Eduqas A Level Psychology Reliability and Validity Exam Questions

About Eduqas A Level Psychology Reliability and Validity

Reliability and validity are central to evaluating any study in Component 2. Reliability is about consistency; validity is about whether a study measures what it claims to and whether its findings generalise. You need the types of each, the threats to validity such as demand characteristics and researcher bias, and the practical ways of assessing and improving both. Examiners often describe a study and ask how you could check its reliability or improve its validity, so give concrete steps such as correlating two observers' tallies or using a double-blind procedure. These ideas also help you evaluate the classic studies and your own investigations.

What you need to know

• Internal and external reliability Internal reliability is the consistency of a measure within itself, such as items in a questionnaire. External reliability is consistency over time or across researchers.
• Assessing reliability Inter-rater: two observers record independently and their scores are correlated (often +0.8 or above is acceptable). Test-retest: the same people take the test twice. Split-half: scores on two halves of a test are compared.
• Internal and external validity Internal validity is whether the study measured the effect of the IV and not other variables. External validity is whether findings generalise to other settings (ecological), people (population) and times (temporal).
• Threats to validity Researcher bias: expectations influence how data are collected or interpreted. Demand characteristics: participants guess the aim and change behaviour. Social desirability: participants answer in a way that makes them look good.
• Dealing with threats Use single-blind and double-blind procedures, standardised instructions, deception where ethical, and anonymous questionnaires.
• Assessing validity Face validity: it looks like it measures the right thing. Content validity: experts judge whether all aspects are covered. Concurrent validity: results match an established measure. Predictive validity: scores predict future behaviour. Construct validity: it measures the underlying theoretical concept.

Common mistakes in Reliability and Validity

• Confusing reliability with validity. A measure can be reliable (consistent) but still not valid, for example a scale that always reads 2 kg too heavy.
• Describing inter-rater reliability without the correlation step. Observers' scores must be compared statistically, not just discussed.
• Calling all lab studies invalid. Lab studies can have high internal validity because of control; the concern is usually ecological validity.
• Mixing up concurrent and predictive validity. Concurrent compares with an existing measure now; predictive compares with an outcome later.
• Saying a double-blind procedure means participants are deceived. It means neither participants nor the person collecting the data know the condition or hypothesis.

Try three Reliability and Validity exam-style questions