Surveys

Definition

A social Science research tool used to study a wide range of characteristics of a population.

Note

Population survey—studies a whole population.
Sample survey—studies a portion, or sample, of a population.

Survey instruments—questionnaires, interview schedules or interview guides (for a brief description of these survey instruments, see Definitions).

A survey is not inherently a participatory research tool. But it is increasingly being used in a participatory manner as community members use it to examine their own affairs.

Materials

- Survey instrument—questionnaires or interview schedules (one copy for each respondent) or interview guides

- Notebooks

- Pens

Purpose

To generate baseline and evaluation data (see Abbreviations and Definitions)—qualitative or quantitative—and to answer questions identified using other methods.

Possible approach

1 Develop a list of information to be gathered. Prioritize the data to be collected.

2 Choose the type of instrument to be used.

3 Develop the instrument (the questionnaire, schedule or interview guide). Use simple language.

4 Pre-test the instrument with the same type of population and in a setting similar to the intended research site. Revise the instrument if necessary. Repeat the pretest to ensure that information generated is reliable (the same quality of results should be generated again and again) and valid (it should gather the type of information for which it is designed). Surveys should generate the same results even if they are conducted by different interviewers.

5 Select the sample. You can select respondents at random or by using other appropriate sampling techniques (see How to draw a sample). The type of sample that you use will determine the type of statistical analysis you can perform.

6 Administer the survey.

- Mailed questionnaires—Send the questionnaire with a courteous cover letter and clear instructions. Send follow-up letters to remind respondents to return the questionnaire.

- Self-administered questionnaires—Give the instrument to the respondent (or group of respondents) and explain it. Ask if the respondents have any questions. Ask them to complete the questionnaire and return it to you. Quickly review the completed questionnaire and obtain any missing information before leaving.

- Personal interviews—Follow the guidelines for interviewing (see topic Interviewing). The researcher should write the responses on the instrument.

- Interview guide—C fine topics in sequence and try to maintain the interview's interest. Note responses an the interview form. At the d of the interview, review the information gathered ensure nothing has been missed.

7 Organize and analyze the

8 Discuss and validate the results with the community. Return the results to the community. (See topic Intellectual property rights.)

Value

- Surveys are useful for identifying and documenting people's indigenous knowledge, practices, and their cultural context.

- Sample surveys can be used when the indigenous knowledge of a large population must be inferred from a small portion of the population.

- Surveys can help determine how widespread is a practice, technique, or belief in a community.

- Surveys with structured questionnaires are useful to capture the "what," "who," "where," "when" and "how often" of IK, but they are less suited to discover details of the "how" and "why." This is true especially if the questions are closed-ended (offering only prescribed response choices).

- Questionnaires designed by local people can provide useful insights into their IK.

Dos and don'ts

- Do train interviewers in how to administer your questionnaire or conduct your interviews.

- Don't use a self-administered questionnaire when surveying illiterate people. Use an interview guide instead.

- Do pretest the questionnaire to make sure respondents understand the questions and can answer them correctly. Revise the questionnaire if necessary.

- Do keep the instrument short and simple. Ask only those questions which you need in order to answer the research question.

Note

Close-ended questions: Where me respondent chooses me answer born a predetermined list Example: Where did you fires learn of this new farming practice:

(a) From family members
(b) From other relatives
(c) From friends or neighbors
(d) From the extension agent
(e) From other sources (specify)

Open-ended questions:

Where me respondent is asked to provide his or her own answer.

Example: How did you fires learn of this new farming practice? Responses to closed-ended questions are easier to analyze. but they may give less insight into the subject than open ended questions.

Compiled by Jit P. Bhuktan
Sources: Frank Lynch 1979, Kerlinger 1982, Kidder 1981