IB Psychology SLTopic 2 — Health ProblemsPaper 1 & 2Research methods~9 min read
How Common Is Obesity?
Prevalence sounds like the easy part of this option — just a number, just counting people. It is not. Before anyone can count, someone has to decide what counts, and the tool used to decide has real flaws. Getting that argument right is where the marks are.
📚 What you need to know
Prevalence shows how common a disorder is within a defined population over a set time period.
It is calculated as the number of cases divided by the total population.
Prevalence tells you the likelihood of any one person being affected, which makes it a useful predictive tool.
Obesity is defined as having a Body Mass Index (BMI) over 30.
UK figures for 2023/24: 26.5% of adults obese, 64.5% overweight or obese combined.
Kyle et al. (2016) is your study, and the comparison in it is more interesting than the headline.
The concept links are responsibility (socially sensitive research) and measurement (BMI).
Working out a prevalence rate
The calculation itself is simple arithmetic. Learn it so you can do it under exam pressure without hesitating.
Prevalence rate
number of cases ÷ total population × 100
So if 4,500 people in a population of 50,000 are obese, the prevalence rate is 9%. That number then has a real job: it tells a government how many hospital beds, clinics and prevention programmes it is going to need.
This topic is about populations, not people. Prevalence data describes groups. It says nothing about any individual, and it should never be used to judge one. Research in this area is genuinely socially sensitive, and if anything on this page feels close to home, talking to your doctor or someone you trust is a far better source of guidance than a statistic.
Kyle et al. (2016)
🔬 Kyle et al. (2016)
Comparing obesity across professions in Scotland
AIM
To investigate obesity prevalence among nurses in Scotland compared with other professions.
PARTICIPANTS
13,483 adults aged 17 to 65: 411 nurses, 320 other healthcare professionals, 685 care assistants, and 12,067 people working outside healthcare.
PROCEDURE
Participants’ BMI was measured directly, rather than being self-reported.
RESULTS
69.1% of nurses fell into the reported category, compared with 51.3% of other healthcare professionals, 68.5% of care assistants and 68.9% of non-healthcare workers.
CONCLUSION
The researchers concluded that prevalence among Scottish nurses was significantly higher than in the other healthcare and non-healthcare groups.
Draw the bars before you read the conclusion. Three of the four are effectively level, and the group that stands apart is the one with the lowest figure.
This is the best evaluation point on the page and almost nobody makes it. The conclusion says nurses were “significantly higher” than everyone else — but 69.1% and 68.9% are not meaningfully different. Look again and the finding is really that other healthcare professionals were unusually low. Always check whether a study’s conclusion is actually supported by its own numbers.
The problem with BMI
BMI is calculated from height and weight only. That makes it cheap, fast and standardised, which is exactly why it is used for whole populations. It also means it is blind to several things that matter.
It does not consider muscle mass, so a very muscular athlete can be classified as obese.
It does not consider bone density.
It does not account for racial or gender differences in body composition.
So BMI cannot identify body fat with 100% accuracy and is not a “one size fits all” measure. Any prevalence figure built on it inherits that imprecision — and because the cut-off is a hard line at 30, people just either side of it get sorted into different categories on a very small difference.
Evaluating prevalence research
Strengths
Limitations
Prevalence data is extremely useful for public health planning, since obesity is linked to serious health risks.
Knowing prevalence is not the same as addressing it — governments may still struggle to deliver effective interventions.
Kyle used a large sample, producing robust quantitative data that is more reliable and generalisable.
Prevalence data lacks depth. It does not explain why anyone is obese.
BMI was measured directly rather than self-reported, removing a major source of error.
More qualitative research is needed to understand the complex causes — lifestyle, socioeconomic status, culture.
Comparing occupational groups within one country controls for national differences.
The nurse group was small (411) compared with the non-healthcare group (12,067).
Linking to the concepts
Responsibility: research into obesity is socially sensitive and can feed a “blame” culture. Reporting these results gives the media opportunities to attribute negative behaviours such as laziness or greed to people who are obese, without reflecting on the factors that contributed.
Many complex variables contribute — socioeconomic status, stress, availability of cheap processed food, education level — so obesity must be considered in the context of a person’s whole life rather than as a “lifestyle choice”.
Measurement: BMI is neither the best nor the only way to measure obesity, and building national policy on a single imperfect number carries risks.
EXAM ANSWER
Discuss the prevalence rates of one health problem. [22 marks]
Define, then show you can use it
Prevalence = cases divided by population. 4,500 in 50,000 = 9%. Obesity is defined as BMI over 30.
Give real figures
UK 2023/24: 26.5% of adults obese, 64.5% overweight or obese combined.
Bring in the study, then interrogate it
Kyle et al. reported 69.1% for nurses, but non-healthcare workers were at 68.9%. The claim of a significant difference is weak.
Attack the measure
BMI ignores muscle mass, bone density and racial or gender differences, so every figure above carries that error.
Conclude: useful for planning, unreliable for blamingcriticising the measuring tool is the highest-level move available here
💡 Exam tips
Show the calculation with the 4,500 in 50,000 example. It proves you understand rather than memorise.
Quote the UK figures with the year attached. Undated statistics look vague.
Always evaluate BMI. It is the single most reliable criticism in this whole sub-topic.
Use the phrase “descriptive, not explanatory” for prevalence data.
The social sensitivity point works for obesity, smoking and mental health — three topics, one argument.
Directly measured BMI is a methodological strength worth naming, since self-reported weight is notoriously inaccurate.
⚠ Common mix-ups
Confusing prevalence with incidence. Prevalence = all existing cases now. Incidence = new cases in a period.
Treating BMI as a measure of health. It is a rough proxy for body composition, nothing more.
Adding 26.5% and 64.5% together. The 64.5% figure already includes the obese group.
Saying prevalence explains causes. It describes how many, never why.
Accepting a study’s conclusion without checking its numbers. Kyle is the perfect example.
Using judgemental language. Write about it as a health outcome shaped by many variables, not a personal failing.
Up next: How Common Is Smoking? — same concept, different problem. Smoking is easier to define than obesity, but far harder to pin down in time, which is where the three types of prevalence come in.
Want this explained one-to-one?
Book a free session with an experienced IB Psychology tutor and get your trickiest topics made simple.
