Chapter 5: Psychological Measurement PDF

Chapter 5: Psychological Measurement What is measurement? The assignment of scores that represent some characteristic of the individual In Psychology, the discipline of Psychometrics concerns the measurement of psychological phenomena e.g. Working memory and digit span ...

Chapter 5: Psychological Measurement What is measurement? The assignment of scores that represent some characteristic of the individual In Psychology, the discipline of Psychometrics concerns the measurement of psychological phenomena e.g. Working memory and digit span  Psychological tests have 3 defining characteristics 1. To sample behaviour (or to collect a systematic sample of behaviour) Features of E.g. writing sample (short essays, psychologi letters, memos) cal tests 2. The sample is obtained under standardized conditions 3. There are established rules for scoring, or for obtaining quantitative (numeric) information from the behaviour sample Psychological Constructs Definition: -represent tendencies on how people think, feel, and behave across a variety of situations; summary of complex set of behavioural and internal processes 2 essential properties: 1. Abstract summaries of some phenomenon in nature (behaviour or internal process) Psychological Constructs (cont’d) -Examples: neuroticism (construct); related to negative emotions (anxiety, anger, sadness) across a variety of situations  observed indirectly; often involve internal processes  E.g.fear: elicits central and peripheral processes; shown in thoughts, feelings, and behaviours (may or may not be observable) Psychological constructs (cont’d) Developing a definition of a construct involves: -proposing conceptual definitions -testing them empirically -revising as necessary Example of modern or not-well-understood constructs? -internet addiction? -social media anxiety disorder? Unique from Operational Definitions  Specifications on how a construct or variable is to be measured 3 broad categories; 1) Self-report e.g. 2) Behavioural e.g. 3) Physiological e.g. Operational Definitions Converging operations: extent to which different and closely related operational definitions produce a similar pattern of results e.g., Stress Levels of Measurement Or scales of measurement 1. Nominal 2. Ordinal 3. Interval 4. Ratio Levels of Measurement Reliability and Validity of Measurement Reliability: Refers to the consistency of a measure 3 types: 1) Test-retest 2) Internal consistency 3) Inter-rater Reliability: Test-retest Test-retest: Extent to which the measurement of a construct is consistent across time -usually measurements are made on the same people -assessed using test-retest correlation (Pearson’s r) between two sets of scores Size of Correlation coefficient r Interpretation r ≥.80 Excellent.70 ≤ r <.80 Good.60 ≤ r <.70 Adequate r <.60 Poor Reliability: Test-retest Test-retest Pearson’s r = +.95 Reliability: Internal Consistency Internal consistency: Extent to which the measurement of a construct is similar across multiple items of a measurement Split-half correlation: assessing whether scores on two subsets of scores are in agreement (e.g. first and second halves; odd and even-numbered items) Assessed by Cronbach’s α Reliability: Interrater Reliability Interrater reliability: extent to which different observers are consistent in their judgments Assessed using Cronbach’s α (quantitative measurements) or Cohen’s κ (categorical measurements) Validity Validity: extent to which the scores from a measurement represent the construct that it intends to measure Types of validity: 1. Face 2. Content 3. Criterion 4. Discriminant Validity: Face  Theextent to which a measurement appears “on the surface” to measure the construct of interest  Good face validity: it is obvious what the questions or measurement is intended to measure Examples: “I believe in the second coming of Christ.” Validity: Content  Thedegree to which a measurement is comprehensive in measuring the construct of interest E.g. Aggression: emotional, physiological, and behaviour dimensions Validity: Criterion  Extent to which scores on a measurement are correlated with other variables (criteria) of the construct of interest e.g. Validity: Criterion 3 types: 1) Concurrent: criterion is measure at same time as construct e.g. 2) Predictive: criterion is measured at some point the future or predict future outcome e.g. Validity: Criterion 3) Convergent: degree to which multiple measures of the construct agree with each other e.g. Validity: Discriminant  Extent to which scores on a measure are not correlated with measures of a variable that are conceptually distinct from it i.e. measurements of conceptually distinct constructs  E.g. self-esteem and mood Threats to reliability and validity Reactivity of participants or experimenters in experimental conditions may affect reliability and validity of results in unintended ways: 1) Socially desirable responding: participants respond or behave in ways in order to be viewed favourably by others E.g Threats to reliability and validity 2) Demand characteristics: researchers provide subtle cues that reveal how participants should behave E.g.