Read the question’s period, unit and answer instruction before choosing a response. Five invented examples show how an accurate answer changes when the question changes. These are reading exercises, not real screeners, qualifying answers or a way to predict acceptance.

1. Use the period the question actually names

In invented example one, the question asks whether you bought a kettle between June 1 and June 30. The exercise gives two purchases: one on May 28 and one on June 2. Only the June 2 purchase falls inside that stated interval. The example supplies both dates; it does not ask you to remember an event that never happened or infer what answer a researcher prefers.

Read a named interval as written. If a real question instead says the past 30 days, do not silently replace that with the previous calendar month. Follow any supplied date definition. If you cannot recall the requested period accurately, use an offered truthful uncertainty option or seek clarification rather than borrowing the example's answer.

2. Keep personal and household units separate

Invented example two gives your own spending as 3 units and another household member's separate spending as 7 units, with no shared transaction. A question explicitly asking for your personal total has an answer of 3 in this exercise. One explicitly asking for the combined household total has 10. The difference comes from the stated unit, not from a change in anyone's circumstances.

Check whose behavior is requested and what belongs in the total. Do not assume a question about you authorizes a guess about every household member. In an actual study, use the researcher's definitions and the information you genuinely know. The arithmetic here is a teaching example, not a household budget, observed survey or earnings result.

3. Follow the selection rule, not the number of boxes

Invented example three says you used both a bus and a train during the named week. If the instruction is to select every listed mode you used, both match the exercise. A different question asking for one most-used mode needs a comparison that those facts alone do not provide. Merely knowing that both were used cannot tell you which was used most.

A checkbox or radio button is part of the interface, but read the written instruction too. Do not add an untrue choice to make a response look more complete. If the prompt and control conflict, leave that ambiguity visible and use a permitted clarification route; this guide cannot repair the questionnaire or authorize ignoring its instructions.

4. Notice when one response combines two judgments

Invented example four asks for one rating of a delivery that was fast and helpful. The exercise describes prompt arrival but unhelpful assistance. Those facts do not determine one combined score. They illustrate a question containing two judgments, rather than a reason to invent a neutral opinion or choose the response that seems safest for screening.

AAPOR recommends that researchers ask about one concept at a time. That design guidance helps explain this difficulty; it is not a participant rule allowing you to rewrite a live study. Follow the study's actual instructions, use an offered comment or clarification option if suitable, or decline a question you cannot answer honestly.

5. Distinguish no, unknown and not applicable

Invented example five offers a list of products bought during a defined week. If you know that you bought none of them, an offered none option describes that fact. If you cannot remember, do not turn uncertainty into a purchase denial. If the question concerns a task you do not perform, use a fitting not-applicable option when one is offered and its instructions fit.

These labels depend on the questionnaire; not every study supplies all of them. AAPOR distinguishes missing responses from none, don't-know and refusal categories. That supports keeping the meanings separate, not assuming that any particular answer earns credit. No button choice here guarantees qualification, a reward or continued access.

Use the exercises to read accurately, then keep your own answers

All five examples were invented for this guide. Their facts are sufficient only for the limited conclusions stated. None was taken from a confidential live study, and none supplies a profile, attention-check key or set of answers to rehearse. Do not paste these examples into research about your own experiences.

This page concerns interpreting one question. Use the separate routine guide for invitations and record keeping, or the eligibility guide for screening outcomes. If a live study is unclear, follow its permitted clarification or exit process without repeatedly trying different personal answers. The notebook cover is a concept image, not a completed questionnaire or participant record.

Sources and further reading

Provider terms and availability can change. This guide does not report a personal payout test.

Join the conversation

No reader comments have been published on this article yet. Log in to check community access.