What a questionnaire is

1. A questionnaire collects data through a set of written questions that the participant fills in themselves.
2. It is a self-report method, so the data are what people say about themselves.
3. Because the questions are written rather than spoken, the same form can be handed to a whole year group in one lesson.
4. Every participant meets identical wording in identical order, which is standardisation.
5. The items used decide the data collected, whether open questions, closed questions or rating scales.

Open questions

1. An open question lets people answer in their own words, with no options to pick from.
2. That gives qualitative data, detailed about how someone thinks or feels.
3. Strength: an answer can raise a reason the researcher never thought to offer as an option, so validity is higher.
4. Weakness: the answers are slow to analyse, and two researchers can summarise the same answer differently, which lowers reliability.
5. An open question suits an aim about why something happens or how it feels, such as why revision makes pupils anxious.

Closed questions

1. A closed question gives a fixed set of options, so the participant ticks rather than writes.
2. That gives quantitative data that can be totalled, tallied into a frequency table and expressed as a percentage.
3. Strength: everyone picks from the same list, so answers are quick to compare and reliability is high.
4. Weakness: when none of the options fits, the person ticks the nearest one anyway, so the answer no longer measures what they think.
5. A closed question suits an aim about how many or how often, such as how many pupils revise on a school night.

Rating scales

1. A rating scale is a closed question that asks for a rating on a fixed run of numbers, most often 1 to 5.
2. Both ends carry a label, such as 1 for "not at all" and 5 for "very much", so everyone reads the numbers the same way.
3. The number of points stays the same on every item, so ratings can be compared across questions.
4. Adding the ratings and dividing by the number of participants gives a mean rating, reported as a figure out of 5.
5. Because each rating is a number, it can also be plotted against a second measurement on a scatter diagram.
6. Strength: it turns a strength of feeling into a number, so a mean rating can be worked out and one group compared with another.
7. Weakness: the number leaves out the reason, so a 2 from one person and a 2 from another may mean different things.
8. Weakness: many people avoid the two extreme ends and bunch in the middle, which flattens the differences the item was measuring.
9. Weakness: an unlabelled run from 1 to 5 leaves each person to decide what a 3 means, so the mean rating is hard to interpret.
10. A scale with three points flattens differences that a scale with five points would show.
11. A rating scale suits an aim about strength of feeling, such as how strongly someone agrees with a statement.

Strengths and weaknesses of questionnaires

1. Strength: they are quick and cheap, so a large sample can be reached in one go, which helps generalisability.
2. Strength: the written wording is standardised, so another researcher can run it again and check whether the results hold, giving high external reliability.
3. Strength: a private form can be filled in without saying an awkward answer out loud to anyone.
4. Weakness: a misread question cannot be explained by anyone, so the answer measures something the researcher did not intend, which lowers validity.
5. Weakness: social desirability bias pushes people towards the answer that makes them look good.
6. Weakness: many forms are never returned, and the people who do reply may differ from those who do not, which weakens population validity.
7. Weakness: when the items pull in different directions the questionnaire is not consistent within itself, which is low internal reliability.
8. A questionnaire suits an aim that needs the same information from a large number of people quickly.

Writing questionnaire items

1. Keep each question about one thing, so nobody has to answer two questions with a single tick.
2. Cut any leading wording that hints at a particular answer, so "Don't you agree exams are unfair?" becomes "How fair do you think exams are?"
3. Match the item type to the data the aim needs before writing any questions.
4. Try the form on a few people first, then reword anything they misread.
5. Promise confidentiality on the front of the form, which reduces social desirability bias and meets the British Psychological Society's guidance.

Turning this into marks

1. When an item type and a reason are both asked for, keep them as two separate points, one naming the type and one naming the data it produces.
2. Write the item out in full, including both scale labels, because "I would use a rating scale" does not secure the knowledge mark.
3. Give a figure with the scale it came from, so a mean rating reads 4.2 out of 5.
4. The mark most often lost is calling a rating scale an open question, when it offers a fixed run of numbers to choose from.