ADVERTISEMENTS:
After reading this article you will learn about:- 1. Types of Questionnaires 2. Kinds of Questionnaire Items 3. Designing the Questionnaire/Schedule 4. Physical Form 5. On Questionnaire Construction 6. Advantages 7. Limitations.
Types of Questionnaires:
The types of questionnaire vary widely. Questionnaires may be classified in a number of different bases. The classification of questionnaires may be based on the variable of structure.
Accordingly, we have:
ADVERTISEMENTS:
(a) structured/standardized questionnaire;
(b) unstructured/non structured questionnaire.
Structured questionnaires are those in which there are definite, concrete and preordained questions with additional questions limited to those necessary to clarify inadequate answers or to elicit more detailed responses. The questions are presented with exactly the same wording, and in the same order to all the respondents. The reason for standardization is to ensure that all the respondents are replying to the same set of questions.
The form of the question may be either closed (i.e., categorical) or open (i.e., inviting free response); the important point is that they are stated in advance, not constructed during the questioning. Standardized questionnaires may differ in the amount of structuring of the questions used.
ADVERTISEMENTS:
They may present fixed alternative answers to questions so the respondent just chooses the appropriate one, or they may leave the respondent free to answer in his own words. Extensive use of structured questionnaire is made, for example, in studies of the cost of living expenditure, and investment practices, etc.
The most famous of the structured questionnaires are all those by the Census Bureau for the population and Housing Censuses. Fixed Alternative Questions are those in which the responses of the subject are limited to the stated alternatives. These alternatives may be simply yes or no.
The following is an example of fixed-alternative question: To what social class would you say you belong:
(a) Middle class,
(b) Lower class,
(c) Working class, or
(d) Upper class?
Open-ended question are designed to permit a free response from the subject rather than one limited to certain stated alternatives. The distinguishing characteristic of open- ended questions is that they merely raise as issue but do not provide or suggest any structure for the response-dent’s reply.
The respondent is given the opportunity to answer in his own terms and in his own frame of reference. When open-ended questions are used in standardized interviews, the questions and their order are predetermined and the task of the interviewer is to encourage the respondent to talk freely and fully in response to the questions included in the interview schedule and to make a verbatim record of his replies.
Open-Ended and Closed Questions:
ADVERTISEMENTS:
Fixed alternative or closed questions have the advantage of being ‘standardizable’, simple to administer, quick and relatively inexpensive to analyse. The analysis of responses to open-ended questions is often complex, difficult and expensive.
Sometimes the provision of alternative replies helps to make clear the meaning of the question. Respondents are more likely to understand the question when the alternative replies are provided.
The function of alternative responses is to clarify the dimensions along which answers are sought. One of the principal values of open-ended question is its use as an exploratory tool before opinion has crystallized or before the research objectives have been clearly defined. The closed question may require the respondent himself to make a judgement about his attitude rather than leaving this to the interviewer or coder.
Most of these advantages of fixed alternative questions have their attendant limitations. One of the major drawbacks of the closed question is that it may force a statement of opinion. Many individuals do not have any clearly-formulated or crystallized opinions on many issues.
Closed questions are ill-equipped to reveal these. Although the wording of questions is the same for all respondents, different respondents are likely to make different interpretations, some of which may be quite different from those intended but it is much more likely to go undetected in the former. Closed questions are more efficient where the possible alternative replies are known, limited in number and clear-cut.
Thus, they are appropriate in securing factual information (e.g., age, home- ownership, income) and in eliciting opinions on matters on which people hold clear views. Open-ended questions are desirable when the issue is complex.
Closed questions have the advantages of focusing the respondent’s attention precisely on the dimensions with which the investigator is concerned.
But they do not provide information about respondent’s own formulation of the issue, the frame of reference in which he perceives it, the factors that are important for him and motivations that underlie his opinions. When these matters are the focus of interest, open-ended questions are justly warranted.
Kinds of Questionnaire Items:
The information sought by the questionnaire may be classified under three heads:
(1) Identificatory Information:
The following items may be included under this category:
Questionnaire case, cross-reference, questionnaire number, name of survey, name of agency sponsoring the survey, name of individual or family interviewed, sex of informant, relationship of informant to family head, address of the case, telephone numbers, interviewer’s name or initials, co-operation of the informant ant notes as to the confidential treatment of returns.
(2) Social Background and Factual Data:
The following items may be included under this category: Age of head of the family and family members, marital status, education, religion, political preference, union membership, family size and composition, occupation of the head of the family or of the respondent, employment, family income, socio-economic status, etc.
(3) Subject-Matter of the Survey:
The informant may be asked a direct question on the facts as he understands or remembers them. Certain information can be obtained rather easily by asking straightforward questions. The opinions which he holds are not so readily ascertained. First, is the so-called “information questions.’
When knowledge about the topic is definitely correlated with the opinions held, these information questions are particularly important. The opinions may be elicited only from those who are in a position to hold them.
A second approach is that of “seeking advice.” The informant is usually flattered by the fact that his advice is considered important. This approach has been used effectively in a study of factors associated with family size.
Another approach which opinion-pollers use is the “exploratory questions.” In questions of this type, the informant is given a sufficient background of information about the survey topic to make sensible judgement.
Designing the Questionnaire/Schedule:
Several considerations should be borne in mind while designing a schedule or questionnaire. Careful planning, about the physical design of the questions, careful selection and phrasing of the questions definitely affect the number of returns as also the meaning and accuracy of the findings.
(1) The physical appearance of the questionnaire affects the co-operation the researcher receives. In a mailed questionnaire, an attractive looking questionnaire is a plus point for co-operation. Conversely, an unattractive one may cause the recipient to put it aside.
(2) Schedules are usually filled by the investigator whereas the questionnaire generally by the respondent himself. This consideration as to who is to record the responses must go into designing a schedule / questionnaire. If a highly-trained investigator is to do the questioning and enter the replies, the form should be different from the one drawn up for the informant to fill out.
The terminology and questions should be adapted to the type of people who will give the information. For example, a questionnaire addressed to experts who are thoroughly conversant with the subject-matter of the survey can be much more technical than one directed to a random sample of the population.
In designing questionnaires that are to be filled out by people from different walks of life, the level of education, the biases, and the interest or other characteristics which affect the ability and desire to fill in the form truthfully and correctly should be taken into consideration.
(3) The choice of words is understandably an important consideration. The information must grasp the spirit of the question rather than its precise wording.
(4) In inquiries of certain aspects, it may be important for certain questions to follow certain others so that proper “set” is developed. Special attention must be given to the sequence of the items/questions.
(5) If the number of questions is small, their arrangement on the questionnaire will not require detailed planning. When the number of questions is large, they should be condensed in a very limited space.
(6) The purpose of the questions is another important consideration. It may be to ascertain facts, to test the ‘knowledge’ of the informant or discover his beliefs or attitudes. If opinions are desired, for example, care must be exercised to see that the questions do not just bring out only the points of fact.
(7) If the questionnaire is to be used for a periodic survey, the questions should be designed with a view to uniformity and comparability of results.
(8) Questions must be designed with the possible difficulties of analysis well in view. Analysis is facilitated to the extent the information is in. a form which is readily amenable to classification and tabulation.
Physical Form of the Questionnaire:
(1) Size:
The size of a questionnaire depends to some extent on the scope of the survey and the number of items to be included. The basic question is, what size of the questionnaire is preferable? This can be best answered by considering the advantages and disadvantages of various sizes.
If the schedules are small, they can be carried easily in pockers and brought out only after the respondent has answered the call. If the questions can be put on handy cards, sorting, counting, filing, checking, etc. in the office are facilitated. Investigators object to carrying a large folder because their identity may be mistaken. The questionnaire should not be too cumbersome for respondents.
It is best to use only one side of the form, the reverse being left blank for special remarks by the recorder. If both sides are to be filled out, filing and sorting the schedules become difficult. Mailed-questionnaires should be large enough to provide adequate space for comments if such supplementary material is desired.
Charts, diagrams and pictorial material may be introduced in main questionnaires to sustain and recreate interest on the part of the recipient, but such material usually requires considerable space. In order to avoid too large a schedule, a booklet may be used.
(2) Quality and Colour of Paper:
The schedule is handled a great deal after it reaches the office. Therefore, the paper should be durable. If sorting and counting are to be done by hand, a strong flexible card with a smooth surface is desirable. The less conspicuous the schedule, the less likely the objections to giving information. Ordinary white and light-coloured schedules are preferable from the collection point of view.
When planning mailed questionnaires, it may be desirable to use a colour which will attract the recipient’s attention. In certain marketing studies, yellow paper was found to have the highest percentage of returns but dark colours were not effective. When several questionnaires are sent out in a sequence, alternation of colours elicits more returns than a single colour.
(3) Arrangement of Items on the Questionnaire:
Questions which belong together should be arranged so. When the question is dependent upon the answer to the preceding one, it should be given a subordinate place. Paying due attention to the appearance of the questionnaire will avoid many errors and ensure a higher percentage of usable returns than would be obtained from a poorly-arranged questionnaire.
Choice of Questions:
(1) The researcher should include only such questions as have a direct bearing on the problem itself or on the evaluation of the methodology adopted for study.
(2) Questions whose answers can be secured more accurately easily and effectively from other source may be excluded.
(3) The selection of questions should be done with an eye on the subsequent tabulation plan.
(4) In drawing up the schedule or questionnaire, other studies or surveys on comparable material should be kept in mind. As far as feasible, identical items, terms, definitions, and quantitative units of measurement should be employed.
(5) Care should be taken while asking personal questions or those which may embarrass the respondent.
(6) Only such factual questions, to which most of the informants can reasonably be expected to know, should be asked. Frequently, duration, vividness, interest, meaningfulness and setting are some of the important factors that help securing the required information.
(7) Questions that are likely to yield inaccurate responses should be avoided. People often resort to a form of wishful thinking in answering questions on such matters as the level of education they have had, the jobs they have had, the jobs they have held, etc.
(8) Questions which involve too much mental effort on the part of the informant should be avoided, e.g., entailing mathematical calculations.
Wording of Questions:
Understandably great care is required in formulating the questions. Reliable and meaningful returns depend to a large on this. Even when factual information is to be secured, certain precautions relating to phrasing of questions are necessary. Even greater care must be exercised when opinions are to be obtained.
Since words are likely to affect responses, respondents with limited vocabularies are likely to be suggestive. If questions are beyond the comprehension of the respondent, he may just choose one of the alternative responses without any idea as to the meaning of his response.
Suggestions for Wording Questions:
(1) Simple words which are expectedly familiar to all potential informants should be employed. This should be done without making the questions appear too elementary for those of higher mental or educational level. This is often possible and it is here that art of question-framing comes in. A question that contains long, dependent or conditional clauses may confuse the informant.
(2) Formulate the questions so specifically as to yield the exact information that is needed. The more specific the question, the greater the usability of the answer to it for tabulation purposes.
(3) Avoid multiple-meaning questions. Such items that will give rise to confusion should be formulated as two or more questions.
(4) The researcher should not make use of ambiguous questions. Ambiguity may arise if terms are beyond the vocabulary of the respondent. If the phrasing is too complex or in general terms, the informants may start with entirely different assumptions. The best means to this end is to pretest the questions.
(5) All questions which produce biased answers may be regarded as ‘leading questions.’ Questions worded in such a way as to suggest the answers may be avoided. For example, to the question, “Do you regularly visit the library?” many people will answer “yes”, even though they may not be doing so.
(6) Answers to questions in which prominent personalities are mentioned will be conditioned by the informant’s personal feeling towards them. Indiscriminate ban on important names is not, however, desirable.
(7) Danger words, catch-words, stereotypes or words with emotional connotations should be avoided. Names of political parties and political figures may colour the responses, hence such use may be avoided.
(8) Chapin suggests the occasional use of “slant side-questions.” For example, when respondents were asked, “Are you married?” most of them replied in the negative but when the question was changed to “Where is your wife?”, a much higher percentage was discovered to be married.
(9) Most people like to feel that they are reasonable, intelligent, penerous, understanding and prestigious members of their community. They thus tend to answer questions in terms of what they “ought” to think or feel about a situation. Hence, caution must be exercised in the use of phrases which reflect upon the prestige of the informant.
(10) The question must allow for all possible responses. Thus, provision for such indefinite answers as “don’t know”, “no choice”, “doubtful”, other (specify) should be made.
(11) The alternatives in multiple-choice questions must be realistic and not far fetched. They should conform more or less to the way people really think and feel about the issues involved. If the alternatives are phrased in terms of concrete situations, the questions will be more meaningful.
When framing the alternatives, it should be borne in mind that people answer questions in terms of the relative value of the choices presented in the questions and not in terms of absolute universe of preferences.
(12) The amount of writing required on the schedule or questionnaire should be reduced to the minimum. Since most handwriting is poor, there is the danger of mis-interpretational errors. Whenever feasible, symbols may be used for replies.
(13) A few questions that will serve as checks on the accuracy and consistency of questions as a whole should be designed. Questions that bring out the same facts but are worded differently and placed in different sections of the questionnaire afford a check on the “internal consistency” of the. replies.
(14) Questions inviting responses toward socially accepted norms or values should be avoided. Such questions often fail to indicate a person’s real opinion.
(15) Seemingly unreasonable questions should be justified by using an explanation why they are worth asking.
On Questionnaire Construction:
The investigator should fist discover the extent to which the desired data are already available in published reports and decide whether all or part of the needed data can be obtained through a formal questionnaire.
The entire process of questionnaire- construction can be divided into following aspects:
(i) Information to be sought.
(ii) Type of questionnaire to be used.
(iii) Writing a first draft.
(iv) Re-examining and revising questions.
(v) Pre-testing and editing the questionnaire.
(vi) Specifying procedure for its use.
(i) The formulation of the problem provides the starting point for developing the questionnaire. The investigator must decide what aspects of the problem are to be dealt with in a particular study.
(ii) The appropriate form of questions depends on the mode of administration, the nature of information sought, the sample of people and the kind of analysis and interpretation intended. The investigator must also decide whether to use closed or open-ended questions.
The use of follow-up questions or probes is advisable at many points in connection, especially, with free responses. The questionnaire should anticipate where they are required and should provide the appropriate wording. Usually, several alternatives are needed, depending upon the preceding responses. For example, if the answer is too general and vague, the follow-up question may be “What do you mean?”
(iii) The best sequence of topics should be carefully considered when framing the questions. Some closely-related questions may be asked in order to measure consistency and for checking the reliability of responses. At this stage of questionnaire-framing all available suggestions need to be utilized. Questionnaires previously drafted on similar lines may prove quite helpful.
(iv) Next, the questionnaire should be scrutinized for technical defects, quite apart from biases and blind spots arising due to personal values.
(v) Pre-testing is necessary to find out how the questionnaire works and whether changes are necessary before launching off the full-scale study. The pre-testing of instrument provides a means for solving unforeseen problems in its administration in the field. It may also indicate the need for additional questions or deletion of others.
If substantial changes are warranted, a second pre-test may be conducted. Sometimes, a series of revisions and pre-testing is required. After pre-testing, the final edition is done to ensure that every element passes scrutiny of content, form, sequence of questions, spacing and arrangement. Editing is intended to make the questionnaire as clear and as easy to use as possible.
The questionnaire itself should contain simple, straight-forward directions, indicating, just what the respondent is supposed to do.
Sequence of Questions:
It is essential to examine the order in which questions are to be asked. Many a refusal and misunderstanding can be avoided by proper arrangement of questions. Questions should be arranged logically. The question arrangement makes it possible to determine the directions of the response. The opening questions should be such as to arouse human interest. The respondent then is less likely to refuse to co-operate.
The opening questions should be such as are easy to answer. Questions which might embarrass the informant should be placed in the middle or at the end of questionnaire. Questions on economic status, or those that test the knowledge of the respondent and those of an intimate personal nature should be put at the end.
Questions seeking personal information should be put at the end. Questions to which the informant may be sensitive should not be put at extreme end because this may leave him with wrong impressions and make it difficult to question him later. Since there is the likelihood of the mail-informant losing interest as he proceeds to fill in the questionnaire, important questions should advisedly be put at beginning.
Advantages of the Questionnaire:
Let us now turn to discuss the typical advantages of the questionnaire as compared to other major methods of collecting data for research:
(1) In so far as the questionnaire is usually mailed to the respondents and contains specific, clear-cut directions, the persons charged with the collection of data need not exert themselves on offering additional explanations or instructions. It is obvious that the questionnaire technique does not call for any special skills of training on the part of investigators in the field.
(2) Since the questionnaire approach makes it possible to cover, at the same time, a large number of people spread over a large territory, it is decidedly more economical in terms of money, time and energy. Other methods do not afford such a facility.
(3) The questionnaire, by its very nature, is an impersonal technique. Uniformity from one measurement situation to another is provided by virtue of its standardized working of questions, standardized sequence of questions and fixed or standardized instructions for recording responses.
This alleged uniformity, viewed from the psychological angle, is often more illusory than real. A given question in spite of its ‘standardized’ wording could have different meanings for different persons.
Careful trial-testing and helping respondents to understand the questions in the course of administration may, however, go a long way toward ensuring uniformity of questions in the questionnaire and as such, making the replies comparable.
(4) Yet another typical merit of the questionnaire is that it ensures anonymity. The respondents have a greater confidence that will not be identified as holding a particular view or opinion. The subjects feel more free to express views that they think would arouse disapproval or get them into trouble.
It has been found that there is often a marked difference between the replies to the questionnaire and those to the interviews. The difference stems from the element of anonymity that is characteristic of the questionnaire approach. Anonymity is not, however, the best method of eliciting replies at all times.
Complex issues like familial adjustment which are bound to have strong emotional overtones may not be inquired into by means of the anonymous tool that a questionnaire is. Here a personal understanding and permissive manner on the part of an interviewer may prove effective.
(5) The questionnaire places less pressure on the respondents for immediate response. The subject, given ample time, can consider each point carefully before actually putting his reply in writing.
If there is some kind of pressure for time on the subject (as is often the case in interview) he may reply with the first thought that comes to his mind. It should, however, be noted that pressure on the subject for immediate response has a certain advantage in situations where spontaneous responses matter.
Limitations of the Questionnaire:
The foregoing discussion also hints at some of the disadvantages or limitations of the questionnaire. We shall deal with these at some length:
(1) One of the major limitations of the education questionnaire is that it can be administered only on subjects with a considerable amount of education. Complex questionnaire requiring elaborate written replies can be used indeed on a very small percentage of population.
It is seen that even the highly educated persons may have little facility for writing and even granting this, very fey have the motivation and the patience to write as much as they might speak out.
Thus, questionnaires are hardly appropriate for a larger section of maintaining interest on the contemporaneous burden of writing and of maintaining interest on the subjects is quite heavy, the number of questions they may be asked as also the fullness of response is severely limited.
(2) In a mailed questionnaire, the proportion of returns is usually low; it may sometimes be as low as 10%. Among the factors that may affect the returns are the sponsoring agency, the attractiveness of the questionnaire, its length, nature of the accompanying appeal, the care of filling out the questionnaire and of mailing it back, inducements for replying and the kind of people to whom questionnaire is sent, etc.
Even under the best of circumstances, a considerable proportion may not return the questionnaire.
(3) In a questionnaire, if the respondent misinterprets a question or writes his reply unintelligibly, there is very little that can be done to correct this. In this approach there is no facility for repeating questions, explaining them or seeking clarification of a particular response.
In questionnaire approach, the validity of respondents’ reports can hardly be appraised. The researcher here is in no position to observe the gestures and expressions of the respondents. He cannot follow-up the inconsistencies or contradictions in the replies.
(4) The usefulness of the questionnaire is restricted to issues on which the respondents have more or less crystallized views that can be simply expressed in words.
The rigidity of questionnaires and the inability to explain elaborately in writing one’s ‘abnormal’, anti-social feelings and behaviour coupled with the fact that the subject has to render his response in writing — all go against frank discussion of socially-tabooed issues raised in a questionnaire.
(5) The success of the questionnaire approach depends upon the ‘sense of responsibility’ among the subjects. A serious attempt at filling out the questionnaire-format presupposes, among other things, the awareness on the part of the subjects of their responsibility to the larger institution of science.
Only then may responsible help be forthcoming. Such an awareness, even in countries where education is quite advanced, is difficult to come by.
(6) A significant limitation of the questionnaire is that the researcher/investigator is not in a position to vary the stimuli or social atmosphere impinging upon the subjects according to his designs.
Certain other approaches to data collection do allow this facility to a greater or lesser extent; the stimuli or atmosphere as he questions the subjects. Such a flexibility characteristic of the interview approach is conspicuously absent in the questionnaire.
This flexibility is indeed a very valuable asset. Johan Galtung has offered a very apt analogy that sums up our discussion. He likens interview to the musical symphony; waves after waves of sound impinging upon the audience.
The questionnaire, on the other hand, has been likened to presentation before subjects of stimuli in the manner of a painting, various shades of colour spread over the canvas. In other words, the interview method involves presentation, one after the other, of stimuli on the continuum of time whereas in the questionnaire, these are presented in space.