Conceptualization and Measurement in Criminology

Vocabulary flashcards covering conceptualization, operationalization, scales of measurement, reliability, validity, measurement error, and crime data sources.

Concept

A mental image or abstraction that summarizes a set of similar observations, feelings, or ideas.

Conceptualization

The process of defining and specifying the meaning of a concept.

The Measurement Process

A multi-step process involving literature review, defining concepts, specifying indicators, spelling out procedures, applying operational definitions to produce data, and assessing quality.

Measurement

Assigning numbers or labels to units of analysis, such as people, places, objects, or events.

Operationalization

Specifying the indicators of a concept and spelling out the procedures by which they will be applied.

Variable

Something that can vary between units in the context of a research study.

Constant

Something that does not vary between units in a study.

Conceptual Definition

A definition specifying the theoretical meaning of a concept, such as defining level of gang involvement as the degree to which a person is bonded to other members and engaged with gang activities.

Operational Definition

The explicit procedure or survey question used to measure a concept, such as asking how many hours a person spends with gang members per week.

Index

A composite measure which uses multiple indicators to capture a single concept.

Unobtrusive Measures

Information collected without direct participation of subjects, such as tracking browser cookies or counting abandoned buildings.

Nominal Scale of Measurement

Measures of attributes with no mathematical relationship to each other other than equal or unequal (i.e., categorical data).

Dichotomous Measures

Nominal measures that have exactly two categories, such as $1/0$ or yes/no.

Ordinal Scale of Measurement

Measures that imply an order but where the distance between cases is unknown, allowing 'less than' or 'greater than' comparisons.

Interval Scale of Measurement

Measures with continuous values and fixed units of measurement, but no absolute zero, meaning distance between cases is known but ratios are not.

Ratio Scale of Measurement

Measures with continuous values, fixed units of measurement, and an absolute zero, meaning ratios between cases are known.

Measurement Reliability

A property of a measurement procedure yielding consistent scores as long as the phenomenon being measured is not changing.

Measurement Validity

The degree to which a measurement actually captures the concept it was designed to measure.

Reliability vs. Validity Example

An illustration demonstrating that consistent but inaccurate responses are reliable but invalid, whereas consistent and accurate responses are both reliable and valid.

Face Validity

The type of validity evaluating whether a measure seems to capture the concept 'on its face'.

Content Validity

The type of validity evaluating whether a measure captures all aspects of a concept.

Construct/Criterion Validity

The type of validity evaluating whether a measure is associated with other measures of the same concept or variable.

Test-Retest Reliability

The consistency or comparability of measurement scores from Time 1 to Time 2.

Inter-item Reliability

The internal consistency evaluating whether multiple measures of the same concept are highly associated with each other.

Alternate-forms Reliability

The consistency of scores across slightly different versions of a measurement procedure, such as split-halves or varying question ordering.

Intra-observer & Inter-observer Reliability

The consistency of measurement scores within a single observer (intra) and between different observers (inter).

Measurement Error

The difference between a measured quantity and its true value.

Random Measurement Error

Errors that make relationships between variables less precise but do not necessarily introduce bias.

Systematic Measurement Error

Errors associated with relevant variables that cause relationships between variables to be biased.

Uniform Crime Reports (UCR)

A standardized crime reporting system where police departments voluntarily send records of crimes known to police monthly to the FBI.

Part 1 (Index) Crimes

The eight primary UCR crime categories: Murder, Rape, Robbery, Aggravated Assault, Burglary, Theft (larceny), Motor Vehicle Theft, and Arson.

Hierarchy Rule

A UCR reporting rule stating that if multiple Part 1 crimes occur in a single incident, only the most serious crime is reported (unless motor vehicle theft or arson is involved, which are both reported).

National Incident Based Reporting System (NIBRS)

An incident-based crime reporting system that captures all crimes per incident and detailed context, but faces lower police department participation rates due to complex reporting requirements.

National Crime Victimization Survey (NCVS)

A nationally representative survey asking Americans about victimizations in the past 12 months, generating crime rate estimates available only at the national level.

Telescoping Effect

A survey memory bias where respondents think a victimization event happened more recently than it actually did.

Aggravated Assault

An unlawful attack by one person upon another for the purpose of inflicting severe or aggravated bodily injury, characterized by dangerous weapons or severe injuries like broken bones.

Rape (Revised Definition)

Penetration, no matter how slight, of the vagina or anus with any body part or object, or oral penetration by a sex organ of another person, without the consent of the victim (revised in 2013).

Homicide Data Sources Comparison

A validation comparison showing close alignment over time between UCR Murders (police reported) and NVSS Homicides (death certificates).

Deaths of Despair

Preventable deaths among middle-class men without a college degree caused by drug overdose, suicide, and liver disease from alcoholism.