4.2.3 - Questionnaires and interviews

What self-report techniques do

Sometimes the most direct way to study thoughts, feelings, attitudes or private experiences is to ask people to report them. That is the logic of a self-report technique. In AQA research methods, the two core self-report techniques are questionnaires and interviews.

The two forms named in this specification point are questionnaires and interviews.

An interview still asks participants about themselves, but it does this through an interviewer-led exchange.

Self-report techniques are useful because many psychological variables are not directly visible. A researcher cannot simply observe how anxious a participant feels before an exam, what beliefs they hold about obedience, or why they chose a particular coping strategy. A self-report method allows the researcher to collect data about those internal states.

The basic trade-off is important:

AO2 example: Dr Mensah wants to study revision stress in Year 13 students. If she emails students a form with rating-scale items, she is using a questionnaire. If she speaks to each student and asks them about their stress experiences, she is using an interview. In both cases, the data depend on what participants say about themselves, so both are self-report techniques.

Self-reports can also be inaccurate without deliberate lying: people may forget details or estimate poorly. Self-reports are not automatically weak. They can be the most ethical and practical method for sensitive topics. The exam skill is to explain when the method fits the aim, and then evaluate whether the reported answers are likely to be valid.

Questionnaires and question types

A questionnaire is made up of items. The type of item determines the type of data collected and how easily the data can be analysed.

A fixed-answer item creates a different kind of evidence.

Open questions invite unrestricted answers. Prompts asking people to describe or explain usually produce qualitative data. For example: "Describe one thing that makes exams stressful for you." This can reveal detail and unexpected themes, but it takes longer to code and compare because each answer may be different.

The response format and the data type are separate decisions: “How many hours did you sleep last night?” allows a free numerical answer but produces quantitative data. A closed item may instead record a nominal category such as a preferred study method.

Closed questions usually produce data that is easier to count and compare. For example: "How anxious do you feel before exams? 1 = not anxious, 5 = extremely anxious." The researcher can calculate frequencies, percentages, medians or means, depending on the scale used. A simple yes/no item produces categories that can be counted, while rating scales produce numerical scores.

Common closed-question formats include:

Good questionnaire construction is not just about choosing open or closed questions. Items should be clear, neutral and unambiguous. A leading item such as "How much do irresponsible students overuse social media?" pushes the respondent towards a judgement. A double-barrelled item such as "Do you revise early and sleep well before exams?" asks two things at once, so a single answer is hard to interpret.

Mixed questionnaires often work well. Closed items give standardised data for comparison, while one or two open questions can explain why participants chose particular answers.

AO2 example: A school wants to measure students' attitudes to homework. "Do you enjoy homework? Yes/No" is quick to analyse, but it may miss the difference between students who dislike homework because it is too hard and students who dislike it because it is repetitive. Adding "Explain your answer" would provide richer qualitative detail.

Evaluating questionnaires

Questionnaires have major practical strengths. They can be distributed to many respondents at once, especially online, so they are often quick and inexpensive. The researcher does not need to be present, which saves time and may reduce embarrassment for sensitive topics. Closed-question questionnaires also produce standardised data, making comparison between groups straightforward.

Questionnaires can also support anonymity. If participants believe their answers cannot be linked back to them, they may be more honest about behaviours such as sleep, diet, prejudice, drug use or mental health.

However, the main evaluation issue is validity: do the answers reflect what participants genuinely think, feel or do?

If a questionnaire asks, "How often do you lie to your teachers?", some respondents may underreport the behaviour because they want to appear honest. This matters because the data may measure the respondent's wish to look good, not the target behaviour.

One common response bias is agreeing too readily with statements.

Acquiescence bias is especially likely with repeated agree/disagree items. A respondent might tick "agree" down the page because it is quick, polite or low effort. The researcher can reduce this risk by using clear item-specific questions, reversing some items carefully, and piloting the questionnaire to check whether respondents understand the wording.