42969 (Research methods to collect primary empirical information)

MINISTERY OF EDUCATION OF THE REPUBLIC OF BELARUS

Belarus State Economic University

REFERAT:

”Research methods to collect primary empirical information”

Minsk 2008

Research methods are strategies or techniques to conduct a systematic research. To collect primary data four main methods are used: survey, observation, document analysis and experiment, but not any of them is adequate or best for all purposes – all should be supplemented and checked.

The most wide-spread method that provides almost 90% of empiric data is survey research. Survey is a poll in which researchers gather facts applying to respondents whose verbal statements are a source of information. Respondents are people who provide data for analysis.

The strengths of surveys are seen in the following: survey research is useful in describing the characteristics of a large population without having to interview each person in that population. It is useful in analyzing social change or documenting the existence of a social problem. It is cheap to organize and makes a maximum use of technical devices to process the obtained data.

A weakness of survey research is that quality of obtained data may be affected by a respondent’s personality – his education, culture, memory, attitudes to the study problem, on the one hand, and the researcher’s personality – his professional level, communicative skills etc., on the other hand.

Survey data are collected by using such methods as questionnaire, interviewing, sociometric survey and expert survey.

Questionnaire is a popular method of data collection with a questionnaire form as a printed research instrument containing a series of items for respondents to answer. It may be self-administered by a respondent or administered by an interviewer in face-to-face encounter or by telephone.

The advantages of questionnaire are well-known: data can be collected fast on specific items; these data can be easily transferred into forms allowing quantified and computerized analyses; the task of data collection can be delegated to less expensive field staff.

However, several problems can arise when using questionnaire. It can impose a rigid, preconceived idea of reality which may be inappropriate for the particular situation. If field enumerators are not supervised properly, errors in recording data can occur. Problems may also arise from respondents concealing, misreporting, or misunderstanding the questions.

The design and preparation of a questionnaire form are extremely important, as they will influence the type of information collected, in somewhat the same way as the mesh-size of a fish-net determines the fish that are caught.

First, the questionnaire form must be strictly designed: it begins with an introduction that should make a respondent interested in participating in the poll; so it informs him about the research aim, way of filling in, guarantees of anonymity etc. The second section is a set of pre-arranged questions. The final part contains demographic data about respondents.

Selecting and phrasing of particular questions is of utmost importance in any survey as their purpose is to discover what people know, not what they do not know. They should follow in a logical order so that the researcher can obtain maximum information, and people should be reminded of aspects on which they might comment. Sometimes it is good to start with a general question, “What do you think about X? ”followed by specific questions. Questions must be carefully phrased so that they can be understood by any respondents who belong to a certain socio-demographic group. They should fit with indigenous knowledge systems, and with local perceptions. The researcher must have enough basic knowledge of the community to know which questions would be meaningful, and how exactly they should be framed so as to minimize the possibility of creating ambiguity or embarrassment.

Besides, the questionnaire form should meet the requirements of validity and reliability. Validity is the extent to which a study or research instrument accurately measures what it is supposed to measure. Reliability is the extent to which a study or research instrument yields consistent results when applied to different individuals at one time or to the same individuals over time.

In questionnaire forms the following types of questions can be used:

• open-ended questions – when a respondent himself formulates the answer, for example, “How did you spend tonight? ”– “I went to the cinema”or “I chatted in the Internet”;

• closed questions – when a respondent is provided with some alternatives, for example, “What do you think of smoking? ”– “It’s bad for health”, “It’s a way of coming down under stress”or “I’m neutral about it”. One and the same question can be made open-ended or closed. Closed questions are easier to computerize, but they need the researcher’s comprehensive knowledge on the issue. Open-ended questions are used when this knowledge is limited;

• semi-closed questions – when a respondent is provided with alternatives and given a chance to express his own opinion on the issue. Normally it’s included as “other”followed by a space for a respondent’s comments;

• scale questions – when a respondent checks a scale (of incidence, preference, or quantity) of 0-5 (1-10 etc);

• menu questions – when a respondent can choose any combination of answers;

• alternative questions supposing to choose only “yes”or “no”answers.

Although the questionnaire ought to cover all questions needed, it should be neither too elaborate nor too long. The number of questions should vary from 25 to 30 as a bigger number creates more accidental and inadequate information because a respondent becomes tired. Consequently, an hour is usually a maximum time period for filling in any questionnaire. Besides, most people have a lot of demands on their time, so they cannot spend too much time on answering questions. Whether to use a closed form (with itemized answers) or an open-ended form questionnaire depends on the researcher’s own needs and requirements.

Where computer facilities are available, it is advisable to frame and code questionnaires so that computer analysis is possible. When the totality of samples is big, such as in a national or other macro-level survey, the use of computers is almost essential. By computer, we refer not only to macro-computers but also PCs, and even some calculators. Manual analysis often can be done quickly and cheaply so that preliminary results are obtained in a few days, instead of waiting at the mercy of the computer for months. When computers are used, a member of the computer staff is recommended to be part of the research team.

Interviewing is a data-collection encounter in which an interviewer asks the respondent questions and records the answers. It is a personal contact between a respondent and an interviewer that differentiates interviewing from questionnaire.

A short-cut method to gather data fast is to interview groups rather than individuals. In applying this method, a problem of representativeness may arise, since any chosen group is unlikely to represent a true cross-section of the local population, although attempts to include individuals of different socio-economic status should be made. The knowledge and experience of several individuals may serve as checks on information given by each others. There is nothing specialized about a group interview as compared to a person interview.

In an interview, it is usual to have a set of questions to ask, and in most circumstances an open-ended type of interview is advised because it allows the conversation to be directed to some extent by the respondent. Closed-forms or standardized interviews, consisting of pre-arranged questions and answers, are seldom the best approach, unless the interviewer already has both extensive, accurate and up-to-date knowledge of the community in general and the energy system. In general, a non-standardized format that allows flexibility is best. The interviewer asks questions from a standardized questionnaire and his task is to record the respondent’s answers in the exact way. Anyway, this kind of interview is complex for both the respondent as it takes a lot of time to think over the question and formulate the answer, and the interviewer who spends a lot of time and efforts to record the answers. Several problems can also arise in processing and decoding the obtained data. That’s why non-standardized interviews are not often used in sociological research, although in some cases they are of great value as they provide most complete, comprehensive, informative answers in a widest range. A semi-standardized interview combines peculiarities of both interviews spoken about.

Sociometric survey is a survey form used in small social groups to discover interpersonal relations between group members by fixing preferences, likings, dislikings etc. The sociometric technique is based on qualitative criteria to measure emotional relations and character of interactions between group members, each member’s status, non-formal group leaders etc. Respondents are asked questions of the type, “Which of your group will you choose for …? ”or “Which of your group will choose you for …? ”The criteria can be formal or connected with organizing joint activities and non-formal or connected with emotional, interpersonal relations, entertainments, leisure time etc.

When members of a group are asked to choose others in the group, everyone in the group makes a choice and describes why he does so. From these choices a description (a drawing, like a map) called a sociogram emerges. The following is a sample of sociogram:

Data processing is supposed to build up various matrices which present results in the form of a matrix or table, calculating coefficients as far as the group’s emotional solidarity is concerned etc. Such table is called a sociomatrix:

Expert survey is a survey form conducted to have expertise on the issue. Its distinction is that its respondents are people who are competent in a definite area of knowledge or practice. That’s why expert survey can’t provide any anonymity.