L2. Data-Driven Approach

Data-driven approach

Profilers use a data-driven approach to work up from criminal evidence and build a statistical database

Data-driven approach: statistical models

Statistical models are used to find relationships between crimes and predict offender characteristics

What do profilers build using the data-driven approach?

A statistical database

Investigative Psychology

Determines patterns/correlations in behaviour. Identifies whether a series of offences are linked and likely to have been committed by the same person.

interpersonal coherence

Engagement with victim correlates with engagement in other relationships.

forensic awareness

Higher awareness (e.g. prior arrests) correlates with higher control of evidence.

Geographical Profiling

Determines patterns/correlations in location. An inference can be made about where the offender lives, following three key principles.

Geographical Profiling: least-effort and distance decay

Least-effort: likely to target locations close to their home base. Distance decay: probability of offence decreases with further distance.

Geographical Profiling: circle theory

Canter and Larkin: the pattern of offending forms a circle around the home base. Marauder: acts in proximity to home. Commuter: travels from usual residence.

Geographical Profiling: spatial decision-making

Planned or unplanned crime, modes of transport, employment status, approximate age.

PECS #1

P- Strength = research support for investigative psychology

E- Canter and Heritage (1990): analysis of 66 s£xual assault cases, identified consistent common behaviours

C- However, relies on accurate databases (only solved crimes)

S- So, not helpful for unusual crimes (where there is less data to correlate/compare)

PEEL #1

P- Strength = research support for geographical profiling

E- Lundrigan and Canter (2001): 120 murder cases, location of each body disposal site created a ‘centre of gravity’ with offender’s base inside

E- Stronger effect for those travelling short distances (marauders)

S- So, geographical information can be used

PEEL #2

P- Limitation = geographical profiling insufficient

E- Success of geo. profiling reliant on quality of data

E- Recording of crime not always accurate, varies between police forces, 75% of crimes unreported

L- So, approach might not be useful

PEEL #3

P- Limitation = investigation vs identification

E- Copson (1995): profiling judged ‘useful’ in 83% of cases, but led to accurate identification in only 3%

E- Rachel Nickell: real killer ruled out as judged to be too tall to fit the police profile given

L- So, little practical value in solving cases