CRIM 2653 (2024-25) Class 8 - Quantitative and Qualitative Measurement PDF

Chapter 3 Quantitative and Qualitative Measurement  The Preliminary Steps of Measurement Measurement has a series of four preliminary steps: 1. Articulating the topic 2. Determining the research problem 3. Identifying a purpose statement 4. Developing r...

Chapter 3 Quantitative and Qualitative Measurement  The Preliminary Steps of Measurement Measurement has a series of four preliminary steps: 1. Articulating the topic 2. Determining the research problem 3. Identifying a purpose statement 4. Developing research questions Reasoning in the Scientific Process Deductive logic o Example: Skogan’s (2009) study of the relationship between concerns about crime and confidence in police. Inductive logic o Example: Dunham and Alpert’s (2009) study of the dynamics of police–citizen encounters.  The Research Question Drives the Method Qualitative purpose statements describe, develop, understand, and discover Quantitative purpose statements identify a theory and may state the directional relationship between concepts The Measurement Process  Defining Crime Varies over times & among countries Multiple precise ways Real-world effects  Measurement Challenges Social construction of crime Varies over time & place Historical, cultural, social influences Power, class, gender, & race Precision & careful processes Measurement challenges 8 Pride March Toronto 1972  Measurement Process Abstract concept - detect empirically Technique or process Process of defining a concept From imprecision to conciseness Specifying meaning Concept of “homelessness,” how to measure?  Conceptual Definition An explicit working definition of a concept Specificity & clarity Precision about aspect of concept under examination Necessarily includes certain aspects & excludes others Concept Conceptual Definition Homelessness Persons living on the streets or without a shelter, or with no usual place of residence 10  Operationalization Translating concept into something that can be measured empirically What & how to measure Common understanding Operationalization 11  Operational definition Operational Definitions Homelessness (I)1 Persons who were without night-time shelter last night, or are expecting to be without housing tonight Chronic Six or more months of homelessness homelessness2 Episodic2 3 or more episodes lasting less than six homelessness months 12 *From Canadian Definition of Homelessness. N.d. www.Homelesshub.ca https://www.canada.ca/en/employment-social-development/programs/homelessness/reports/highlights-2018-point-in-tim  Homelessness Empirical indicator: use of shelters nightly Surveys: who included & excluded Choices appropriate for research design Operational Definition Homelessness (II) Individual or family who lacks a fixed, regular, and adequate night-time residence 13 *From Canadian Definition of Homelessness. N.d. www.Homelesshub.ca  Operational Definitions Unsheltered People who are sleeping in places unfit for human homelessness1 habitation, including the following locations: streets, alleys, parks and other public locations, transit stations, abandoned buildings, vehicles, ravines and other outdoor locations Sheltered People sleeping in the following locations: homelessness emergency shelters (general & specific to men, women, youth, etc.), extreme weather shelters, Violence Against Women (VAW) shelters, & transitional shelters Op definitions 14 https://www.canada.ca/en/employment-social- development/programs/homelessness/reports/highlights-2018-point-in-time-count.html#1  Setting Priorities How do police set investigational priorities? ‘Harm’: concept referring to negative consequences from crime Organized crime groups Purpose of project: a tool police can use Setting Priorities  Conceptualizing harm How to conceptualize ‘harm’? Process of specifying meaning Multiple dimensions Alternative conceptual definitions  Operationalizing harm Concept Conceptual Definition Harm Harm refers to the negative consequences resulting from organized crime in terms of direct harms (e.g., physical and property damage) How to operationalize conceptual definition of harm? Process of making decisions about measurement Specific choices of measures Operational Definition Harm Harm is the monetary cost of damage from crime and costs of responses to crime  Concept to variable From abstract to empirically observable Determine empirical indicators E.g., survey of business owners on losses to crime Choices must be appropriate to research Level of measurement 18 E.g., use of scale to measure harm by criminal offence type  Variables Variables give concepts measureable characteristics. All variables have attributes, which are a variable’s categories or numeric values. o Dependent variable o Independent variable o Control variable  Hypotheses Hypotheses are untested statements that specify the relationship between two theoretical concepts (directional or non-directional). Alternative hypothesis (H1) Null hypothesis (H0)  Criteria for a Causal Explanation A causal explanation must meet three criteria: 1. Empirical association 2. Temporal order 3. Non-spuriousness Confounding variable Intervening variable (z)  Causality in Qualitative Research Understanding an outcome’s context: how do a series of events in a certain set of conditions lead to a particular outcome? Focus on situations, events, and behaviour contextualized within the beliefs of the individuals observed Reliability & Validity Quality Reliability in Quantitative Research Reliability: the extent to which repeated testing of a concept produces the same result. Tests of reliability include: o Test–retest method o Cross-test method o Split-half method o Inter-rater testing  Validity Truthfulness Precisely matches concept to data What is measured matches definition Concept Operational Definition Homelessness (II) Individual or family who lacks a fixed, regular, and adequate night-time 26 residence  Validity in Quantitative Research Validity: the extent to which a measure measures what it intends to; depends on how well a construct is operationalized. There are eight forms of validity: 1. Face validity 2. Content validity 3. Criterion validity 4. Concurrent validity 5. Predictive validity 6. Construct validity 7. Convergent validity 8. Discriminant validity  Operationalizing Fear Conceptualized in various ways No consensus in literature Operational definition based on survey question Fear factor threshold 28 Reyns, Bradford W. and Christine M. Englebrecht. 2012. “The Fear Factor: Exploring Predictors of Fear Among Stalking Victims Throughout the Stalking Encounter.” Crime & Delinquency. 59(5): 788-808.  Operationalizing Gangs Delinquent activity a prerequisite? Organizational structure? Self-nomination? Involved illegal activities? Level of organization? Operationalizing Gangs 29 Esbenen et al. 2001. Red Scorpions (B.C.), allied with Bacon Brothers  Qualitative Research Qualitative research uses paradigm-specific criteria to assess rigour: o Dependability o Trustworthiness Level of Measurement  Quantitative Variables Two types of quantitative variables: 1. Categorical (discreet) variable § Nominal variable § Ordinal variable 2. Continuous variable § Interval variable § Ratio variable  Summary Categorical Continuous Assigns into particular categories Numerical measure Nominal (no ranked order) Interval (numeric distance between points; no true zero) E.g., gender, marital status, province E.g., IQ scores (specifies distance of residence between scores, no true zero) Ordinal (ranked order, but no exact Ratio (true zero, precise distance between orders) measurement) E.g., highest level of education E.g., income, temperature completed, rank in police department Higher-level variables can be turned into lower-level variables 33 Lower-level variables cannot be turned into higher-level variables Choosing Level of Measurement 1) Mutually exclusive 2) Exhaustive 3) Uni-dimensionality Consistency in fit for all elements Choosing level of measurement 34  Composite Measures Composite measures combine multiple variables to measure a concept that is too complex to capture with a single indicator. o Index o Scales § Criminology and criminal justice research uses five scales: Likert scale, semantic differential scale, Guttman scale, Bogardus social distance scale, and Thurstone scale.