PSY 311 - Week 5

Behavioral Measure

method of data collection involving the direct observation and recording a participant’s actions

Construct Validity

extent to which a measurement tool accurately captures the theoretical construct it is intended to measure

Content Validity

subjective assessment of whether a measure represents all facets of a defined construct

Convergent Validity

an empirical check to see if a measure correlates strongly with other measures of the same or similar constructs

Criterion Validity

degree to which a measure is correlated with a relevant, real-world behavioral outcome

Discriminant Validity

empirical check ensuring a measure is not overly correlated with measures of different constructs, proving its distinctness

Face Validity

subjective assessment of whether a measure appears, on the surface, to measure what it claims to measure

Internal Reliability

degree to which various items within a single scale or test consistently measure the same construct

Operationalization

process of turning a conceptual variable into a specific, measurable value or procedure

Test-Retest Reliability

consistency of scores when the same measure is administered to the same people at different times

Measurement Process

1. Operationalize the Variable

2. Test for Reliability

3. Test for Validity

Types of Measurements

• Self-Report Measures

• Behavioral Measures

• Physical (Physiological) Measures

Subjective Assessments

• Face Validity

• Content Validity

Empirical Assessments

• Criterion Validity

• Convergent Validity

• Discriminant Validity

Positive Relationships/Correlation

when 2 variables are moving in the same direction - either up or down

Negative Relationships/Correlation

when 2 variables are moving in the opposite direction - one goes up and the other goes down

Curvilinear Relationships/Correlation

when variables don’t follow a consistent increasing or decreasing pattern; the pattern is a U shape

Zero Relationship/Correlation

when no linear relationship between two variables exists

Scatterplots

shows both the direction + strength of the relationship between variables

Coefficient of Determination (r²)

tells you how well one variable predicts the other variable

Statistical Significance

the probability of a result this extreme (or more) if the null is true

Correlation does

NOT EQUAL causation

3 Requirements for Clausal Claim

1. Temporal Precedence

2. Covariation between the two variables

3. Eliminating plausible alternative explanations

Temporal Precedence

one variable must come before the other

Reverse Causation

causal direction is actually the opposite