives in the test-material represents in some way, externalization or projection of process within himself.
Tomkins-Horn Picture Arrangement Test:
This test has a somewhat more specific focus. It is designed for group-administration. It consists of 25 plates each containing three sketches that may be arranged in various ways to portray a sequence of events.
The subject is asked to arrange them in a sequence which he considers to be most reasonable. The responses are interpreted as providing evidence regarding conformity to norms, social orientation of the respondent, optimism-pessimism, etc.
Word-Association Test:
The subject is presented with a list of words; after each one, he has to respond with the first word that comes to his mind. For example, if the stimulus-word is “black” the subject may respond immediately by saying “white”, or someone may respond by saying “sheep” or “Negro.” The content as also the rate of the subject’s response may indicate the likely areas of emotional disturbance.
Sentence-completion Test:
In this test, the first few words of a possible sentence are given and the subject is asked to complete it. Different subjects may complete the sentence differently. This also affords clues to areas of the subject’s emotional disturbance and his attitudinal structure.
Doll Play Test:
This test is especially suited for children. The child-subject is given a set of dolls representing adults or children of both sexes or members of different ethnic groups. The subject may be asked to show how these dolls would act in specified circumstances.
The children may be allowed to play with the dolls freely. The manner in which they organize the dolls and work out various arrangements would indicate their attitudes or prejudices, etc., towards the class of persons represented by the dolls.
It is important to note that each of these tests, in addition to the specific function ascribable to each, also affords a basis for broader interpretation of the individual’s personality. Many of these techniques have been subjected to critical scrutiny.
Consequently, standardized methods of administration, scoring and interpretation have come to be established. Nevertheless, questions have been repeatedly raised about their validity, and evidence on this point to date is from conclusive.
Despite this limitation, the projective tests have been used with advantage in studies concerned with relation of individual personality to social and cultural factors.
For example, Adorno and colleagues in their study entitled, “The Authoritarian Personality” used T.A.T. pictures as one of the means for assessing the personalities of individuals who scored low on the scale of anti-Semitism and ethnocentrism, i.e., those who were less prejudiced.
With a view to identifying the relation between culture and personality, anthropologist Du Bois in her study of the people of Alor administered a Rorschach test to a number of villagers of Alor to see the correspondence a Rorschach test to a number of villagers of Alor to see the correspondence between the personality patterns and cultural behaviour.
The projective techniques have also been devised and utilized to investigate the content of an individual’s attitudes some social objects. These techniques share some of the characteristics of the projective methods already described. They encourage a free response on the part of the individual; they do not ask him to talk about himself directly or about his own views and feelings.
In so far as the purpose of these tests is to tap specific attitudes, the test-materials usually provide a more specific subject-matter than do those used in tests for assessing broad personality patterns. In these tests as in others ,the aim is to prevent the transparency of the test as best as possible. It is well worth noting that transparency of purpose (if masking is not feasible) is not necessarily a serious disadvantage.
Even when the purpose of a projective to the subjects, the projective tests are definitely preferable (if attempts at interpretation do not suffer owing to incompetence) for the following reasons:
(a) The subject may find it easier to express himself if he is not talking about his own feelings and attitudes explicitly)
(b) The subject may be unable to describe his feelings and attitudes as accurately as they may be discerned in the projective tests.
(c) It is possible that sometimes access to certain populations of potential subjects may be withheld if the topic under investigation is made explicit to subjects.
(d) The projective test may produce more extensive information than a questionnaire or even an interview, even if its purpose is not concealed to the subjects.
Quite a few projective techniques for the study of attitudes have also been devised. These very in the extent to which they can camouflage their purpose and requisition a certain measure of skill at recording and analyzing the responses.
5. Scaling Technique:
We know that social research necessitates that distinctions of degree rather than of kind be measured. The researcher may want to ascertain, for example, whether Mr. X is more favorably disposed toward an issue than Mr. Y. Although making such distinctions of degree is justly a function of analysis rather than of data collection, the desire to be able to identify such distinction influences the form in which data are collected.
This means that the questions asked to the respondents must be such as to give information on which judgements of degree can be based. Mostly the distinctions of degree are incorporated in the measuring instruments themselves.
Broadly, techniques for registering differences in degree are of two types. In the first type, one makes a judgement about some characteristic of an individual and places him directly on a scale defined in terms of that characteristic.
A scale is a continuum extending from the highest point (in terms of a characteristic, e.g., favourableness) agreement, etc. and the lowest point, i.e., the lowest degree in terms of the characteristic; there being several intermediate points between these two poles.
These scale positions are so related to each other that the second point indicates a higher degree in terms of a given characteristic as compared to the third.
In the technique of the (first) type we shall be presently considering how the rater places the individual on a rating scale set up in such a way that various degrees of the characteristic in question (for example, the favourable or unfavorable attitude toward co-education in colleges) are indicated.
The person making the judgement as to where to assign a particular response of the individual on the scale, may be the particular individual himself or an observer, an interviewer or a coder, etc. The second type of technique for registering differences of degree consists of questionnaires constructed in such a way that the score of individual’s responses assigns him a place on a scale.
For example, if the researcher is interested in an individual’s attitude toward co-education in colleges, the individual respondent is asked to reply to a series of questions relevant to co-education or to indicate his agreement or disagreement with a series of statements.
From his replies to these statements or questions, a score is computed; this score is taken as indicating his/her position on a scale representing various degree of favourableness or un-favourableness toward co-education.
The rating scales and attitude scales, both have the object of assigning individuals to positions with different numerical values in order to make possible the distinctions of degree. Let us now consider some of the major types of rating scales in which the rater places the person or object being rated at some point along the continuum, a numerical value being assigned to each point.
Graphic Rating Scales:
This perhaps is the most widely used rating scale. In this type, the rater (who may be the subject himself) indicates his rating by simply making a mark (✓) at the appropriate point on a line of statements that run from one extreme of the attribute or characteristic in question to the other extreme.
Scale-points with brief descriptions may be indicated along the line, their function being to help the rater in localizing his rating. The following scale may illustrate a graphic rating scale. Let us say the characteristic we wish to ascertain is peoples’ views in regard to participation of workers in the management.
One of the major advantages of these scales is that they are relatively easy to use and provide scope for fine discriminations of degree. A reference must be made to certain precautions to be taken in designing and using them.
Statements so extreme that they are not likely to be used should be avoided. Secondly, descriptive statements should be ordered to correspond as closely as possible, to the numerical points on the scale.
Itemized Rating Scales:
These are also known as numerical scales. In this type, the rater selects one of a limited number of categories that are ordered in terms of their scale positions. Scales with five or seven categories have generally been employed but some have used even as many as eleven points.
Barker, Dembo and Lewin in their study of the effects of frustration on constructiveness of play among young children constructed a seven point scale for rating constructiveness. They drew specific illustrations of points on the scale indicating degree constructiveness.
In the above study, “The toys are examined superficially”, the fourth point indicating moderate manipulation of the toys, and the seventh point indicating the highest degree of constructiveness was “play showing more than usual originality.”
In general, the more clearly defined the categories, the more reliable the ratings are likely to be. Of course, how much specification is needed depends on the fineness of distinctions warranted by the purpose of the study and, the nature of the material, etc.
Comparative Rating Scales:
In this category of rating scales, the positions on the scale are expressly defined in terms of a given population, a group or in terms of people with known characteristic.
The rater/respondent, for example, may be called upon to indicate whether an individual’s problem-solving skill or some other attribute most closely resembles that of Mr. X or Mr. Y or of Mr. Z, etc., all of whom may be known to him (the rater), in the matter of skill or attribute.
Or again, a rater may be asked to estimate the ability of an individual to do a certain kind of work in the context of the ability to the total group of persons engaged in the above kind of work and whom the rater has known. The rater then may indicate whether the individual is more capable than 10% of them or 209c of them, etc.
Rank Order Scale:
Here the rater is required to rank subjects/persons specifically in relation to one another. He indicates which person rates the highest in items of the characteristic being measured, which person is next highest and so on.
In the rating scales, the rater himself may be the subject to be rated. This is called self-rating. Self rating has certain typical advantages. The individual (rater himself) is often in a better position to observe and report his feelings, opinions, etc. than anyone else is.
But if the individual is not aware, as is not unusual, of his biases, beliefs or feelings or is aware of such feelings but does not wish to express them for certain reasons (such as fear or image conversation) then self-rating procedure may prove to be of little value.
It must be reckoned that an individual’s conception of what constitutes a particular position, say extreme position, may be quite different from that of the others making comparable self-rating.
Despite these failings, self-rating has proved useful in measurement of attitudes. With respect to certain attributes or attitudes, e.g., intensity, importance, etc., self-rating has come to be reckoned as the only satisfactory source of information.
Clear specifications of the dimensions to be rated and definition of the frame of reference or standard against which ratings are to be made, may reduce the possibilities of distortion in self-ratings.
6. Sampling Method:
We shall be addressing ourselves to an important problem concerning the practical formulation of social research. This problem relates to the estimation of certain characteristics of a ‘universe’ or ‘population’ on the basis of a study of the characteristics of a portion (or a sample) of it.
The method consisting of the selecting for study, a portion of the ‘universe’ with a view to drawing conclusions about the ‘universe’ or ‘population’ is known as sampling. Sampling, however, is not typical of sciences only. In a way, we frequently practice in our day-to-day lives some crude versions of sampling.
The housewives, for example, press a few pods of boiled rice in the cooking pot to be able to declare that it is ready to be served. Understandably it is not feasible to examine each and every grain in the pot, and more important, doing so is not necessary either.
Our day-to-day experience testifies to the fact that, by and large, it is possible to make some kind of a general statement about the ‘universe’ by observing only a few items or elements, i.e., a sample drawn therefrom.
Statistical sampling thus, is only a methodological version of our everyday experience and quite commonly employed procedure.
A statistical sample ideally purports to be a miniature model or replica of the collectivity or the ‘population’ constituted of all the items that the study should principally encompass, that is, the items which potentially hold promise of affording information relevant to the purpose of a given research.
A.L. Bowley, whose pioneering work on sampling statistics in the realm of social sciences won him academic and official recognition in the early twenties of the last century, arrived at certain conclusions about the ‘universe’ of his study by resorting to the method of sampling. Bowley sampled out for his study, one family for each group of twenty families.
His conclusions, based on the sample were found to be to a considerable extent consistent with the subsequent findings of Charies Booth and B.S. Rowntree who worked on a much more comprehensive canvas. Bowley’s work demonstrated very clearly that the sampling technique effecting, as it typically did, considerable economies of time, money and effort, also afforded worthwhile conclusions.
The use of sampling in social sciences has steadily increased ever since. During the last few decades, sampling techniques have assumed great importance.
A sample is a portion, selected from the ‘Population’ or ‘Universe.’ The terms ‘Population’ and ‘Universe’ have been used here in a very specific sense. ‘Population’ is not necessarily synonymous with a population of a community or a state.
‘Population’ in the manner it is used in sampling statistics is constituted of all the individuals, things, events, documents or observations (on a single or many individuals) etc., belonging to a designated category characterizing specific attributes which a particular study should principally cover.
The ‘population’ or the ‘universe’ of a study dealing, for instance, with ‘the views of college students of city about co-education’, will consist of all the students studying in various classes in the colleges of the city.
A ‘population’ contains ‘sub-populations.’ Thus, the female college students in the city form a ‘sub-population’ or a stratum of the ‘population’ consisting of the entire college students of the city.
A sub-population or a stratum may be defined by one or more specifications that divide a ‘population’ into mutually exclusive sections or strata consisting of (a) colleges, and (b) girl students in women’s colleges and male students from colleges meant for males only. A single unit or member of the ‘population’ is referred to as a population element.
It is well to take heed of an important issue raised by J.L. Simon. Sample for him is a collection of observations for which one has data with which he/she is going to work. Almost any set of observations for which one has data constitutes a sample.
Every sample principally corresponds to a ‘Population’ or ‘Universe’ behind it. But ‘Universe’ is typically harder to define because it is often an imaginary concept. A universe may be said to be a collection of things or people that one would like to say, his sample was selected from.
A universe can be finite or infinite and ill-defined. Infinite universes are harder to understand and it is often difficult to decide which universe is appropriate for a given purpose.
For instance, if we are interested in studying a sample of homicides, the issue to decide or settle is which universe the sample comes from. Depending on our objectives, the appropriate universe might be all homicides now living or it might be all homicides who might ever live.
The latter concept of homicides is imaginary since some of the items in the universe do not exist. It is infinite too. Those not agreeing with this notion of universe, would be inclined to regard it not as the collection of people/objects that they would say the sample was drawn from, but the collection from which the sample was actually drawn.
This view equates the universe to the sampling frame which is an empirical representation of the theoretic universe in which one is interested. The sampling frame is always finite and existential. The former notion of the universe is pragmatic.
A ‘census’ refers to a count or a study of all of the elements in the ‘population.’ As is quite obvious it is generally more economical in time, efforts and money to get the desired information for only some of the elements (sample) than for all of them, i.e., the population.
When we select some of elements (sample) with the intention of finding out something about the ‘population’ from which they are taken, we refer to that sub-group of elements as a ‘sample.’ Our expectation, in studying the sample, is of course, that what we find out from the sample, will be true of the ‘population’ as a whole. Actually this may not be the case, since after all, a sample is only a part of the ‘population.’
How far would the information or finding we get from the sample, approximate the finding we would get if the totality, i.e., the given ‘population’, were studied and whether or not our finding based on the study of a sample is likely to differ from the finding that we would get if the given ‘population’ as a whole was studied, by more than a specific margin, would depend greatly on the way sample is selected.
Of course, we can never have the full assurance that our sample returns reflect the state of the ‘population’ with respect to the characteristics we are studying, unless we have simultaneously conducted a complete comparable study of the ‘population’ (in which case the very purpose of and gains accruing from sampling would be nullified).
We can, however, devise sampling plans which if properly executed, can guarantee that if we were to repeat a study on a number of different samples, each of the same size, drawn from the given ‘population’, our findings would not differ from the true findings which we would get if the given ‘population’ as a whole were studied, by more than a specified value in at least a specified proportion of samples drawn from the population.
That is, it is possible to devise a sampling plan about which we can have a good measure of confidence that the findings based on our sample of a given size drawn from a given ‘population’ will not differ or deviate from the ‘true’ finding, i.e., the ‘population’ finding by more than a certain value, so that a tolerably reliable picture of the state of affairs in the population can be had from the sample-findings.
In actual practice, however, we do not go on repeating the study, i.e., go on recording responses or measurements for the same set of items on an indefinite number of samples drawn from the given ‘population.’
But the mathematical knowledge of what would happen in repeated studies on these samples, enables us to infer that with a given sample there is a probability that a certain proportion of estimates based on samples drawn from a population will be close to the population value, i.e., true value (i.e., will not deviate far from this value) and thus give out a reasonably good or dependable estimate of the population value which is the true value.
For a researcher who decides to study a sample with the intention, naturally of arriving at a reliable estimate about the ‘population’, it is very important that he should be able to say with a substantial measure of confidence that his sample-finding/estimate closely ‘approximates the ‘true’, i.e., population finding; otherwise studying a sample will have no meaning.
A sample is studied with a view to drawing conclusions about the ‘population’ or ‘universe’ that the sample is assumed to represent.
Thus, the measure of confidence that the researcher would like to place in his sample findings must be ‘substantial.’ This means that the probability of the sample finding being a reliable indicator of the ‘true’ finding, i.e., finding that would have been arrived at, if the ‘population’ in its entirety were investigated, must be quite high.
A sampling plan for a study is devised largely taking into view the level of accuracy and confidence in the findings of the study. Research projects differ in regard to the levels of aspiration for accuracy of and confidence in their findings (based on study of a sample.
A sampling plan which warrants the insurance that the chances are great enough that the selected sample is sufficiently representative of the population to justify our running the risk of taking it as a basis for estimating the characteristics (of researcher’s concern) in the population, may be called representative sampling plan.
Representative sampling plan is one major strategy employed by scientists to decrease the likelihood of misleading findings.
In social sciences, the measure or level of confidence is conventionally fixed at 95 (i.e., 95 Quite obviously, no purpose will be served by fixing the level of confidence at .5 50 r since this would simply amount to saying that there are 50% chances that the sample finding will be a very close approximation of the ‘true’, i.e., population value and again, that there are 50% chances that the sample value will not be a good estimate of the ‘true’ value.
It is, like saying that there is one chance in two that it will rain and also the same chance that it will not rain. Such an equivocal statement does not have any worthwhile import, for it is so pointless.
On the contrary, a 95% level of confidence with regard to the sample would lend us assurance that one can safely assume that the sample value will most probably afford a good estimate of the ‘true’ (population) value; since, 95% level of confidence would mean that the researcher has at this probability or confidence level the assurance that there are 95 chances out of 100 that his sample- finding will be a close estimate of the true finding and conversely, the odds are 5 against 100 that his sample-finding will be a bad estimate of the population findings.
There is another way of looking at this, too. Suppose 100 samples, each of the same size as the one actually selected by the researcher were drawn from a ‘population’, then 95% level of confidence or probability would mean that out of these 100 samples, 95 samples will be good estimates of the ‘population’ while only the remaining 5 samples will be bad or foul estimates of the population.
Thus, the researcher aiming at 95% level of confidence carries a great deal of assurance that the sample selected will give out findings which will represent the state of affairs (in regard to his specific concerns) in the ‘population.’
The underlying assumption is, of course, that the researcher’s sample belongs to the category of the 95% good samples and not to the 5% bad samples. The researcher’s sample happening to belong to the category of the 5% bad samples is a possibility which, however rare, cannot be overlooked.
It is helpful to understand in a general way the general advantages and limitations of sampling:
(1) Obviously, a sample can afford an estimate of the characteristics of the population in a much shorter time than would be possible otherwise. This time- saving advantage is especially important for studies of our modern dynamic society which is characterized by rapid changes.
Unless shortcut methods, e.g., sampling strategies, are devised for measuring social situations, the measurement is out of date before the study on the ‘population’ is conducted.
(2) Sampling makes the study much less expensive. Fewer people need to be interviewed. A smaller staff is required to collect, process and tabulate the data. Money saved by sampling procedure may be used to dig out more details about the cases under study and to intensify analysis of data.
From the administrative point of view, it is often impossible to conduct a study of the total ‘population.’ The typical difficulties in such a case relate to hiring of a large staff, the task of training and supervising them, etc.
(3) When small samples are used, it becomes possible to give more attention to each return that is received and to check their accuracy. This contributes significantly to the trustworthiness of tabulations and analysis.
It should be remembered, as was suggested earlier, that sampling in some sense is always employed in all studies, since it is clearly impossible to study all the manifestations of phenomena for all times and places.
It is well worth noting that even the census is but a sample of the country’s population at a given point of time. No sooner it is taken than it is a sample of the past. Hence, quite so often, there is no alternative to sampling.
Sampling, however, is not without its limitations. Here we may point out the major ones. Sampling demands exercise of great care and caution, otherwise the results obtained may be incorrect or misleading.
When the characteristics to be measured occur only rarely in the population, a very large sample is required to yield cases that will give statistically reliable information about it. Often, small samples hamper analysis of data since there are not enough cases for breakdown tables and sub-classifications.
We may do well to note that complicated sampling plans may in the long run require more than a complete ‘population’ count. This is particularly true if the sample is a large proportion of the total population and/or if complex weighing procedures are used.
7. Case Study Method:
The method of exploring and analyzing the life of a social unit/entity, be it a role-incumbent (person), a family, an institution or a community, is customarily known as case study method. The aim of case study method is to locate or identify the factors that account for the behaviour patterns of a given unit, and its relationship with the environment.
The case data are always gathered with a view to tracing the natural history of the social unit, and its relationship with the social factors and forces operative and involved in its surrounding milieu. In sum, the social researcher tries, by means of the case study method, to understand the complex of factors that are working within a social unit as an integrated totality.
Looked at from another angle, the case study serves the purpose similar to the clue- providing function of expert opinion. It is most appropriate when one is trying to find clues and ideas for further research.
Burgress has highlighted the special potency of the case materials for understanding complex behaviour and situations in specific detail. He refers to these data as a social microscope. The major credit for introducing case study method to the field of social investigation must go to Fredrick Leplay.
The English social philosopher, Herbert Spencer, was among the first to use case materials in his comparative studies of different cultures. William Healey, resorted to the case study method in his study of juvenile delinquency.
Healey realized that the problem of juvenile delinquency was too complex to be understood simply on the basis of available statistical data. Hence, he declared himself in favour of the case study method which afforded a deeper and rounded understanding of the phenomenon.
Anthropologists and ethnologists interested in the systematic description and study of the primitive as well as modern cultures have liberally utilized the case study method.
Cora Dubois, Robert Redfield and Oscar Lewis, to mention some of the prominent names, have liberally employed the case study method. Historians have resorted to this method for portraying some historical character or a particular historical period and describing the developments within a national community.
Many a novelist and dramatist has used some semblance of the case study method for presenting a word picture of characters.
The case study method received the necessary impetus and recognition as a systematic field-research technique in sociology with the well-known study, “The Polish Peasant” by Thomas and Znaniecki. In the course of this study, made extensive use of life history documents and made them their chief instrument in reaching out to the actual experiences and attitudes of individuals and groups as well as in securing “a cross-section of the entire process of their social becoming.”
They scrutinized a large number of personal diaries, letters, autobiographies and other types of case materials with a view to getting at the concrete details of the individual and collective behaviour of persons in a given cultural context.
Thomas and Znaniecki aimed at reconstructing a chronologically continuous and complete word-picture of the feelings of individuals subjected to particular experiences, of their ideas about the relations they have with others and the impact of these on them.
Thomas and Znaniecki maintain that the case data constitute ‘the perfect type of sociological material’ in so far as they represent a more enlightening and fundamentally more reliable record of personal experiences, with a wealth of concrete detail, vivid memories, tension situations, and multifarious reactions to social situations which would escape the attention of the most skilled investigators using other techniques.
Thomas and Znaniecki contend that the social science has to resort to the use of data other than the case-history or life history simply because of the practical difficulty in securing, at the moment, a sufficient number of such records encompassing the totality of sociological problems and the enormous amount of work involved in an adequate analysis of all the personal data necessary to fully characterize the life of a social group.
In India, quite a few monographs on rural and tribal communities have resorted to the case study method.
Social scientists ultimately aim at some kind of generalization or theory-building. Whether the case data can be regarded as sufficiently typical or representative affording a secure basis for theory-construction is a question that has been plaguing the social scientists for quite some time.
The issue has been a subject of controversy among the social researchers. Hence, it is important to thrash out the question whether the materials offered by case history may be considered an adequate basis for generalizing with respect to the category of cases that the particular case under study represents.
Stouffer, Kinsey and Adorno among others, have had occasion to study a large number of cases. These social scientists found a remarkable uniformity among independently conducted studies of large groupings in different socio-cultural and temporal contexts.
The opinions of Stouffer, Kinsey etc., in regard to the reasonably high generalizability afforded by case data are in tune with those of Thomas and Znaniecki. Anthropologist Franz Boas too, on the basis of his several case studies of the ‘primitive’ groups, came to the conclusion that human nature anywhere is of one piece.
But, the fact of uniformity among cases does not warrant the conclusion that the cases under study are the typical representatives of the larger category of cases they were drawn from. It is indeed improper to overstress the element of uniformity since the similarity among cases hardly, if ever, extends to all the dimensions there are to life.
While human behaviour may vary according to situations, it is usually possible to identify the ‘basic’ human nature in the midst of such variations. This is the assumption that underlies collection of case data. All human beings experience certain physiological tension; certain experiences are ubiquitous, e.g., birth, death, sex drive, fatigue, etc.
As Dubois an anthropologist, rightly points out, that the comparative studies of personalities as determined by variations in culture are possible simply because of certain basic homogeneity or similarity evidenced in mankind.
Psychologist G.W. Allport contends that some statements about human nature broadly apply to each individual or to each member of a larger group. As such there does not appear to be any reason why a quest for identifying the innate human tendencies cannot capitalize on personal case data.
Various media and techniques have been utilized by researchers in the course of certain excellent case studies, they managed to turn out. Nels Anderson, who conducted a case study of the ‘Hobos’, got to know about their inner lives through the medium of their poetry, folk-songs, ballads and other cultural manifestations. Anderson collected their photographs published in journals and newspapers.
He also collected from several institutions the statistical and other types of information about life of the Hobos, collecting relevant bits of information from such diverse sources, Anderson could offer a systematic account of the inner life of the Hobos and the practical ethics of their organization. Warner and associates have, in the course of their case studies entitled.
The Yankee City Series, made use of various methods and techniques of data collection. Personal interview, observation, questionnaire, statistical records, etc., were the diverse means employed by them. In sum, various researchers have employed a number of different means and techniques to get at data substantiating, supplementing and verifying the information gained through the case study method.
The specific method of case study would depend upon wit, commonsense and imagination of the person doing the case study. The investigator makes up his procedure as he goes along. Saturating oneself in the situations is very important.
Some anthropologists believe that case studies of less than several year’s duration are likely to be misleadingly superficial. Bronislaw Malinowski, a doyen among anthropologists, gives a vivid argument on this point.
“Living in village with no other business but to follow native life, one sees the customs, ceremonies and traditions over and over again, one has examples of their beliefs as they are actually lived through and full body and blood of actual native life fills out soon the skeleton of abstract constructions. That is the reason why working under such conditions, the ethnographer is enabled to add something essential to the bare outline of tribal constitution, and to supplement it by all the details of behaviour, setting and small incident.”