Lecture 1 – Introduction
Amy’s cats’ names (bonus question at the exam!):
- Bobby (Bobbalicious)
- Sammy (Samuel L Catson)
Introduction to criminology
Criminology: the scientific study of crime. It’s different from criminal justice studies. It’s
multidisciplinary:
- Sociology
- Political science
- Psychology
- Economics
- Geography
- History
It comprises the actors, causes, consequences, reactions, policies and programs of crime.
The goal is to understand, prevent, and reduce crime.
Crime can be defined in different ways:
- Legal: behaviour prohibited by law
- Moral: behaviour that offends shared morals and beliefs
- Social: behaviour that violates social norms
- Humanistic: behaviour that denies human rights
Crime is a social construction! There are three elements of ‘a crime’:
- Harm
- Social consensus
- Official societal response
This means that crime can change over time: in most Western countries, homosexuality isn’t
a crime anymore (while it remains a crime in many Middle-Eastern and Asian countries).
Patterns of crime
- Spatial trends: patterns over space, for example urban spatial patterns. Some areas
can be known for certain kinds of criminal behaviour, but as soon as you zoom in to
the street level, you see that this criminal behaviour is confined to one particular
street (e.g., sex work).
- Temporal trends: patterns over time, for example urban temporal trends. Some
offenses are more likely to happen later in the day when people have been drinking
and become aggressive.
- Demographic trends: patterns across populations
The “Great American Crime Decline”
Crime dropped by (around) 50% since 1990. You can ask yourself a few questions:
- Which crime dropped?
- Did crime fall everywhere?
- Did social, demographic, and economic factors play a role?
- Did any policy changes play a role?
,Rudy Giuliani (former mayor of NYC) claimed that he helped bring down crime rates more
than anyone else while he was in office. However, the same decline in crime rates happened
in other cities as well where different policies were into place, so it’s not too sure whether
Rudy Giuliani was right in his claim.
Also, the Great “American” Crime Decline also happened in Canada and G7 countries, not
just in North America.
Since the 1990s, crime rates have indeed gone down. However, they are rising again. What
this means for our future? We do not know: we’re very bad at predicting patterns of crime.
On an individual level, it’s possible, but on an aggregate (macro) level this becomes
increasingly unreliable.
Studying and reducing crime
How do we know what causes crime so we can prevent/reduce it?
1. Study crime in its environment
a. Who? What? When? Where?
2. Examine the explanations
a. Literature, previous research
b. Theories
c. Evidence
3. Review current strategies
a. What works?
Crime prevention and crime control
There are four pillars of crime reduction:
- Police
- Courts
- Corrections
- Prevention
There are three crime prevention strategies:
- Developmental
- Situational
- Community
Evidence-based crime prevention
Key strength is in its scientific approach:
- Effectiveness: assess evidence of impact
- Methodological quality: assess the evaluation
o E.g., Maryland Scale of Scientific Methods
- Review the collective evidence: tally/rank studies
- Draw conclusions:
o What works?
o What doesn’t?
o What is promising?
o What don’t we know?
,Assessing methodological quality
Internal validity: the extent to which there is certainty that the intervention caused the
change.
External validity: generalisability of conclusions.
Other aspects of assessing methodological quality are the statistical conclusion validity, the
construct validity, and the descriptive validity.
The Maryland Scale
The Maryland Scale evaluates studies by their internal validity. Looks at:
- Control of other variables
- Measurement
- Statistical power
It ranks articles according to their research design, from 1 (weakest) to 5 (strongest):
- Level 1: simple correlation, single point in time
o CCTV areas have lower crime than non-CCTV
- Level 2: before and after, no control group
o Crime decreased after installing CCTV in area
- Level 3: before and after, with control group
o Crime decreased after installing CCTV in area, but not in control (non-CCTV)
area
- Level 4: before and after, multiple control groups
o Crime decreased in CCTV area compared to control area, controlling for
factors related to crime in those areas
- Level 5: random assignment to comparable groups
o Crime decreased in randomly-assigned CCTV areas compared to control
Multiple studies at levels 3-5 are preferred to draw conclusions. Reviewing the collective
evidence using the Maryland Scale:
- What works?
o At least two L3-L5 evaluations showing effects
o Other evidence
- What doesn’t?
o At least two L3-L5 evaluations showing no effects
o Other evidence
- What’s promising?
o At least one L3-L5 study showing effects
o Other evidence
- What’s unknown?
Studying crime: Medellín
Very high and persistent levels of violence in homicide in Medellín. There are theories and
evidence on social inequality and economic exclusion:
- Strain: lack of access to legitimate jobs, wages
- Social control: weak community cohesion, bonds
- Situational: lack of street lighting, police stations far away
There were multiple possible strategies/programs, which basically all came down to
improved access to economic opportunities:
, - Redistributive policies
- Employment programs, job training
- More police patrol
- Improved public space
The city board installed the Medellín metrocable: a gondola that connected a lot of different
areas, also the areas that were previously cut off from the rest of the city. The Metrocable
evaluation reported the following:
- Metrocable was built according to topography
- Gondola+
- Police station, patrols
- Better lighting, pathways
- Schools, libraries, recreational centres
- 66% greater decline in homicide
- 75% decline in reported violence
It was difficult to conclude whether these gondolas actually caused the decrease in
homicide and violence rates, or whether this was due to other factors which coincided with
the instalment of these gondolas. Therefore, both the internal and external validity are
probably not very high.
Policy implications
Evidence-based crime prevention can best inform policy. However, this brings along
questions about:
- Political challenges
o Electability: will this cost me potential voters?
- Cost effectiveness
- Common-sense thinking
o “We are all entitled to our own opinions, but not to our own facts”
o Most people have some ‘common sense’ knowledge on crime, but this often
goes against what researchers observe about crime
What can we do?
- Adhere to scientific standards
- Collect and evaluate studies
- Examine theoretical mechanisms
- Test, retest, repeat
Bad science, bad policy
A good example of bad policy originating in bad science is the “more guns, less crime”
argument that is often used in the USA. John R. Lott, Jr. is a big proponent of guns and a
writer of books on the subject. However, he has been critiqued in many ways:
- Results not robust
- Questionable data
- Extended time period -> no effects
- Numerous studies ignored that link firearms to violence:
o Interpersonal violence (assault)
o Homicide
o Suicide
o Accidents
Amy’s cats’ names (bonus question at the exam!):
- Bobby (Bobbalicious)
- Sammy (Samuel L Catson)
Introduction to criminology
Criminology: the scientific study of crime. It’s different from criminal justice studies. It’s
multidisciplinary:
- Sociology
- Political science
- Psychology
- Economics
- Geography
- History
It comprises the actors, causes, consequences, reactions, policies and programs of crime.
The goal is to understand, prevent, and reduce crime.
Crime can be defined in different ways:
- Legal: behaviour prohibited by law
- Moral: behaviour that offends shared morals and beliefs
- Social: behaviour that violates social norms
- Humanistic: behaviour that denies human rights
Crime is a social construction! There are three elements of ‘a crime’:
- Harm
- Social consensus
- Official societal response
This means that crime can change over time: in most Western countries, homosexuality isn’t
a crime anymore (while it remains a crime in many Middle-Eastern and Asian countries).
Patterns of crime
- Spatial trends: patterns over space, for example urban spatial patterns. Some areas
can be known for certain kinds of criminal behaviour, but as soon as you zoom in to
the street level, you see that this criminal behaviour is confined to one particular
street (e.g., sex work).
- Temporal trends: patterns over time, for example urban temporal trends. Some
offenses are more likely to happen later in the day when people have been drinking
and become aggressive.
- Demographic trends: patterns across populations
The “Great American Crime Decline”
Crime dropped by (around) 50% since 1990. You can ask yourself a few questions:
- Which crime dropped?
- Did crime fall everywhere?
- Did social, demographic, and economic factors play a role?
- Did any policy changes play a role?
,Rudy Giuliani (former mayor of NYC) claimed that he helped bring down crime rates more
than anyone else while he was in office. However, the same decline in crime rates happened
in other cities as well where different policies were into place, so it’s not too sure whether
Rudy Giuliani was right in his claim.
Also, the Great “American” Crime Decline also happened in Canada and G7 countries, not
just in North America.
Since the 1990s, crime rates have indeed gone down. However, they are rising again. What
this means for our future? We do not know: we’re very bad at predicting patterns of crime.
On an individual level, it’s possible, but on an aggregate (macro) level this becomes
increasingly unreliable.
Studying and reducing crime
How do we know what causes crime so we can prevent/reduce it?
1. Study crime in its environment
a. Who? What? When? Where?
2. Examine the explanations
a. Literature, previous research
b. Theories
c. Evidence
3. Review current strategies
a. What works?
Crime prevention and crime control
There are four pillars of crime reduction:
- Police
- Courts
- Corrections
- Prevention
There are three crime prevention strategies:
- Developmental
- Situational
- Community
Evidence-based crime prevention
Key strength is in its scientific approach:
- Effectiveness: assess evidence of impact
- Methodological quality: assess the evaluation
o E.g., Maryland Scale of Scientific Methods
- Review the collective evidence: tally/rank studies
- Draw conclusions:
o What works?
o What doesn’t?
o What is promising?
o What don’t we know?
,Assessing methodological quality
Internal validity: the extent to which there is certainty that the intervention caused the
change.
External validity: generalisability of conclusions.
Other aspects of assessing methodological quality are the statistical conclusion validity, the
construct validity, and the descriptive validity.
The Maryland Scale
The Maryland Scale evaluates studies by their internal validity. Looks at:
- Control of other variables
- Measurement
- Statistical power
It ranks articles according to their research design, from 1 (weakest) to 5 (strongest):
- Level 1: simple correlation, single point in time
o CCTV areas have lower crime than non-CCTV
- Level 2: before and after, no control group
o Crime decreased after installing CCTV in area
- Level 3: before and after, with control group
o Crime decreased after installing CCTV in area, but not in control (non-CCTV)
area
- Level 4: before and after, multiple control groups
o Crime decreased in CCTV area compared to control area, controlling for
factors related to crime in those areas
- Level 5: random assignment to comparable groups
o Crime decreased in randomly-assigned CCTV areas compared to control
Multiple studies at levels 3-5 are preferred to draw conclusions. Reviewing the collective
evidence using the Maryland Scale:
- What works?
o At least two L3-L5 evaluations showing effects
o Other evidence
- What doesn’t?
o At least two L3-L5 evaluations showing no effects
o Other evidence
- What’s promising?
o At least one L3-L5 study showing effects
o Other evidence
- What’s unknown?
Studying crime: Medellín
Very high and persistent levels of violence in homicide in Medellín. There are theories and
evidence on social inequality and economic exclusion:
- Strain: lack of access to legitimate jobs, wages
- Social control: weak community cohesion, bonds
- Situational: lack of street lighting, police stations far away
There were multiple possible strategies/programs, which basically all came down to
improved access to economic opportunities:
, - Redistributive policies
- Employment programs, job training
- More police patrol
- Improved public space
The city board installed the Medellín metrocable: a gondola that connected a lot of different
areas, also the areas that were previously cut off from the rest of the city. The Metrocable
evaluation reported the following:
- Metrocable was built according to topography
- Gondola+
- Police station, patrols
- Better lighting, pathways
- Schools, libraries, recreational centres
- 66% greater decline in homicide
- 75% decline in reported violence
It was difficult to conclude whether these gondolas actually caused the decrease in
homicide and violence rates, or whether this was due to other factors which coincided with
the instalment of these gondolas. Therefore, both the internal and external validity are
probably not very high.
Policy implications
Evidence-based crime prevention can best inform policy. However, this brings along
questions about:
- Political challenges
o Electability: will this cost me potential voters?
- Cost effectiveness
- Common-sense thinking
o “We are all entitled to our own opinions, but not to our own facts”
o Most people have some ‘common sense’ knowledge on crime, but this often
goes against what researchers observe about crime
What can we do?
- Adhere to scientific standards
- Collect and evaluate studies
- Examine theoretical mechanisms
- Test, retest, repeat
Bad science, bad policy
A good example of bad policy originating in bad science is the “more guns, less crime”
argument that is often used in the USA. John R. Lott, Jr. is a big proponent of guns and a
writer of books on the subject. However, he has been critiqued in many ways:
- Results not robust
- Questionable data
- Extended time period -> no effects
- Numerous studies ignored that link firearms to violence:
o Interpersonal violence (assault)
o Homicide
o Suicide
o Accidents
