Validity of Qualitative & Quantitative Methods (OCR GCSE Psychology): Revision Note

Validity

• Validity refers to how true or accurate a piece of research is — in other words, whether it measures what it claims to measure

• A valid study reflects real behaviour, thoughts, and experiences rather than artificial or biased results

• Reliability and validity are linked

  • A study can be reliable (consistent) without being valid (true)

    • E.g. if a test produces the same results repeatedly but measures the wrong thing, it is reliable but not valid

Validity in psychological research

• Validity ensures that findings are realistic, accurate, and generalisable beyond the research setting

• Different research methods vary in their level of validity:

  • Lab experiments often have low ecological validity because they use artificial settings or tasks

  • Field experiments and natural experiments usually have higher validity, as they take place in real-world environments

  • Qualitative methods (e.g. interviews, case studies) tend to have high validity because they explore behaviour in depth and capture real experiences.

• To maintain validity, researchers use controls such as random allocation and counterbalancing to reduce bias and extraneous variables

Types of validity

Ecological validity

• Ecological validity refers to how far the results of a study reflect real-life settings

• If participants behave differently in a lab than they would in everyday life, ecological validity is low

  • E.g. testing memory in an artificial environment may not represent how memory works in daily situations

• Researchers can improve ecological validity by

  • conducting studies in more natural environments

  • using realistic tasks or materials that mirror real-life behaviour

Population validity

• Population validity measures how well the sample represents the target population

• If a study uses a narrow or biased sample (e.g. only university students), it may not generalise to wider groups such as older adults or people from other cultures

• Representative sampling methods (like random sampling) improve population validity

Construct validity

• Construct validity concerns how accurately a variable is measured compared to the overall concept

  • E.g. an “aggression test” that only measures how often a person shouts is not a valid measure of aggression as a whole

• High construct validity means that the test or measure fully captures the behaviour or concept being studied

Threats to validity

Demand characteristics

• Demand characteristics occur when participants guess the aim of the study and change their behaviour

  • E.g. they might try to please the researcher or behave in a way they think is expected

  • This makes responses artificial and reduces validity

• Researchers can minimise demand characteristics by:

  • using single-blind procedures where participants don’t know which condition they are in

  • keeping instructions neutral and avoiding cues that reveal the hypothesis

Social desirability bias

• Social desirability bias occurs when participants alter their answers to appear more socially acceptable

  • E.g. they may exaggerate positive behaviours (like being kind or honest) or hide negative ones (like aggression or prejudice)

  • This often affects self-report methods such as questionnaires and interviews

• Researchers can reduce social desirability by:

  • using anonymous questionnaires.

  • including unambiguous questions

  • reminding participants that there are no right or wrong answers

Improving validity

• Researchers can improve validity by:

  • using standardised instructions and procedures

  • selecting representative samples to generalise findings

  • conducting pilot studies to identify and fix validity issues