Week 3: Measurement, Variables, and Unit of Analysis

Vocabulary flashcards covering key concepts from Week 3 notes on measurement, variables, and unit of analysis.

Measurement

The process of turning abstract ideas into objectively measurable variables by developing a system to quantify them.

Construct

An idea with multiple conceptual elements, typically subjective and not directly empirical.

Concept

An abstract element of a notion that is not yet operational for observations.

Operationalization

The process of converting a concept into a measurable variable.

Variable

Something that varies and is measurable, used by researchers to collect observations.

Value label

A category used to label the possible values of a variable (e.g., Strongly agree, Agree, etc.).

Unit of Analysis (UofA)

The entity the variable measures or collects information about (e.g., individuals, neighborhoods, cities, states).

Measurement strategies

Methods used to collect data at different units of analysis (e.g., surveys for individuals; rates for neighborhoods, cities/counties, and states).

Nominal variable

A variable with categories that have names but no inherent order or distance (e.g., gender, race).

Ordinal variable

A variable with ordered categories but no meaningful numeric distance (e.g., Likert scales).

Interval/Ratio variable

Variables with ordered categories and meaningful distances; ratio implies a true zero (e.g., rates, age in years).

Collectively exhaustive

All potential value labels are accounted for by the variable (often including an 'all others' category).

Mutually exclusive

No observation can be categorized into more than one value label; categories do not overlap.

Index

A score created by combining multiple variables to measure a broader concept (e.g., self-control).

Independent variable (IV)

The presumed cause in a cause-and-effect relationship (the variable that influences the outcome).

Dependent variable (DV)

The outcome in a cause-and-effect relationship (the variable that is affected).

Validity

The extent to which a variable accurately represents the concept it is intended to measure.

Reliability

The consistency of measurements across time, across researchers (interrater reliability), and across contexts.

Rate

An interval/ratio measure of incidents per population, allowing standardized comparisons (e.g., crimes per 1,000 residents).

Prior Conviction

A nominal variable indicating whether a person has a prior conviction (typically Yes/No).

Prior Offenses

A variable that can be coded as ordinal (0,1,2,3 or more) or as actual numbers (interval).

Coding scheme

The numeric or symbolic coding assigned to value labels (e.g., 0=No, 1=Yes; 1=1, 2=2).