IB Psychology SLTopic 5 — Methods of ResearchPaper 1 & 2Core skill~10 min read
Interviews and What They Reveal
An interview is a questionnaire with a human being attached. That one difference gives you everything a questionnaire cannot reach — follow-up questions, tone, hesitation, the thing they mention at the end — and it costs you the thing a questionnaire is best at, which is standing still while you measure it.
📘 What you need to know
An interview is a self-report method carried out face to face, by phone or online.
Structured: every question written in advance, asked in the same order, no deviation.
Semi-structured: a prepared list plus the freedom to follow up.
Unstructured: no fixed questions; it runs like a guided conversation.
More structure means more reliability. Less structure means more depth and richer validity.
Focus groups interview six to twelve people together, usually with a moderator.
Group settings add two problems of their own: dominant respondent bias and groupthink.
The one axis that explains all three types
Structured, semi-structured and unstructured are not three unrelated methods. They are three points on a single line: how much of the interview was decided before anyone sat down.
The blank bars are the point. Every space you leave in the script is space for something you did not think to ask about.
Structured interviews
The interviewer reads a fixed set of questions in a fixed order and does not go off script. It is essentially a questionnaire read aloud, and it inherits a questionnaire’s strengths.
Different researchers can run the same interview, so it can be replicated and checked.
The researcher effect is reduced, because nobody can chase the answers they find interesting.
Most of the data comes out as counts and scores, so it can be analysed statistically.
The weakness sits in the same place as the strength. If a participant says something that opens up the whole topic, the format will not let you follow it. You end up with tidy data about the questions you happened to think of.
Semi-structured interviews
The researcher brings a list of questions but is allowed to leave it. If something interesting or unexpected comes up, they can chase it. If a participant is nervous or stuck, they can slow down and encourage.
Why this is the most common type in real psychology: the prepared questions keep every interview covering the same ground, so comparison is still possible, while the freedom to follow up means you can find out why someone answered as they did.
The cost is messier data. Some participants get asked things others were not, so the interviews are not perfectly comparable, and analysing the mix of numbers and stories takes far longer.
Unstructured interviews
No prepared list. The researcher opens with something broad — “tell me about the year after you moved here” — and follows the participant. These are sometimes called narrative interviews, because the participant ends up telling a story rather than answering a list.
They produce qualitative data only, and it is often extraordinarily rich: the details, contradictions and turning points that a fixed script would never have reached. That is the case for high ecological validity.
🤔 Why the depth also creates the danger
The closer the researcher gets to the participant, the more honest the account can be — and the more likely the researcher is to feel involved. Over several sessions they may start to identify with the person, hear what fits their expectations, and let it colour how they write it up. That is a loss of objectivity, and it is the standard limitation to give for unstructured interviews.
Type
Script
Data
Best at
Worst at
Structured
Fixed, in order
Mostly quantitative
Replication and reliability
Depth and follow-up
Semi-structured
Prepared, but flexible
Both types
Balance of the two
Messy, harder to analyse
Unstructured
None
Qualitative only
Depth, individual experience
Reliability and objectivity
Focus group
Usually guided by a moderator
Qualitative
Shared views, group dynamics
Dominant voices and groupthink
Focus groups
A focus group interviews several people together, usually six to twelve. Fewer than that and the discussion dies; many more and people stop getting a turn. A moderator starts the conversation and keeps it roughly on topic without steering the content.
The reason to use one is that people build on each other. Someone mentions a small detail, three others recognise it, and a shared experience surfaces that no individual interview would have produced. You also get to watch the group dynamics themselves.
Two named weaknesses live in this picture: dominant respondent bias on the right, and groupthink pulling the quieter members towards the majority view.
If a question mentions a focus group, you already have two evaluation points ready before you have read the rest of the stem. Name them properly — dominant respondent bias and groupthink — and explain what each does to validity.
Worked examples
WORKED EXAMPLE
Choose and justify an interview type
A psychologist wants to understand how adults who moved country as teenagers now describe their sense of belonging. Suggest a suitable type of interview and justify your choice. [4]
Step 1: What does the aim need?
Personal experience in the participant’s own words, not counts.
Step 2: Rule out structured
A fixed script would cut off anything the researcher had not predicted.
Step 3: Choose and justifyUnstructured or semi-structured, because the participant can lead, and follow-up questions can chase what matters to them.
Semi-structured, for depth plus comparabilitysay one weakness too: rich data but low reliability, since no two interviews are the same
WORKED EXAMPLE
Explain a limitation of a focus group finding
In a focus group of eight students, one participant states early on that the new timetable is unfair. By the end, seven of the eight agree. Explain why the researcher should be cautious about this finding. [3]
Step 1: Notice the order of events
The strong opinion came first, and agreement followed it.
Step 2: Name the two effectsDominant respondent bias shaped the discussion, and groupthink pulled the rest towards the stated view.
Step 3: State the consequence
The agreement may record group pressure rather than eight real opinions.
Validity of the finding is reduceda fix worth adding: collect private written views before the discussion starts
💡 Exam tip
State the data type each interview produces. It is the fastest way to show you understand the difference.
Justify a choice by working backwards from the aim. “The aim is to explore experience, so depth matters more than comparison.”
Reliability and depth pull in opposite directions here. Say that sentence in any evaluation of interviews.
Recording and transcribing interviews is a real strength worth mentioning: nothing depends on the researcher’s memory.
For focus groups, always name the moderator’s role. It shows you know it is a managed method, not just a chat.
Social desirability bias applies harder here than in a questionnaire, because a real person is listening.
⚠️ Common mix-up
Unstructured does not mean unplanned. The researcher has thought hard about the topic; they have just not fixed the questions.
Semi-structured is not “half a structured interview”. It is a prepared list plus permission to follow up.
A focus group is not a group interview by accident. The interaction between members is part of the data.
Confusing interviewer bias with social desirability bias. One comes from the researcher, the other from the participant.
Saying interviews are always qualitative. A structured interview with closed questions gives quantitative data.
Forgetting that rapport cuts both ways. It gets honest answers and it costs objectivity.
Up next: Correlational Studies and Their Limits — what to do when you cannot manipulate anything but you still have numbers.
Want this explained one-to-one?
Book a free session with an experienced IB Psychology tutor and get your trickiest topics made simple.
