POLI 110 week 3 (final version)

Concept

An abstract idea that represents a phenomenon in the real world, like democracy or power.

Concept Specification

The process of clearly defining a concept to distinguish it from other similar terms.

Why Concepts Matter

Clear concepts improve communication, comparison, theory-building, and theory-testing; vague ones weaken arguments.

Operationalization

Transforming abstract concepts into measurable and observable variables.

Manifest Variable

A variable that can be directly observed or measured (e.g., age or income).

Latent Variable

A variable that cannot be directly observed and must be measured using indicators (e.g., democracy).

Disjunct Variables

Variables whose categories do not overlap.

Exhaustive Variables

Variables that cover all possible values or categories.

Absolute Variable

A variable that can be measured independently from other units (e.g., height in cm).

Relational Variable

A variable that must be measured in relation to something else (e.g., age percentile).

Measurement

The process of linking theoretical concepts to empirical indicators or data.

Goals of Measurement

Objectivity, reliability, and validity.

Objectivity

A measure is independent of who collects or analyzes the data.

Implementation Objectivity

Standardization during data collection to reduce bias (e.g., avoiding interviewer effects).

Evaluation Objectivity

Ensuring data analysis is consistent and unbiased (e.g., text interpretation rules).

Reliability

The degree to which repeated measurements yield the same result.

Test-Retest Method

Measuring the same object twice to test reliability; sensitive to changing characteristics.

Parallel Test Method

Using different methods to measure the same concept to test reliability.

Validity

The degree to which a measure accurately captures the intended concept.

Face Validity (Prima Facie Validity)

A measure appears to assess what it is supposed to, based on intuition or first impression.

Content Validity

A measure captures all relevant dimensions of the theoretical concept.

Criterion Validity

A measure correlates strongly with an established external standard.

Construct Validity

A measure aligns with theoretical relationships between concepts.

Convergent Validity

A measure correlates highly with other measures of the same concept.

Discriminant Validity

A measure shows low correlation with unrelated concepts.

Nominal Scale

Classifies data into categories with no inherent order (e.g., party labels).

Ordinal Scale

Ranks categories in order but with unclear spacing between them (e.g., support levels).

Interval Scale

Ranks with equal spacing but no true zero (e.g., temperature in °C).

Ratio Scale

Ranks with equal spacing and a true zero (e.g., income, population).

Categorical Scales

Nominal and ordinal; used for classification without numeric interpretation.

Numeric Scales

Interval and ratio; allow for arithmetic operations and precise comparisons.

Reliable but Not Valid

A measure is consistent but does not measure the correct concept.

Valid but Not Reliable

A measure captures the right concept but does so inconsistently.

Valid and Reliable

Ideal measurement: both accurate and consistent.