You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
The Edexcel application topics culminate in two pieces of work that ask students to use the psychology they have learned rather than simply recall it: a key question of societal relevance, analysed with the topic's concepts and theories, and a practical investigation, in which students design and reason about their own small-scale, ethical study. This lesson develops both for Health Psychology. Part (a) takes the key question "How can psychology be used to reduce the harm caused by substance misuse?" — a genuine issue of contemporary importance — and analyses it using the biological and learning explanations, the treatments, the persuasion literature and the behaviour-change models covered across the topic. Part (b) then models a practical investigation: a correlational study of the relationship between an attitude/risk factor and self-reported substance use, worked through from hypothesis and design to sampling, ethics (with the heightened sensitivity that substance-misuse research demands) and analysis, including a named descriptive statistic and a named inferential test (Spearman's rho, with Mann–Whitney as the alternative for a two-group design). Throughout, substance misuse is treated objectively and sensitively, in the standard academic register expected at A-Level.
Key Definition: A key question in the Edexcel specification is a real, societally relevant issue examined through the lens of the topic's psychology; a practical investigation is a small-scale, ethical study designed and analysed by the student to develop research-methods skills in the context of the topic.
This lesson addresses the Edexcel 9PS0 — Paper 2, Topic 8: Health Psychology requirements to analyse a key question of societal relevance using concepts, theories and research from the topic, and to design and reason about a practical investigation appropriate to the topic (including hypotheses, design, sampling, ethics, and descriptive and inferential analysis). In assessment-objective terms, you should be able to describe the chosen key question and relevant methods (AO1), apply topic concepts to the key question and to the design of a practical investigation — selecting appropriate design, sampling and statistics (AO2), and evaluate the psychology's contribution to the key question and the strengths, weaknesses and ethics of the practical investigation (AO3).
Connects to…
Substance misuse — smoking, harmful drinking, and behavioural addictions such as gambling — imposes very large costs on individuals, families, health services and society. It is strongly patterned by circumstance (it tends to concentrate in more disadvantaged groups, deepening health inequalities), it is frequently accompanied by stigma, and it is notoriously resistant to simple exhortation to "just stop". Because psychology offers explanations of why addiction develops and persists, and evidence-based tools for prevention and treatment, the question of how psychology can reduce the harm is both socially urgent and squarely within the topic — which is exactly what makes it a strong key question.
The value of the topic's psychology is that it supplies a layered answer: harm can be reduced at the level of the brain, of learning, of the individual's beliefs, and of the population.
| Level | Psychology from the topic | How it reduces harm |
|---|---|---|
| Biological | Dopamine reward pathway; neuroadaptation; tolerance/withdrawal | Explains why willpower alone fails and justifies drug therapies (e.g. nicotine replacement, substitute and antagonist drugs) that manage withdrawal and craving |
| Learning | Classical conditioning of cues; operant reinforcement; social learning | Justifies cue-exposure/avoidance, contingency and aversion approaches, and relapse prevention; explains uptake via modelling |
| Individual cognition | Health Belief Model; Theory of Planned Behaviour; stages of change | Guides stage-matched, belief-targeted interventions that raise readiness, reduce barriers and build self-efficacy/control |
| Population / persuasion | Hovland–Yale; fear-appeal + efficacy; boomerang effect | Designs effective campaigns and, combined with structural measures, shifts norms and reduces uptake |
The key argument is that psychology reduces harm most effectively when these levels are combined. No single tool is sufficient: drug therapy manages the biology but not the conditioned cues or the social context; a persuasive campaign may shift attitudes but not craving; a behaviour-change model may build intention but leave habit untouched. The topic's own evidence makes this concrete — the fear-appeal research shows that frightening people without offering efficacy can backfire; the stages-of-change model shows that help mistimed to a person's readiness fails; and the history of tobacco control shows that sustained harm reduction came from combining persuasion with structural measures (pricing, advertising restrictions, smoke-free laws). A psychologically informed answer therefore argues for multi-level, evidence-based, well-timed intervention rather than any single "magic bullet".
Two worked applications make the multi-level argument concrete. Consider, first, helping an established smoker to quit. The biological explanation tells us nicotine dependence involves genuine neuroadaptation, so a purely psychological approach that ignores withdrawal is likely to fail — which is why nicotine replacement (managing the biology) improves success rates. But biology is not enough: the learning account identifies conditioned cues (the coffee, the break, stress) that trigger craving and relapse, so relapse-prevention skills and cue-management are needed; the Theory of Planned Behaviour highlights building perceived control and self-efficacy, especially after previous failed attempts; and the stages-of-change model warns that all of this must be timed to the smoker's readiness — offering a quit plan to a precontemplator will fail. The most effective help therefore layers the tools: pharmacological support for the biology, cue-management for the learning, control-building for the cognition, and correct timing for the process.
Consider, second, reducing uptake of vaping among teenagers — a prevention rather than a treatment problem. Here the population-level psychology dominates: the Hovland–Yale model tells us to use credible, similar (peer) sources and to tailor the message to a teenage audience; the fear-appeal literature warns that graphic threat without efficacy risks defensive avoidance, and the boomerang effect warns that a heavy-handed or "everyone's doing it" message can provoke reactance or inadvertently normalise vaping. Social-learning theory adds that teenagers model peers and media figures, so denormalisation and positive-norm messaging matter. And because attitudes do not automatically become behaviour, campaigns work best combined with structural measures (restricting availability and marketing). The same lesson recurs: psychology sharpens each lever, but harm falls furthest when the levers are combined with policy.
Psychology's contribution is genuine but partial and contested, and a strong answer says so. Its strengths are clear: it explains why addiction resists willpower (reducing stigma and justifying treatment), and it generates interventions — pharmacological treatments derived from the reward account, stage-matched behaviour-change programmes, efficacy-paired campaigns — that demonstrably improve outcomes. But there are real limits. The attitude–behaviour gap means changing beliefs does not guarantee changing behaviour; interventions built on rational cognition under-weight the craving and habit that drive entrenched addiction; and psychology alone cannot alter the structural drivers (poverty, availability, marketing) that shape substance misuse at the population level. There is also an ethical dimension: using psychology to change behaviour shades into social control, raising a genuine debate about the balance between protecting health and respecting individual freedom — some interventions (information, support) clearly respect autonomy, while others (heavy nudging, coercive framing) are more contested. The defensible conclusion is that psychology is a necessary but not sufficient contributor: it reduces harm most where its multi-level tools are combined with structural policy and applied with attention to ethics and autonomy.
The second component models a practical investigation appropriate to Health Psychology. Because researching substance misuse directly raises acute ethical and practical difficulties, a well-judged student investigation typically studies an attitude, belief or risk factor related to a health behaviour rather than trying to manipulate or elicit the behaviour itself. The worked example below is a correlational study; a two-group experimental/comparison variant is then outlined to motivate the alternative inferential test.
Why a correlational design here? It is worth being explicit that the choice of design is driven by the research question and by ethics, not by convenience. The question — whether perceived control is associated with reported use — is naturally a question about a relationship between two variables that already vary in the population, which is precisely what a correlational design measures. Crucially, the obvious experimental alternative is not ethically available: to test causation one would have to manipulate teenagers' substance use (or actively try to change their control and then measure resulting use), which no ethical committee would permit for a harmful behaviour in young people. Studying naturally occurring variation therefore sidesteps the ethical impossibility of manipulating the behaviour, at the known cost that correlation cannot establish causation. Recognising this trade-off — that the design is chosen partly because the more powerful experimental design is ethically ruled out — is exactly the kind of methodological reasoning that distinguishes a strong practical investigation from a merely competent one.
Aim: to investigate whether there is a relationship between a person's perceived behavioural control over avoiding a health-risk behaviour and their self-reported engagement in that behaviour — for example, between students' rated confidence that they could resist peer pressure to vape (a control/self-efficacy measure drawn from the Theory of Planned Behaviour) and their self-reported frequency of vaping.
The hypotheses are stated in terms of a correlation:
Key Definition: A correlational design measures two variables (without manipulating either) to see whether they co-vary. It can establish that a relationship exists and its direction and strength, but — because neither variable is controlled — it cannot establish that one causes the other.
For the two-group comparison variant: instead of correlating, the study could compare self-reported use between two independent groups — for example, those high versus low in perceived control (a median split) — giving an independent-groups design with one IV (control group: high/low) and one DV (self-reported use), analysed with Mann–Whitney U.
Substance-misuse research carries heightened ethical sensitivity, and a strong practical investigation foregrounds this. Beyond the standard principles, specific safeguards apply:
| Principle | Application in this sensitive context |
|---|---|
| Informed consent | Full information about the study's purpose and the nature of the questions; for participants under 18, appropriate consent arrangements (e.g. parental/guardian consent and institutional approval) |
| Right to withdraw | Emphasised strongly, including the right to withdraw data afterwards — important where questions touch on personal, possibly risky behaviour |
| Confidentiality / anonymity | Anonymous responses are essential: disclosing substance use must never be traceable to an individual, which also improves honesty |
| Protection from harm | Non-judgemental wording; awareness that questions may cause discomfort or disclose a problem; signposting to support (e.g. a helpline or school counsellor) provided to all participants |
| Confronting participants / safeguarding | If responses suggest a participant is at risk, the researcher must have a clear, ethical safeguarding route — an issue a teacher must oversee |
| Debriefing | Full debrief explaining the study and reiterating support information |
The deliberate choice to study attitudes/control rather than to manipulate substance use is itself an ethical decision: it avoids inducing or normalising a harmful behaviour, keeping the study within the bounds appropriate for a student investigation.
A subtle but important point is that, in this topic, good ethics and good validity actually reinforce one another. Guaranteeing anonymity is an ethical requirement (protecting participants who disclose a stigmatised behaviour), but it is also a validity measure: participants who trust that their answers cannot be traced to them are far less prone to social-desirability bias and so report their behaviour more honestly. Similarly, non-judgemental wording protects participants from discomfort and reduces the pressure to under-report. This alignment is worth stating in an evaluation, because it shows that the ethical safeguards demanded by a sensitive topic are not merely a constraint on the research but part of what makes its self-report data credible in the first place.
Descriptive statistics. Because the data are ordinal, the appropriate measure of central tendency is the median (not the mean), reported for each variable, with the range or interquartile range as a measure of dispersion. A scattergraph should be plotted, with perceived control on one axis and self-reported use on the other; a downward-sloping cloud of points would visually indicate the predicted negative relationship, and the scattergraph also reveals any outliers or non-linearity.
Inferential statistic — Spearman's rho. The relationship is tested with Spearman's rank-order correlation coefficient (ρ), the appropriate non-parametric test because the design is correlational, the data are ordinal, and normality cannot be assumed. Spearman's ρ ranks the two sets of scores and correlates the ranks, yielding a coefficient between −1 and +1:
ρ=1−n(n2−1)6∑d2
where d is the difference between the ranks of each pair and n is the number of pairs. A value near −1 indicates a strong negative relationship (supporting H₁), near 0 no relationship, and near +1 a strong positive relationship. The decision rule compares the calculated coefficient with the critical value for the given n, the chosen significance level (conventionally p≤0.05) and the one-tailed test: for Spearman's ρ, the result is significant when the calculated value is greater than or equal to the critical value. If it is, H₀ is rejected in favour of H₁ — there is a statistically significant negative correlation between perceived control and self-reported use.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.