Preview
IB DP Psychology — Research Methods | SL & HL | Paper 3
🎯 Learning Intentions By the end of this note, you will be able to:
- Define internal validity, external validity, ecological validity, and construct validity using IB-aligned terminology.
- Explain how confounding variables threaten internal validity and how controls are used to address this.
- Distinguish between different types of validity and explain why each matters for interpreting research findings.
- Describe the role of experimental controls — including random allocation, standardisation, and control groups — in strengthening causal claims.
- Evaluate the validity of research studies using supporting and opposing evidence in ACAS format.
- Apply concepts of validity and control to a real-world research scenario.
- Analyse research designs using the concept areas of Causality, Measurement, and Bias.
Introduction & Context
Imagine a study claims that students who listen to classical music before an exam score higher than those who do not. The headline sounds compelling — but before accepting this conclusion, a careful thinker would ask: were the two groups similar to begin with? Did both groups sit the same exam under the same conditions? Were the researchers aware of which students had listened to music? These questions are not pedantic — they go to the heart of whether the study actually tells us what it claims to tell us.
This is the problem of validity. In psychological research, a finding is only meaningful if the study was designed in a way that genuinely measures what it set out to measure, and if the conclusions drawn are actually supported by the data. Researchers can collect mountains of data and still reach entirely misleading conclusions if their design contains hidden flaws. Understanding validity — and the controls used to protect it — is therefore not just a technical skill; it is the foundation of scientific reasoning in psychology.
🔍 Analogy — The Leaky Bucket Think of a research study as a bucket designed to collect water (evidence). Validity is about whether the bucket actually holds water — whether what you collect is what you intended to collect. Confounding variables are holes in the bucket: they let other things in (or let your intended evidence leak out) so that by the time you examine what you have collected, you cannot be sure it represents what you think it does. Controls are the patches you apply to seal those holes before the study begins.
Core Concepts
Before we can evaluate whether a study's conclusions are trustworthy, we need to ask a more fundamental question: did the study actually measure what it claimed to measure? This is the question of construct validity — and it is the starting point for understanding all other forms of validity in psychological research.
📖 Key Definition — Construct Validity Construct validity refers to the degree to which a test, measure, or procedure actually measures the psychological construct it is intended to measure. A construct is an abstract concept — such as intelligence, anxiety, or memory — that cannot be directly observed but must be operationalised (translated into measurable form).
Consider the concept of "stress." Stress is not a thing you can see or touch — it is a theoretical construct. To study it, a researcher must decide how to operationalise it: perhaps through self-report questionnaires, cortisol levels in saliva, or heart rate measurements. Each operationalisation captures a different aspect of stress, and none is perfect. If a researcher uses a questionnaire that only asks about work-related worries, they may be measuring occupational stress rather than general stress — meaning their