Survey in Research Survey in Research

Survey in Research

survey-in-research

Survey in Research

The word survey has been derived from the words 'Sur' or 'Sor' and 'Veeir' or 'Veoir', which mean 'over' and 'see' respectively. Therefore, the literal meaning of a survey is to take over something from a high place but it has come to be used as a term with the specific connotation of its own. Although the term is used for the technique of investigation by direct observation of phenomena or collection of information through interviews, questionnaires, etc ., it is currently being used in those investigations also where published data is used, e.g ., a survey of joint-stock enterprise in Nepal. Some writers still believe the term survey can be applicable only when direct contact is made, and does not include any study from the libraries or archives.

General Components of Survey Research

A survey often begins by identifying several individuals considered representative of the group to be studied (what is referred to as a sample) and deciding what questions they should be asked. The survey can be used to test accepted explanations or theories, and to develop new ones. They can lend support to a theory by indicating that a relationship that is postulated actually occurs among the group studied, or they can throw doubt on the applicability of a theory by offering little evidence that the relationships suggested by the theory actually exist. If the sample is statistically representative of a larger population, then the findings can offer greater support, or greater denial, of the theory.

1. Modes of eliciting information
There are two primary modes of doing a survey: using questionnaires or giving interviews. Both methods are based on a set of questions. In the questionnaire, these questions are written down and the respondent reads them and gives written answers. In an interview, the interviewer asks the questions as they are written in an interview schedule and then records the respondent's answers either by writing them down or recording them electronically. Interviews may be face to face, or they may be carried out on the telephone.

2 . Modes of selecting respondents
The first consideration in choosing among them is whether a potential set of respondents will be able to give answers to the types of questions to be asked. The relevant issue here is, what is the appropriate population to which we questions apply? If we are doing a study of farm abandonment, we will probably want to survey farmers, exfarmers, and others who live in rural areas.

Second, it will be important to design a survey, which will be appropriate for this sample. It should include questions, which such a group could and would answer (Fink, 1995a). That is the questions must be presented in language familiar to the sample and phrased so those members of the group will understand them.

3. Modes of returning information
Once a questionnaire is completed, there must be clear instructions as to how it is to be returned. In most mail surveys, return self-addressed envelopes-which are usually stamped-are included with the questionnaires. In this case, the questionnaire needs to contain very few instructions about its return. Interviews, once terminated, need to be fully converted into information, which may be processed as a part of the study.

Characteristics of Survey Research

Survey research is but not one of many research tools available to social researchers. It bears repeating that survey method are not appropriate to many research topics nor do they necessarily provide the best approach to topics to which they might be reasonably applied. Nevertheless, survey research can be used profitably in the examination of many social topics and can be especially effective when combined with other methods. Survey research provides the best teaching example for instruction in social science methodology.

1. Logical
Survey research is guided by all the logical constraints. In practice, moreover, survey data facilitate the careful implementation of logical understanding. While this topic will be explored in much greater detail.

2. Deterministic
Whenever the survey researcher attempts to explain the reasons for and sources of observed events, characteristics, and correlation, he must assume a deterministic posture. And where the survey format permits a clear and rigorous elaboration of a logical model, this clarifies the deterministic system of cause and effect.

3. General
Sample surveys are rarely conducted for purposes of describing the particular sample under study. Rather, they are conducted for purposes of understanding the larger population from which the sample was initially selected.

4. Parsimonious
Because the survey researcher has a large number of variables at his disposal, he is in an excellent position to carefully examine the relative relevance of each. Like all scientists, he would like to obtain the greatest amount of understanding from the fewest number of variables.

5. Specific

In the context of this scientific characteristic, the crustacean-like nature of survey research is most important. Ironically, this characteristic also opens survey methods to the greatest amount of criticism.

Types of Surveys

Selecting the type of survey we are going to use is one of the most critical decisions in many social research contexts. A few simple rules will help us make the decision; we have to use our judgment to balance the advantages and disadvantages of different survey types. Surveys may be classified into various types according to their subject matter, the technique of data collection, regularity, etc.

1. General or specific surveys
When a survey is conducted for collecting general information about any population, institution, or phenomena without any particular object or hypothesis it is known as a general survey. The government for supplying regular data on many socioeconomic problems mostly undertake such surveys. Censusofpopulation every tenth year is a typical example of such a survey.

Specific surveys are conducted for specific problems or for testing the validity of some theory or hypothesis. These surveys are naturally more pointed and only such information as is directly related to the specific purpose is collected. Information gathered through these surveys is generally of very little value outside the problem under study.

2. Regular and ad-hoc survey
Some surveys are regular in nature and must be repeated after regular intervals. For such surveys, permanent machinery for collecting information has to be set up. Most of such surveys are mainly economic surveys. Economic Survey of Ministry of Finance, Industrial Survey, Agricultural Survey, Census Survey by Central Bureau of Statistics (CBS) are few examples of regular surveys in Nepal.

Ad-hoc surveys are undertaken one for all. They may also be conducted in phases if the area of investigation is large, the whole survey is finished in two or more installments. This would not entitle them to be termed as regular or repetitive type unless the same information is collected over and over again. Ad-hoc surveys are undertaken mostly for testing a hypothesis or supplementing some missing information regarding any research problem.

3. Preliminary and final surveys
A preliminary survey is generally known as a "pilot study" and it is the forerunner of the final survey. The purpose of this survey is to get first-hand knowledge of the universe to be surveyed. It helps a person to get acquainted with the problem and the nature of respondents from whom the information is to be collected. It is, therefore, very useful in preparing the schedule or questionnaire and organizing the survey on proper lines. The final survey is made after the pilot study has been completed.

4. Official and non-official survey
The government for finding certain facts for official as well as philanthropic purposes can conduct a survey. Such a survey is called an official survey. Individuals as well as social and research organizations with or without the help of the government also collect surveys. Such surveys are called nonofficial surveys.

5. Direct and indirect survey
Surveys have also been classified as direct and indirect surveys. In the former case, facts can be quantitatively interpreted whereas in the latter case that is not possible and results are to be concluded out of the available data by some indirect method.

6. Personal or postal surveys
A personal survey is one in which the surveyor himself is required to move about and collect information personally but on the other hand in the case of the postal survey, the surveyor gets information through the post.

7. Primary or secondary survey
A primary survey is one in which the surveyor starts work on a particular subject and personally collects all data and facts and figures. On the other hand in the case of secondary data, the surveyor himself does not start the work but depends on the work, which has already been done by someone else. Obviously, primary data is more important and dependable than secondary data (Fink, 1995c).Evo

8. Census and sample surveys
In a census survey, every single unit in the universe is to be contacted and information collected. In the case of a sample survey, only a small part of it is taken as representative of the whole, and data collected from it are made applicable to the whole universe. Sample surveys are becoming more popular these days because of their convenience, time-saving and low cost.

Objectives of a Social Survey

A survey may be organized simply by a need for administrative facts on some aspect of public life, or be designed to investigate some cause-effect relationship or to throw fresh light on some aspect of sociological theory. The surveys may thus be motivated by several objectives. The chief of them may be classified as follows.

1. Supply of information on any problem
The purpose of many surveys is simply to provide someone with information. That someone may be a government department wanting to know how much people spend on food, a business concern interested to find out what detergents people are using, a research institute studying the housing of old-age pensioners. Most of the surveys are, thus, utilitarian in nature and are meant to provide information regarding practical problems. The researcher himself may carry on a survey, but there are also specialized bodies that carry on such surveys for others.

2. Description of a phenomena
Surveys are also used for a detailed description of phenomena. To a social scientist, a survey may equally have a purely descriptive purpose as a way of studying social conditions, relationships and behavior. No description of phenomena would be accurate and complete unless we come face to face with it. Surveys help the researcher to come in direct contact with the phenomena under study and thus provide him with all the details that he needs. Socioeconomic surveys describing lining conditions of a geographical area may be cited as illustrations of this kind.

3. Explanation of a phenomena
This is the theoretical importance of the survey. Before we embark upon such surveys we must have a hypothesis to test or a problem upon which we require some specific information to establish some causal relationship. To illustrate the point let us suppose a hypothesis is formed that 'slums give rise to delinquency.' Now, this can be proved or disproved only by collecting data regarding delinquents both from the slum and non-slum areas.' Necessarily the information collected would be highly specific and purposive and not a general nature as in the case of description of a phenomenon.

A survey may be thus general or specific. It may be purely utilitarian in nature or many have academic importance aimed at verification of some established theory or any of its corollaries. It thus helps in the refinement and expansion of old theories and in the establishment of new ones. Even those theories that are based upon logical inferences of accepted principles have to stand the test of verification through surveys.

Please leave your comment

If this article has helped you, please leave a comment.

Previous Article Next Article
close