The Questionnaire and the Schedule

Its limitations

Restricted to the literate, which in Indian conditions excludes a large part of many populations and, crucially, is not a random part of them; low response rates, and non-response is systematic rather than random, which can destroy the study; no help available if a question is misunderstood, and the misunderstanding is invisible; incomplete answers and skipped questions; no certainty who answered it; no non-verbal information; and inflexibility, since nothing can be followed up.

The schedule

What it is

The same list of questions, administered by a trained investigator who asks, explains where necessary, and records the answers. It is the instrument of most large Indian surveys, including the Census and the National Sample Survey, for the obvious reason that it does not require the respondent to read or write.

Its merits

Reaches the illiterate, which is decisive here; high response rate; misunderstanding can be corrected on the spot; the identity of the respondent is certain; non-verbal and contextual information is available to the investigator; incomplete answers can be pursued; and it can be combined with observation of the household.

Its limitations

Costly, since investigators must be recruited, trained, deployed and supervised; slow; limited coverage for a given budget; interviewer bias, in all the forms set out in [The Interview]; no anonymity, so social desirability bias is stronger; and it requires training and supervision, without which standardisation fails and the data are not comparable.

The comparison

The one-line answer: the difference is who writes the answers, and everything else follows. Choose a questionnaire for a literate and dispersed population, especially on a sensitive subject where anonymity matters; choose a schedule for a general Indian population, and always where literacy cannot be assumed.

A worked example

Study one: the working conditions of advocates' clerks in a district court. The population is literate, concentrated in one building, and the subject includes earnings and treatment by employers, on which people will not speak freely to a stranger taking notes. A questionnaire, distributed and returned anonymously, is the right instrument: the anonymity is worth more than the follow-up would be.

Study two: awareness of maintenance rights among women in a rural block. Literacy cannot be assumed; the questions require explanation; the researcher needs to know who actually answered; and the household context is itself data. A schedule is the only workable instrument, administered by trained women investigators, and privacy during the interview is a design requirement for the reasons in [The Interview].

Note that the subject matter did not decide it. Both studies concern sensitive matters. What decided it was literacy, dispersion and whether anonymity or clarification mattered more, and that is the reasoning an examiner is looking for.

Quick revision

1. Questionnaire: the respondent fills it in. Schedule: the investigator fills it in. Everything else follows from that.
2. Question forms: closed, open, mixed or cafeteria, contingency, scale.
3. Construction rules: short; simple language; one idea per question, no double-barrelled questions; no leading questions; no ambiguous words such as often; easy questions first and sensitive ones late; provide for "do not know"; and pre-test it.
4. Questionnaire merits: cheap, wide reach, no field staff, no interviewer bias, time to reflect, anonymity.
5. Questionnaire limits: literate respondents only, low and non-random response, no clarification, incomplete answers, uncertain identity, no context, inflexible.
6. Schedule merits: reaches the illiterate, high response, clarification possible, identity certain, context and non-verbal data, incomplete answers pursued.
7. Schedule limits: costly, slow, narrow coverage, interviewer bias, no anonymity, and it needs training and supervision.
8. Choose by literacy, dispersion, and whether anonymity or clarification matters more.

Test yourself

1. Distinguish a questionnaire from a schedule. A questionnaire is a list of questions sent or given to respondents who read and answer it themselves and return it. A schedule is a similar list carried by an investigator who puts the questions orally and records the answers. The distinction lies in who administers the instrument, and every other difference follows: literacy is required of the respondent only for a questionnaire, cost and response rate move in opposite directions, interviewer bias is absent from one and present in the other, and anonymity is available only with the questionnaire.

2. State the rules for constructing a good questionnaire. Keep it short, since length causes non-response; use simple language in the respondent's own vocabulary; ask one idea per question and avoid double-barrelled questions such as whether the police are prompt and courteous; avoid leading questions; avoid ambiguous words such as often and usually and ask for numbers instead; put easy questions first and sensitive ones late; group related questions logically; provide for "do not know" and "not applicable"; state the purpose, sponsor and assurance of confidentiality; and pre-test the instrument before using it.

3. What is the chief advantage of the questionnaire, and the chief advantage of the schedule? The questionnaire's chief advantage is anonymity, which matters most on sensitive subjects, since a respondent will record in writing what they would not say to a stranger's face, and it is also free of interviewer bias. The schedule's chief advantage is that it does not require the respondent to read or write, which in Indian conditions decides the matter for most general populations, and it permits a misunderstood question to be explained at once.

4. Why is a low response rate a serious defect rather than merely a small sample? Because non-response is not random. Those who do not answer differ systematically from those who do, typically in literacy, in interest in the subject and in their view of the sponsor, so the returns are a biased selection rather than a smaller unbiased one. A study with a low response rate cannot be corrected by increasing the number sent out, since the same selection operates on the additional recipients.

5. How should the choice between the two instruments be made? By literacy, dispersion and the relative value of anonymity against clarification. A literate, dispersed population studied on a sensitive subject calls for a questionnaire, where the anonymity is worth more than the ability to follow up. A general population whose literacy cannot be assumed, studied on questions that need explanation, calls for a schedule administered by trained investigators. The subject matter alone does not decide it, since both instruments are used on sensitive subjects.