Psychological Measurement

What is measurement in psychology?

• Measurement is assigning values from an instrument to individuals that represent a characteristic

• Psychometrics refers to the study of measuring psychological variables and constructs

What is a construct?

• Constructs refer to variables that are not directly observable but can be measured using instrumentation

• Examples of constructs

• Personality traits, emotional states, attitudes, learning, thinking patterns

• Operational definitions

• Method of ascribing a value (quantitative) or description (qualitative) to the measurement of a construct

• Quant: stress → Using the scores of Perceived Stress Scale (PSS)

• Qual: stress → using thematic analysis to differentiate coping strategies

Operational Definitions

• Operationalization refers to providing a definition of a variable or construct in terms of how it will be measured in your study

• Types of measures:

• Self-report: participants report their own thoughts, feelings, and behaviors

• Behavioral: researcher observes and records behavioral patterns

• Physiological: research measures, or records, a physiological process or reaction

Levels of Measurement

Nominal (categorical)

Ordinal (rank-order)

Interval

Ratio

Levels of Measurement: Nominal (Categorical)

Variables that consist of only categories with no inherent numerical value

Ex: Gender, sex, religion

Levels of Measurement: Ordinal (Rank-order)

Variables that consist of categories with a numerical rank

Ex: Class standing, finishing a race, letter grade

Levels of Measurement: Interval

Numeric variable that has equal distance between scores; 0 is arbitrary

Ex: GPA, depression inventory,

Levels of Measurement:

Interval scale with an absolute zero

Ex: Height, weight

Reliability

the extent to which scores on a measure are consistent across time and within items on a measure

• Test-retest reliability: across time

• Internal consistency: Cronbach’s Alpha

• Interrater reliability: comparison of raters’ scores

Validity

the text to which scores on a measure actually represent the variable or construct intended

• Face validity – does it look valid

• Content validity – does it measure all aspects of the construct its measuring?

• Criterion validity – how related are the scores on the measure to the scores on another related measure?

• Discriminant validity – how dissimilar is the measure compared to other constructs’ measures?