9                SURVEY METHOD                 (QUESTIONNAIRE TECHNIQUE)

Surveys are probably the most common form of research method for collection of primary data, but not always properly understood and carried out.

The first purpose of a survey is to describe; to count the frequency of some event or to assess the distribution of some variables such as proportion of the population of different age groups, sex, religion, castes and languages, knowledge, attitude and adoption of practices about particular issues, and other information of similar nature about the population. You will conduct this type of survey for your baseline study. 

The descriptive survey of the population is valuable in understanding the audience, and in the definition of the existence and magnitude of the problems. 

The survey data are also helpful in determining cause and effect relationships between variables, for example, between level of education and frequency of watching educational programmes on TV.

The preliminary descriptive survey results can prove useful for planning more sophisticated survey studies with a view to identifying areas where problems occur or where changes are required, to understand why people behave in a certain manner and what can be done to provide alternate solutions to the problems. Here, an attempt is made to understand the relationships between different variables, and the purpose of survey becomes to diagnose or analyse the situation rather than just describe the situation. 

Surveys may also be done to measure the extent and nature of effect and impact of a project after the population has been exposed to media for a reasonable length of time. 

Salient Features of Survey Method 

Surveys are done to gather data from the field in order to generalize results from a sample to a larger population. The primary purpose and advantage of surveys is, therefore, generalization of the results. 

In surveys, we are usually interested in gathering data from many respondents than in obtaining intensive, detailed information from a few individuals. 

While designing a survey research study, you have to take special care of the sample and the sampling procedure.

 The sample size should be adequate to allow generalization of the results. 

The sampling procedure should also be such that small sub-groups within the population (such as landless farmers) are properly represented in your sample. Errors in sampling procedures may not justify generalisation of the results, lowering the value of the survey.  Basic knowledge and descriptions of populations and geography etc. are preconditions for survey research, particularly for sample construction and designing of tools for information gathering. It is essential for you to first collect pilot data through case studies, observations and other methods before you mount meaningful survey studies.  Further, most survey data-gathering tools are structured to facilitate quantification of the responses. The major tools used for the survey consist of questionnaires and structured interviews. We shall first talk about the questionnaire as an information gathering tool.

Questionnaire as Data Collection Tool 

When using the questionnaire, information is gathered by asking selected members of the audience to answer a set of questions posed to them in a written form. 

A questionnaire consists of a number of questions in a definite order on a form(s). The forms are usually mailed to the respondents who are expected to read and understand the questions, and to reply to these by writing in the relevant space provided for the purpose on the forms. The respondents are also expected to return the filled-in questionnaire. 

In certain situations, you may hand over the questionnaire to the respondents individually or in a group, and get these filled personally, offering necessary explanations with reference to the questions, if and when necessary. 

In other cases, you may put the questions as they appear in the schedule to the respondents, and also record their replies. 

You may use your medium such as radio and TV in putting some questions across the audience, and requesting them to send their replies to you. Telephone is also used to conduct the surveys. 

Strengths of the Questionnaire as a Tool 

It is far more convenient and economical to collect information this way. Large amounts of data can be collected from small or large population groups in this way. It is a suitable method when it is intended to collect some specific information from a large number of people. (For example, information about what the women know about the childcare and early childhood diseases can be quickly obtained through a questionnaire). The questionnaire technique is impersonal, and avoids bias, which can develop as a result of interaction between the researcher and the respondent. It ensures some degree of anonymity to the respondents. The respondents feel free to express their views through a questionnaire than they would do personally to the researcher. It places less pressure on the respondents for immediate response. They can complete it at their own time and pace. They can also look through the whole questionnaire and form an idea of the nature and scope of questions before replying to these.

Limitations of the Questionnaire as a Tool 

This can be administered only to the respondents who are able to read and write fairly well. Hence, it may not be appropriate for the rural people in general. However, it can still be very well used for organized literate groups in the rural areas such as students, teachers, government and non-government extension functionaries, social workers, NGOs etc. 

By itself, this is not a preferable method when you are trying to explore the situation, and doing an in-depth study of the needs, constraints, context and solutions to the problems of the audience. The respondents may not give their full attention to the job of replying to the questions. The answers may lack depth resulting in superficiality. If the respondent misinterprets a question, there is little that can be done to correct. There could then be inconsistencies in the replies.  They may not reply to all questions for one reason or another and leave some blanks.  Generally, mailed questionnaire survey does not produce a high rate of recovery. Quite a few respondents may not care to return the filled- in questionnaire even after reminders.  Collection and compilation of information is time consuming.

Forms of Questions 

However, the more important matter for you when you use this tool is to design a really good questionnaire comprising relevant, meaningful and easy to understand questions. How you frame your questions makes all the difference in the quality of the responses. If a questionnaire is not well designed, the information gathered will be of poor quality. It will also affect the recovery of the questionnaires. Let us look at the main forms of questions, and the uses to which these are put.

Closed Questions 

The usual format of a closed question is to ask a question, then provide the range of answers, and ask the respondent to tick the appropriate answer. It is also called multiple-choice question. An example is given below: 

What are your reasons for not going to the TV centre to watch TV? 

Lack of time
Long distance
Not safe to move in the night
Programmes not interesting
No proper seating arrangement

 Any other reason (Please specify)____________________________________
______________________________________________________________ 

Please note that the responses include a safety net in the form of the final response category namely, any other reason just in case anyone has a different stand. 

Open Ended Questions 

Here the range of possible answers to the question is not provided. The onus is placed on the respondent who is expected to formulate and record answers in his or her own words. For example, the above mentioned closed question on reasons for not watching TV can be framed in the open ended form as under:- 

Q. If you don’t go to TV centre to watch TV, what are your reasons for that? _________________________________________________________________ _________________________________________________________________ _________________________________________________________________ 

Open-ended questions may be used when you are not sure of all the possible range of responses, and you are trying to understand the situation. These questions can produce detailed answers to tricky situations and complex problems, and are, therefore, good in eliciting the feelings and opinions of the respondents.  However, open-ended questions are effective only when administered to the respondents who are good in expressing themselves and are willing to do so. Since it requires more effort from the respondents, the risk is that these are less likely to be fully attempted and completed.

It produces a wide range of answers posing difficulty for the researcher to categories and analyse the responses.

It would, therefore, be better if your questionnaire combines both closed and open-ended questions according to the nature of the study. Indeed, it is often desirable to follow a closed question with one or two open ended questions to obtain an insight into the problem. Some examples are given below: 

Q. (Closed) Do you find TV programmes interesting? 

Yes, most of the time
Yes, sometimes
No, mostly boring

Q. (Open Ended) If you find most or some of the programmes not interesting, why do you think so? ______________________________________________________________ ______________________________________________________________ ______________________________________________________________ 

Q. (Open Ended) Are there any suggestions you would like to make to make the programmes interesting? _________________________________________________________________ _________________________________________________________________ _________________________________________________________________ 

Rating Scale

Coming back to the closed questions, in some instances a simple range of responses is not adequate, if you are trying to get at the shades of opinion or the levels of importance. In such cases, it is profitable to use a rating scale. An example of a rating scale administered to teachers is given below: 

Q. You have been teaching standard V for quite sometime. Please indicate the difficulty level of different subjects as experienced by most students of this standard. 

Ranking Scale 

When the objective is to know the degree of preferences or the relative weightage given by the respondents to different items you may design a ranking scale. For example, to measure the credibility accorded by the farmers to different sources of agricultural information, a question may be framed as under: 

Q. There are generally four sources of agricultural information namely, demonstrations, television, radio and extension workers. Please rank these in order of credibility from the most credible to the least credible. 

The other way is to provide these sources in different possible pairs and asking the respondents to select one source over the other from each pair having more credibility. The question could then be framed like this:

Q. The following list provides important information sources in different pairs. Please select one source over the other from each pair having more credibility.?

1.	Television
	Radio
2.	Demonstrations
	Extension workers
3.	Radio
	Demonstrations
4.	Extension Workers
	Television
5.	Demonstrations
	Television
6.	Radio
	Extension workers

Yet another way to put the same question in a more simplified manner could be: 

Q. Which source would you choose from the following four sources for obtaining agricultural information if you could have only one of these? Please     tick ( ) in the box against it. 

1.	Demonstrations
2.	Television
3.	Radio
4.	Extension workers

 

There could be several ways of putting a question. Designing a good questionnaire, which is easy to understand, appealing and motivating to the respondents is a skill, which can be acquired with some practice and patience.  Some more hints on using the questionnaire method are given below.

Hints on Using Questionnaires 

In the beginning, state concisely what the survey is about, and how the findings will be used. Request the respondents for their co-operation.

Keep things short and to the point. You may have to edit and revise your questionnaire several times to make it concise or else you may end up collecting more information than what you need. Write the questions in plain and simple language, which is understandable to the respondents.  Write as if you are talking to the respondent. Aim for a warm, friendly tone. Keep the sentences short.

Avoid ambiguous questions. If need be, define certain words which can create ambiguity. For example, regarding the question on credibility of information sources in the previous section, the word credibility may be defined as ‘both trust worthiness and expertness’ for the respondents. 

In other instances, ambiguity can arise through the construction of a question. This often happens when a question contains double negatives. For example 

Q. Don’t you think that telecast duration should not be reduced? 

Avoid leading questions or questions, which encourage a particular response. For example, the following question is more likely to end with ‘yes’ responses. 

Q. Do you think the quality of TV programmes needs improvement ? 

• With ‘Yes/No’ answers, it may also be necessary to make provision for ‘Don’t know’ or ‘Can’t say’ response to enable the respondents to express themselves more precisely. 
  
• While structuring the questionnaire schedule, proceed from general to specific. 
  
• For developing a good questionnaire don’t forget to first pre-test and pilot-test the questionnaire. This is important not only to ensure that the questions are good and unambiguous but also to check that nothing important has been overlooked. 

Enclose a self-addressed, stamped or pre-paid cover with the questionnaire if it is mailed. This will help increase the response rate. There will, however, still be a need to chase the respondents. A polite reminder enclosing another copy of the questionnaire may help better recovery.  For designing appropriate questionnaire and structured interview schedules for data gathering, it helps, if you first know your audience and have good contact and rapport with them. It would therefore be better if you spend some effort in using relatively unstructured data gathering techniques - unstructured interviews, focused group discussions, observations, case studies etc. before moving to such structured techniques as survey methods. You will find discussion on these methods and techniques in the following chapters.