Evaluate the claim "we can predict who will become a criminal"
Sharonhy
Lecture1-Introfinal.pptxCriminal Behaviour CRM203
Dr David Keatley
Senior Lecturer in Criminology
Chair of Crime Science
Director of Researchers in Behaviour Sequence Analysis (ReBSA)
Director of Cold Case Review
Lecture 1: Introduction
1
1
Lecture Overview
Course overview
Assessments
What is Criminal Behaviour?
2
2
Assessment 1: Essay
Title: Evaluate the claim “we can predict who will become a criminal”.
Due Date: 1st October, 2021, 5pm.
Max. word count: 1500 (not including reference list or tables etc)
THIS IS A REVIEW OF RESEARCH LITERATURE
Articles (to get you started)
Herrenkohl, T. I., Jung, H. J., & Lee, J. O. (2017). Effects of child maltreatment, cumulative victimization experiences, and proximal life stress on adult crime and antisocial behavior. Summary Report for US Department of Justice.
Keatley, D. A., Golightly, H., Shephard, R., Yaksic, E., & Reid, S. (2018). Using behavior sequence analysis to map serial killers’ life histories. Journal of interpersonal violence, 0886260518759655.
Monahan, J. (1984). The prediction of violent behavior. American Journal of Psychiatry, 141(1), 10-15.
3
3
What is (criminal) Behaviour?
4
4
How do we measure behaviour?
(and how do we apply this in real world?)
5
5
What is Crime?
Where do we get our knowledge/understanding of crime from?
6
6
Criminal Statistics
Where do we get our stats from?
Official sources, mostly police and courts.
Do these reflect the actual level of crime?
Probably not
7
7
Criminal Statistics
There is no universal, standard categorisation scheme for crime
Different sources/assessments give different data
Theft and property crimes = majority
British Crime Survey
Police recorded crime
8
8
Criminal Statistics: Sources
Source
Benefits
Limitations
Surveys
of general population
Researcher
controlled;
Incl. events
not in
court/reported
Recall;
accuracy;
exaggeration; low response
rate; ethics
Police stats
Better evidence?;
systematic?
Reporting bias; reported
vs
recorded;
Court
stats
Validity?
Accessible?
Prosecution
vs
no
charge
Prison
Stats
Breakdown of
criminals/crimes;
Biases in sentencing;
change in policy and law;
validity?
General population
offender
surveys
Random sample;
overcome police and
court limitations
Bias (serious
criminals
arrested); deception
9
9
Task: Was there a robbery?
Case 1
Man and woman talking for approximately 10 seconds…
Man walks away
Case 3
Man and woman talking for approximately 10 seconds…
Man runs away
Case 2
Man and woman talking for approximately 30 seconds…
Man walks away
Case 4
Man and woman talking for approximately 30 seconds…
Man runs away
10
10
Task: Was there a robbery?
Langsdale and Greenberg (2006)
Man running after 10 seconds more likely to be a crime than if ran after 30 seconds
Walking away = less chance of crime?
In real life, this may be the difference between a crime being reported or not “interpretive processes”
Remember those ‘shuffling bodies’ – context is important!
11
11
Reporting of crime: Usefulness
Limited usefulness in limited circumstances
Result of complex social and cognitive processes
That we need to be aware of!
Problems of source, nature, and interpretation of data
There are three kinds of lies:
Lies, damned lies, and statistics
(Mark Twain [maybe])
12
12
40% of crime is reported
What is likely to be reported? Why?
What is least likely to be reported? Why?
Who is likely to be reported? Why
Who is likely to report? Why
13
Using official statistics
Crime definitions vary between and within jurisdictions.
Crime statistics do not reflect the actual level of crime.
Most of what we know about offenders comes from those who are adjudicated
Are these adjudicated offenders a reasonable representation of offenders; a representative sample?
What does it mean?
14
Victimisation Surveys
Another method of measuring crime is through victimisation surveys.
Victimisation surveys are regularly conducted in many jurisdictions. In Australia, the Australian Bureau of Statistics publishes these.
These figures confirm that not all crime is reported and not all reported crime is recorded
15
Victimisation Surveys
Probably an underestimate because:
Some people are difficult to survey (children, remote communities, people who live with the offender)
Some people do not know they have been victims
But are useful because they give a better/different picture and they can investigate why some crimes are not reported
16
Self-Report Measures
Another method is to simply ask a representative sample about their offending.
Turns out there is much more crime than we thought – seems we are all criminals
17
Using Crime Statistics
Crime statistics are used in a range of areas such as testing criminological theories, evaluation of the criminal justice system, setting criminal justice priorities and forecasting.
You need to be aware of two key (related) issues
Ecological fallacy
Distinction between incidence and prevalence
18
Ecological Fallacy
Making judgements about an individual based on aggregate data
For example, low income areas have higher rates of crime than high income areas
You might infer that low income individuals commit more crime
BUT higher income individuals who live in low income areas tend to commit more crime AND low income individuals in high income areas tend to commit less crime
19
Prevalence and Incidence
Need to distinguish between incidence and prevalence
Prevalence refers to the number of offenders in a given area over a given time.
Incidence refers to the number of offences committed by offenders in a given area over a given time.
Crime rate refers to the number of offences in a given area over a given time.
20
Prevalence and Incidence
Area A has 3,000 burglars per 100,000 who commit 2 crimes per year giving a crime rate of 6,000 burglaries per year
Area B has 1,000 burglars per 100,000 who commit 7 crimes per year giving a crime rate of 7,000 burglaries per year
21
Correlations!!
Mostly our knowledge of crime is based on correlations
Correlations: crime and poverty are correlated. What does this mean?
22
Correlations!!
Mostly our knowledge of crime is based on correlations
Correlations: crime and poverty are correlated. What does this mean?
Very few randomised control studies
Why?
23
Correlations!!
Mostly our knowledge of crime is based on correlations
Correlations: crime and poverty are correlated. What does this mean?
Very few randomised control studies
Why?
24
Correlations!!
Mostly our knowledge of crime is based on correlations
Correlations: crime and poverty are correlated. What does this mean?
25
What does it all mean?
Our data are not very good:
Not random samples, especially for some offences. Why does this matter?
Crime statistics are poor.
Crime statistics can be difficult to interpret. Does increasing crime rates mean increasing crime? Why is there Indigenous over-representation in prisons?
26
Question: What are the chances?
On my walk to work today, I saw a car with the licence plate: CR1ME P4YS
27
27
Court proceedings:
There has been a hit and run accident at night involving a taxi cab. The only eyewitness identified the taxi cab as being blue. There are two taxi companies operating in the city. One company has green cabs which make up 85% of the total cabs in the city. The other company has blue cabs, comprising the remaining 15% of taxis.
Why does it even matter? -Researcher-Practitioner-
The eyewitness was tested by the court under similar night time conditions and was able to accurately distinguish the colour of cab 80% of the time.
What is the likelihood that the eyewitness correctly identified the cab as blue?
28
Why does it even matter? -Researcher-Practitioner-
People frequently estimated the likelihood of proper identification of the cab at 80%, consistent with the eyewitnesses' demonstrated reliability.
Even those who attempted to calibrate a more accurate estimate judged that the eyewitness would be correct over 50% of the time.
The reality is that given the underlying percentages of cab colours (only 15% are blue), the eyewitness is likely to be incorrect in their testimony nearly 60% of the time. That is, they are more likely to be wrong than right! taxis.
Source: Tversky and Kahneman (1982)
29
Why does it even matter? -Researcher-Practitioner-
30
Why does it even matter? -Researcher-Practitioner-
CRIME RATES
“theft increases 50%”
Shock, horror, panic… 50%... Help!!! RUN..
But wait… 50% … OF WHAT????
31
Why does it even matter? -Researcher-Practitioner-
“theft increases 50%”
That sounds a lot, so let’s call it an increase of 2% and sleep easy
Imagine: 4 crimes last year…
50% (relative) and a 2% (absolute) increase both = 6
32
cf. Spousal homicide
Final thoughts…
You can make Criminology better (and I want you to)
Behaviour is not easy to define (let alone measure)
Beware statistics
33
