Data Analysis Project
andyinla
CrimeSpr191.pptxCrime
Crime
Summary data compiled by the FBI from local police departments.
Issues
Legal definitions of crime.
“Downgrading crimes” by agencies to create more favorable stats.
Hierarchy rule (only the most serious offense recorded).
National Incident-Based Reporting System (NIBRS)
More detailed incident-based data on offense, offender, victim, property, location, etc.
Compiled by the FBI beginning in 1989 to replace the UCR by 2021.
Records 52 major crimes classifications and up to 10 offenses per incident (no hierarchy rule).
Issues: still in development and lacking data to study national crime patterns.
National Crime Victimization Survey (NCVS)
Annual household survey compiled by the Census Bureau on behalf of the Justice Department.
Includes both reported and unreported crimes.
Issues:
Excludes murder, commercial burglary and robbery, victimless crimes, and white-collar crimes.
Excludes crimes against the homeless.
One household respondent: may exaggerate their own, versus other HH members, victimizations.
Telescoping: bringing past victimizations closer to the present.
Crime
Lower Crime Rates
1991-2000: U.S. experienced a 44% drop in homicides, 47% drop in robberies, and a 42% drop in burglaries.
2001-2008: stabilized, then dropped again in 2009 and 2010.
Explanations:
Good economic conditions
Higher incarceration rates
Improved policing efforts
Expanded availability of abortions after 1973
Increased Ratio of Female Criminals
Since the mid-1990s the ratio of offenses committed by women relative to men has increased.
The number of women offenders decreased but not as fast as that for men.
Explanations:
Higher single (vs. collaborative) offender incidents.
More equal gender roles and increased freedom of women.
Welfare reform.
Crime
Highest Crime Regions
Adjusted vs. Unadjusted Statistics
Adjusted Data: correlated variables such as poverty, income, and unemployment are controlled for to better evaluate law enforcement counter-crime policies.
Unadjusted Data: Appropriate for evaluating quality of neighborhood and overall safety.
Crime typically measured per 100,000 people.
Overestimates of population can artificially lower crime rate.
Underestimates can artificially raise crime rate.
Unreliable data: deficient reporting and lack of a clear definition.
2012 & prior: UCR: “The carnal knowledge of a female forcibly and against her will”.
Post-2012 UCR: “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”
Includes: attempts of such, male-male rape, and non-consensual/non-violent rape
Excludes: statutory rape and incest
Crime
Poverty & Crime
Correlated at group level (city, neighborhood, etc.) but not at individual level.
Implications: government policy designed to counter poverty may not reduce crime.
Correlation not consistent with the Great Depression and Great Recession.
Offsetting factors: less to steal, decreased mobility, and greater unity in extreme economic conditions.
High Black Crime
Despite being 13% of population, 27% of arrestees and 39% of prison population are black.
Blacks age 25-44 are 8 times more likely to be murdered than whites.
Disproportionate black on black crime.
Type of crime may increase probability of being arrested
Blacks more likely to commit conspicuous street crimes.
Whites more likely to commit inconspicuous burglary and drunk driving.
“Crime” in UCR omits white-collar crimes.
May be discrimination and bias in the law enforcement and judicial system.
Crime
Does Prison Pay
US has the highest incarceration rate in world.
In 2008 over 2% of US males age 18-64 were incarcerated.
Cost-benefit analysis
Cost to imprison an individual per year (approximately $30-$40 thousand).
Benefit is the value of reduced crime.
Criminals often commit a larger number of crimes than they are apprehended for, raising benefits.
3-strikes law increased imprisonment rates and costs but did it decrease crime?
Gun Control
Estimates indicated there are over 300 million guns in America.
Highly concentrated: half of households do not own guns.
Correlation studies are difficult since ownership is both legal and illegal.
Who kills and would gun control stop them?
≈15% of homicides are committed by family members.
≈55% of homicides are committed by individuals known to the victim (includes family).
≈12% of homicides are committed by strangers.
