Measurement, Reliability and Validity

These flashcards cover key concepts related to measurement, reliability, and validity in psychology, including types of measures, validity types, survey methods, response biases, and observational research.

What are the three common types of psychological measures?

Self-report, observational, and physiological measures.

What are self-report measures used for?

To assess subjective experiences that can’t be directly observed.

What are the strengths and weaknesses of self-report measures?

Strengths: Access to internal states. Weaknesses: Prone to bias, social desirability, and limited self-insight.

What are observational measures?

Direct recordings of behavior by watching participants.

What are the strengths and weaknesses of observational measures?

Strengths: Capture real behavior, avoid self-report bias. Weaknesses: Cannot access internal states; possible observer bias.

What are physiological measures?

Biological data used to infer psychological states.

What are the pros and cons of physiological measures?

Pros: Objective. Cons: Costly, may need interpretation.

What are the two main types of measurement scales?

Categorical (nominal) and quantitative.

What are the three types of quantitative scales?

Ordinal: Rank order without equal intervals; Interval: Equal intervals, no true zero; Ratio: Equal intervals with true zero.

What is reliability?

The consistency of a measure across time, items, or observers.

What are the three types of reliability?

Test–retest, inter-rater, and internal reliability.

What is validity?

The accuracy of a measure—whether it captures what it claims to measure.

How are reliability and validity related?

Reliability is necessary but not sufficient for validity.

What is face validity?

Whether a test appears to measure what it claims to.

What is content validity?

Whether all relevant parts of the construct are included.

What is criterion validity?

Whether the measure predicts relevant real-world outcomes.

What is convergent validity?

Correlates with related constructs.

What is discriminant validity?

Does not correlate with unrelated constructs.

What is deficient validity?

Missing key elements.

What is contaminated validity?

Includes irrelevant elements.

What determines the construct validity of a survey?

The clarity, neutrality, and order of its questions.

What is the purpose of Likert and semantic differential scales?

To measure attitudes or agreement along graded continua.

What are response sets (non-differentiation)?

Answering all items in a similar way without thoughtful engagement.

What is acquiescence and how is it controlled?

Agreeing with everything; controlled with reverse-worded items.

What is fence-sitting and how is it reduced?

Always choosing neutral options; reduced by removing middle choices.

What is socially desirable responding and how is it mitigated?

Responding to appear favorable; reduced via anonymity or control items.

What does 'reporting more than they can know' mean?

People give explanations for behavior even when unaware of true causes.

What are memory errors and flashbulb memories?

Memory distortions; vivid flashbulb memories feel accurate but often degrade.

What are major threats to construct validity in observation?

Observer bias, expectancy effects, and reactivity.

How can observer bias and expectancy effects be reduced?

Use blind observers, detailed codebooks, and multiple raters.

What is reactivity and how can it be minimized?

Participants change behavior when watched; minimized by blending in.

When is observational research ethical without consent?

When behavior occurs in public spaces and participants expect to be observed.