IB Psychology HLTopic 8 — Health & WellbeingPaper 3 (HL)Reading data~10 min read
Does Virtual Reality Therapy Work?
Sooner or later Paper 3 hands you a graph and asks for one limitation. Three marks, two minutes, one point. Students throw away marks here by writing four half-points or by criticising the therapy instead of the data. This page fixes that.
📚 What you need to know
Graph questions test whether you can read evidence, not whether you like the treatment.
Always state which condition did better and by roughly how much. That is the analysis.
Measures such as self-efficacy, fear or confidence are subjective variables — a 7 out of 10 does not mean the same thing to two different people.
A bar chart shows differences. It does not show cause, significance, sample size or how much individuals varied.
If the x-axis has labels like “post-test” and “follow-up” with no time intervals, you cannot conclude anything about long-term effects.
“Explain one limitation” means exactly one. Extra points earn nothing and cost you time.
Only use what is on the graph. Never invent what the therapy involved or who took part.
Start by reading it properly
Before you criticise anything, describe what is there. Examiners want to see that you can extract a finding from a chart in a sentence or two.
Practice describing a chart out loud: who started where, who changed, by how much, and what stayed the same.
🧩 How to read any Paper 3 graph in 30 seconds
Read both axes. What is measured, and on what scale? A 0–10 scale is a rating, not a count.
Find the comparison. Which groups or conditions are being set against each other?
State which did better and roughly by how much. Use the numbers on the axis.
Look for what stays the same. A flat comparison group is often the most important part of the picture.
Ask what is missing. Sample size, spread, significance, time intervals, units of the rating.
Step 4 is the one students skip. If the comparison group does not change while the treatment group does, that is your evidence. If both change equally, the treatment has told you nothing.
What a graph like this cannot tell you
Every limitation you might write comes from one of these gaps. Learn the list and you will always have an answer ready.
Gap in the data
Why it limits your conclusion
The measure is subjective
Fear, confidence and self-efficacy mean different things to different people, so the numbers may not be comparable across participants
No sample size shown
A large difference from six people is not the same as a small difference from six hundred
No measure of spread
An average hides the participants who got worse
No significance test
You cannot say whether the difference is bigger than chance
No time intervals on the x-axis
If “follow-up” could be one week or one year, no claim about long-term effects is possible
Only means shown
The mean uses every value, but a median would hide the extremes; either way, individual patterns disappear
Subjective variables in one line: the scale is objective, the thing being scaled is not. Two people who both write “6” may be describing completely different experiences, which weakens the validity of any comparison between them.
How to structure a 3-mark answer
Three marks is three moves. Nothing more.
The same three-step shape works for every “explain one limitation” question in the paper.
Worked examples
WORKED EXAMPLE
Explain one limitation linked to the interpretation of the data in this graph. [3 marks]
Version A — the subjective measure
Fear is measured on a self-rated scale from 0 to 10. This is a subjective variable, because two people describing the same level of distress may choose different numbers, and one person’s ratings may drift over the course of treatment.
That means an apparent drop from 8 to 4 may partly reflect a change in how the participant uses the scale rather than a real change in fear, which weakens any conclusion drawn about how effective the therapy was.
Version B — the missing time information
The x-axis gives stages rather than time intervals, so we do not know how long passed between the measurements. Without that, it is impossible to say whether the improvement lasted a meaningful length of time, and no conclusion about long-term effectiveness can be drawn from the graph.
Pick one. Writing both is not rewarded — the question says “one”.
WORKED EXAMPLE
Analyse the findings shown in this graph and state a conclusion based on them. [6 marks]
Describe the data
The graph shows self-rated fear on a 0 to 10 scale for two groups at three points. Before therapy both groups scored around 8, so they began at a similar level. After therapy the VR group had fallen to roughly 4, while the waiting list group remained above 7. At the three-month point the VR group was still below 4 and the waiting list group had barely changed.
Interpret it
The condition that performed better is clearly the VR therapy group, and the flat waiting list line matters just as much: it suggests that time alone did not reduce fear. The gap also did not close at follow-up, which points to a maintained rather than a temporary effect.
Conclusion, with a caution
The findings suggest VR exposure therapy reduced self-reported fear and that the reduction held for at least three months. However, the graph shows no sample size, no spread and no test of significance, so the conclusion should be treated as a promising pattern rather than proof that the therapy caused the change.
💡 Exam tip
Name the axis and the scale in your first sentence. It shows the examiner you actually read the graph.
Always say which condition did better. It is an easy mark and students skip it constantly.
Use the comparison group as evidence, not decoration. A flat control line is a strong argument.
Limit yourself to the data. Do not speculate about what the therapy involved or who the participants were.
Match the answer to the marks: 3 marks means one developed point; 6 marks means describe, interpret, conclude.
Keep three stock limitations ready — subjective measure, no sample details, no significance test. One of them fits almost any graph.
⚠ Common mix-up
Criticising the therapy instead of the data. “VR is expensive” is not a limitation of the graph.
Listing every limitation you know. The question asked for one, and only one will be marked.
Inventing missing detail. If the sample is not stated, say so — do not guess it.
Treating a difference as proof of cause. A gap between bars shows a difference, not a mechanism.
Ignoring the follow-up bars. They are usually where the interesting finding is hiding.
Writing a paragraph of praise. If the question asks what limits interpretation, strengths earn nothing.
Up next: Technology and the Developing Self — the argument about whether growing up with screens builds a sharper mind or a lazier one.
Want this explained one-to-one?
Book a free session with an experienced IB Psychology tutor and get your trickiest topics made simple.
