PSYC 3402: Measuring Crime

Contents

1. 1. Looking at Crime
   1. 1.1 Definitions of Crime
   2. 1.2 A Review of Terms
2. 2. Measuring Crime
   1. 2.1 Three Types of Crime Rates and One Index
   2. 2.2 Three Ways to Measure Crime Rates
   3. 2.3 Measuring Crime Correlates
   4. 2.4 Why should we bother measuring crime?
3. 3. Crime Trends – A Breakdown of Results

Updated July 14, 2010:
This post contains notes from the book.

1. Looking at Crime

1.1 Definitions of Crime

Crime: acts that are prohibited under the law and render the actor subject to intervention by justice professionals (legal);  violation of societal norms (social); violation of religious and moral norms (moral); acts that reward perpetrator but harm others (psychological).

• That being said, though inter-culturally, people agree on basic crimes, criminal acts can change over time and place (e.g. abortion laws, gay marriage)

Crime Desistance: as age increases, criminal activity decreases

Psychological Focus: intra-individual differences – variations in criminal conduct for an individual over time and situations – and inter-individual differences – variations of criminal behaviour between individuals.

PIC-R (Personal, Interpersonal and Community-Reinforcement Model): (to explain better later) posits that criminal behaviour reflects the immediate factors that combine to influence a decision to engage in criminal behaviour – this is further influenced by antisocial attitudes, history of criminal behaviour, cost-benefit analysis and the presence of social support for crime. (Learning Theory)

1.2 A Review of Terms

Meta-Analysis: an aggregate of many studies over time.

• Advantages: the aggregate effect size provides a quantitative summary of a large body of research; formulas can be used to combine statistical results to one method, therefore allowing a direct comparison between studies.
• Disadvantages: publication bias (a tendency for studies with significant results to be published) – resolve this by seeking out unpublished papers; low-quality studies should not be as important in your results.

ROC (Receiver Operating Characteristic) Analysis: gives you a particular accuracy measure (AUC – area under the curve) – the probability that an outcome (re-offence) will occur.

• AUC: the higher the AUC, the more probable.

Base Rates: the percentage of people within a given population who engage in a specific behaviour; it tends to be easier to predict frequent events than non-frequent events.

2. Measuring Crime

2.1 Three Types of Crime Rates and One Index

Most common measures are aggregated crime rates (at the level of large areas consisting of populations of people) which are different from individual crime rates, which is what we look at:

• Prevalence Rates: where we get the number of offenders in a given area over a given time period divided by the population of people in that area in that time period (i.e. # of criminals/population)
• Incidence Rates: where we get the number of incidents committed in a given area during a time period divided by the number of offenders in that area in that time period (i.e. # of incidents/offenders)
• Per Capita Crime Rates: where we get the prevalence rate multiplied by incidence rate in a given area during a given time period (i.e. [# of criminals/population] * [# of incidents/offenders])
• Crime Severity Index: this index tracks two components of crime – the changes in volume of a particular crime and the relative seriousness of the crime, where more serious crimes have more weight.
• Weight: composed of two components – proportion of people convicted of the offence who are sentenced to time in prison and the average length of prison sentence given for a specific offence.

Issues with Crime Rates: see disadvantages mentioned in Official Statistics (next section).

2.2 Three Ways to Measure Crime Rates

There are three major ways to measure crime (below). For best results, we try to combine and triangulate the results from these methods in order to get the most accurate measure of crime.

1. Victim Surveys: where we interview people through community samples or mail surveys by asking them about their experiences of being victims of crime.
   Advantage: helps reveal more crime than official statistics (alleviating the issue of the Dark Figure).
   Disadvantage: people still under-report certain crimes (such as domestic assaults, minor theft and rape).
2. Official Statistics: primarily crimes recorded by police (ex: FBI Uniform Crime Rates) but can be collected other ways: clear-up rates; charges; convictions.
   Advantage: most accessible form of crime statistics.
   4 Disadvantages (WHUK):
   1. Winnowing Effect: the number of crimes is reduced as we travel through the CJS.
   2. Hierarchy: when multiple crimes occur at the same time, only the most serious crime is reported.
      Ex: when rape is committed after a B&E, only the rape is reported.
   3. Under-recorded: police under-record crimes either because the offender is a child; crime is trivial or participant is to blame; or claim is false, mistaken or withdrawn. (Think: police discretion.)
   4. Know (Dark Figure): crimes must be known and recorded. (Think: victim discretion).
3. Self-Reports: where individuals record their involvement in crime by means of anonymous questionnaires or interviews by specifically looking at performance of specific acts within specific periods of time.
   Advantage: significantly reduces the Dark Figure.
   4 Disadvantages (ODUS):
   1. Over-reporting: respondents may over-report their involvement in crime (i.e. showing off)
      Ex: Craig’s RA’s experience with Liverpool delinquents.
   2. Desirable: respondents can answer in socially desirable ways – especially prisoners.
   3. Understanding: respondents often don’t understand the items (tends to be remedied nowadays).
   4. Seriousness: self-reports tend to identify the less serious offenders.

2.3 Measuring Crime Correlates

Sociologists and Criminologists believe that socio-economic status, age, gender, and race cause crime but Psychologists tend to believe that the Big Four (anti-social personality, attitudes, peers and history of anti-social behaviour) and Central Eight cause crime.

Individual Correlates: psychologists are particularly interested in understanding what causes crime at the level of the individual offender (as opposed to criminologists and sociologists).

3 Major Problems with Correlates (REC):

1. Reliability of Correlations: single estimates are unreliable so we have to have multiple measures of the relationship between A and B.
2. Ecological fallacy: this is the assumption that aggregate correlations imply individual correlations – not always the case! (Wikipedia Examples)
3. Correlation does not imply causation.

Solution: use meta-analytic techniques to examine effect sizes at the individual level.

First and Second Order Correlates of Crime

The list below outlines the major variables related to criminal activity:

1. First Order – Central Eight:
   • Big Four:
     • Criminal history: static
     • Antisocial personality: impulsivity, aggression
     • Antisocial attitudes: dynamic, stable
     • Antisocial associates: dynamic acute and dynamic stable
   • Moderate Risk Factors:
     • Family/marital
     • School/work
     • Leisure and Recreation
     • Substance Abuse
2. Second Order – Minor Risk Factors: persona/emotional distress, major mental disorder, physical health issues, fear of official punishment, physical conditioning, low IQ, social class of origin, serious of current offence, other factors.

Research Points:

• Addressing the Criminogenic Needs (big four) is empirically shown to reduce recidivism rates;
• Minor risk factors don’t explain criminal conduct;
• Take-away point: research empirically supports a cognitive social theory model of criminal behaviour.

2.4 Why should we bother measuring crime?

• Social policy issues, including effective distribution of police, proper resource allocation, and knowing which groups are victimized more.
• Evaluation of CJS, including ensuring crime interventions are effective which indicates whether or not our tax payer money is being spent well (same applies to offender interventions).
• Construction of theories, meaning in order to test our crime theories, we need to be able to measure crime before and after.

Examples

1. Crime Hot-spotting: used to efficiently distribute police resources across a geographical area based on higher probabilities that crime will happen – which is based on historical records.
2. Treatment Evaluation: CJS attempted to intervene with annoying prisoners but showed a low success (we need to evaluate our theories and interventions via crime rates).
3. Theories of Crime: the most widely accepted part of crime theories suggest that men, more than women, commit crimes and that the older you get, the less likely you are to commit crime.

3. Crime Trends – A Breakdown of Results

• Public Perception:
  • Least confidence: police preventing crime, courts appointing appropriate sentences, corrections rehabilitating offenders and parole supervising offenders in the community.
  • Greatest confidence: police solving crimes, courts determining guilt, corrections preventing escapes.
• Sentence Lengths: 80-90% of offenders spend less than 6 months incarcerated (2% spend two or more years incarcerated);
  • Too-short sentences means that offenders have less time to spend being assesses and participating in rehabilitative prison-programs.
• 30% of Canadians (per a phone survey) reported being a victim of crime.
• Crime costs too much money ($60 billion in 2003) – that’s just less than half our direct and indirect health care costs for 2003, $123 billion.