The purpose of risk assessment is to predict future crime and manage offender risk throughout the course of the criminal justice process. Risk assessment is the process whereby offenders are assessed on several key variables empirically known to increase the likelihood of committing an offense.
These variables, known as risk factors, are subdivided into static and dynamic factors. Both types of factors are known to share a causal relationship with criminal behaviour, however, static factors are historical and unchangeable, while dynamic factors are current and changeable. Examples of static factors include age at first offence, history of prior convictions, gender, type of victim, and motivation for committing past crimes. Examples of dynamic factors include present economic situation, marital status, attitudes supportive of crime, faulty cognitions, sexual deviant preference, family condition, leisure activities, criminal friends, substance abuse, and employment status. Examples of factors not empirically related to crime (but often mistakenly associated with crime) include self-esteem, depression, and fear of official punishment.
Concurrent with one of the major themes of the Psychology of Criminal Conduct, individual differences and heterogeneity among offenders means that things are not always so simple. Some risk factors for one type of offender may not be the same for another. For example, juvenile delinquents are heavily influenced by their family life during development, and improper parenting practices, such as physical or mental abuse and inconsistently strict punishment, significantly predispose a child to involvement in crime. However, these same factors are obviously absent with adult offenders (or most of them, at least). Similarly, offenders with certain personality disorders or mental disorders will possess risk factors that do not exist in non-mentally-disordered offenders.
The assessment of sex offenders, in addition, include items neither applicable nor valid to non-sexual offenders, such as sexual deviant preference ratings measured by phallometric assessment. Issues such as these form the core content of the psychology of criminal conduct, a theory committed to recognizing the uniqueness of individuals and opposed to relying on the stereotypical viewpoint that all offenders represent one single population inherently different than the rest of us. (See the section entitled Correctional Rehabilitation, which specifies the different treatment targets for different offenders.)
Correctional personnel typically approach risk assessment in one or more of three ways: through clinical opinion, actuarial prediction, or structured-professional judgment; a third type of approach that is becoming more popular today is called the anamnestic approach, but the former three approaches are the most common. Clinical opinions have historically fared quite poorly in terms of accuracy, with an effect size of .52 compared to matched controls (with complete chance being at .50). While some clinical practitioners can achieve greater than .52 (and some undoubtedly less), clinical opinion is still dependent on the competence and acuity of the assessor, not the assessing instrument. Because of this it is inherently unreliable. It may be more suited to on-the-fly assessments, where speed and cost are more of an issue than accuracy and reliability.
Actuarial prediction involves a strictly evidence-based selection of only those risk factors empirically related to the cause of criminal behaviour. These risk factors, known as "criminogenic needs," construct the basic building blocks, or precursors, to antisocial behaviour, and can include poor parenting, impulsivity, substance abuse, criminal peers, and attitudes that condone the use of violence or justify the breaking of law. Through the use of meta-analytical studies (large-scale reviews that integrate a number of individual, independent studies' findings to determine an average effect-size across all of them), criminogenic needs can be identified, and placed accordingly into the risk assessment scale. Working from the same principle, but combining certain elements of the 1st generation approach, is the 3rd generation approach known as "structured professional judgment" (SPJ).
The SPJ approach combines the objective evidence-based selection of preset and predetermined risk factors of 2nd generation instruments with the subjective, professional interpretation of the severity, frequency, or duration of those predetermined risk factors by the individual clinician. While slightly less accurate that 2nd generation approaches, Structured Professional Judgment allows the clinician to take into account individual-specific details of the case. For example, all risk factors should in many cases not be treated equally for all individuals. While it is likely for most people, on the average, that a history of criminal violence will lead to a future of criminal violence if no treatment is administered, that same history of violence may not be so likely a factor in someone who has recently married, fathered children, and landed a highly secure, well paying job. In this instance, history would be weighted less than someone who had just recently been paroled for manslaughter, received little or no treatment in prison, and was not complying with his parole conditions. Structured Professional Judgment states that is more efficient and accurate to assess individuals on a case-by-case basis, while identifying dynamic, changing, or current risk factors that could be more easily implemented into a treatment plan. In addition, many risk factors, although applicable to population at large, possess very little utility when attempting to assess special or rare populations offenders, such as frotteurists, exhibitionists, or adolescent substance abusers.
Risk Assessment is used in several different stages in the criminal justice process, the most important contexts being courts, sentencing, prison classification, prison reclassification, and conditional release.
Risk Assessment instruments have proliferated in the past two decades until virtually every sub-population of offender can have a specific assessment instrument that targets his or her specific needs. A sample of some of the more popular ones are listed below.
SONAR (Sex Offender Need Assessment Rating)
LSI-R (Level of Service Inventory-Revised)
SAVRY (Structured Assessment for Violence Risk Among Youth)
SARA (Spousal Abuse Risk Assessment)
CRS (Custody Ratings Scale)
SIR-R (Statistical Inventory on Recidivism-Revised)
Applications in Courts
Risk Assessment for the use in courts is gaining popularity, even though its effective influence on jury-decision making is still under debate. In order for testimony from an expert witness to be accepted in a Canadian court of law, four guidelines must be met.
· Testimony must be made by an expert
· Testimony must offer some predictive or probative value to the judicial decision
· Testimony must be based on reliable measures or evidence
· Testimony must impart information that in some way expands the knowledge of the court
Applications in Civil Settings
Risk Assessment is also used widely in the civil setting, as opposed to a criminal setting either in the courts or in correctional institutions. It is used in civil commitment cases, where an individual with an acute or severe mental illness requires commitment to a mental health facility because the individual represents a danger to herself or others. It is also used in child protection cases by organizations such as the Childrens Aid Society, where a child is being abused or neglected by his or her caregivers, and it is necessary to assess the risk posed to that child by his or her caregivers. It is also used to screen for immigration applicants, and can be used in school or labour organizations.
Applications during Sentencing
Risk assessments were first required in Canada when dangerous offender legislation was introduced in 1977. Since then there has also been long-term offender designation, similarly requiring risk assessments for violence by a mental health practitioner. Risk assessment is also considered when an offender is applying for conditional release or nearing his or her eligibility dates. Canadian serial killer, Paul Bernardo, for example, scored 36 on the PCL-R checklist at the time of his prosecution, while Clifford Olson scored 37.
It is also used in correctional settings, during initial classification with an instrument such as the Custody Rating Scale. This instrument encodes several risk factors under the headings of institutional adjustment, escape risk, and public safety to make a final prediction of how likely that offender will engage in prison misconduct. This determination of risk is crucial when determining a security level. It is also used in reclassification, as offenders often change many skills, attitudes, and personality features as they progress throughout their sentence. For example, the Level of Service Inventory Revised (LSI-R) is used during prison reclassification to more efficiently place offenders to security levels. This scale substitutes many of the static factors of the Custody Rating Scale (during initial classification) for more dynamic, recent factors such as the offenders motivation, performance in treatment, or obedience to rules.