Dynamic RFs Lecture Flashcards
Dynamic Risk Factors:
Are __ constructs…
Do RFs denote real psychological processes?
Are “composite constructs” (i.e., they are
general terms which have multiple
subfactors that excert an independent
influence on offending) that are designed
to predict reoffending.
Risk factors do not exist in the sense they
do not refer to or denote a real
psychological process which increases risk
of offending (i.e., it’s an arbitrary category
which summarizes lots of contextual
factors, symptoms/mental states, potential
causal factors).
The distinction between static and
dynamic risk factors are reflected in
current assessment practices and the goal
of risk reduction.
The recruitment of DRFs to formulate
cases and to inform treatment is now
standard in clinical practice.
*the dual purpose DRFs to predict risk of recidivism and used as explanatory variables to explain “why” and “how” the offender’s risk of recidivism is increased and can be managed.
Risk Assessment uses (4) Different Types of Risk Factors:
Monahan & Skeem’s typology of RFs-
- Fixed Marker:
These are fixed and unchangeable risk
factors (i.e., static demographic RFs such
as gender). - Variable Marker:
These are fixed factors which change
overtime but are not amenable to
change (i.e., age). - Variable Risk Factor:
These are risk factors which are variable
and are amenable to intervention (i.e.,
employment). - Causal Risk Marker:
These are potential causal RFs which
are changeable and reduces risk of
reoffending (i.e., DRFs).
(4) Conceptual Problems with DRFs:
- Lack Coherence:
DRFs are composite constructs which
are composed of multiple subtypes of
causal factors which exert an
independent influence on crime and
therefore they’re inherently vague.
The Subtypes of RFs vary in types (i.e.,
contextual factors, mental
state/symptoms, potential causal factors)
and using an overarching DRF
categorical term is misleading and
implies that it is homogenous when it’s
filled with heterogenous factors. - Lack Specificity:
What is the causal mechanism?
The presence of alternative causal
explanations within the same DRF
category means that they’re not useful
for guiding treatment unless theory is
used to tease out the causal
mechanism[s].
DRFs on their own are too vague and
lead to case formulations making
offenders all sound the same (i.e.,
homogenous) when they’re
heterogenous group! - Lack Precision:
DRFs suffer from the Grain Problem.
High-level variable extraction, such as
DRF, has limited utility because it
includes multiple lower-level variables
which requires further analysis to specify
which causal mechanism is important.
What level of abstraction is desirable? It’s
important to specify in advance.
Contextual factors, states, behaviours
and physiological responses. - Lack of Factualness:
DRFs are normative concepts.
DRFs and risk bands are a classification
system aimed to facilitate inductive
inferences and predict risk of future
crime.
They’re legal and social categories
which predict risk of recidivism rather
than explain causal mechanisms that
increase their risk of offending.
*DRFs are predictive not explanatory
constructs!
Example: DRF- Emotional Congruence:
Definition:
Definition:
o Emotional congruence with children refers
to feeling that relationships with children
are more emotionally satisfying than
relationships with adults. Individual who is
emotionally congruent with children may
find children easier to relate to than adults,
may feel he is still like a child himself, and
may believe that children understand him
better than adults do. He often feels
himself to be “in love” with his child victims,
as if the relationship was reciprocal.
o Within this definition there are multiple
lower-level RFs present (i.e., contextual,
state/symptoms, potential causal factors)
that exist independently and what is
present will be different for each individual.
o Which elements are important? What level
of analysis is desirable?
Deconstructing Emotional Congruence:
o Symptoms (behaviours and feelings arising
from contexts and causes):
Feelings of fear and anxiety around
adults.
Engages in courtship/grooming behaviour.
Avoids adult intimacy and relationships.
o Situational/Contextual Factors:
“Contact” with children.
No adult romantic relationships.
Activities with children.
Spending time with other sexual
offenders.
o Causal Factors:
Negative Affective Systems (i.e., the
offender views adults as threatening and
employs avoidance strategies to manage
fear and anxiety). E.g., adults can’t be
trusted.
Positive Affective Systems (i.e., views
children as sources of reward and more
likely to provide him with love sexual
pleasure and care). E.g., sees children
positively as good social and sexual
partners.
Cognitive Systems (i.e., displays
attentional bias toward signs of affection
from children or indicators that it might be
possible to become involved with them
sexually and emotionally). E.g., cues
signifying vulnerability and lack of
supervision.
Intrapersonal Social Processes (i.e., views
himself as vulnerable and unsafe, living in
a dangerous world, lacks understanding
of his motives due to expectancy and
interpersonal biases). E.g., dangerous
world and entitlement etc.
Self-Regulation Systems (i.e., lacks
capacity to soothe themselves and
effectively control their negative
physiological arousal.
Interpersonal Social Systems (i.e.,
dysfunctional internal working models in
which affiliation seeking strategies
directed towards interactions with
children).
May also belong to deviant social
networks that approve of adults having
sex with children.
*all (7) of these are potential causal factors within the DRF of Emotional Congruence.
Key Point:
It’s possible to analyses emotional
congruence with children into different
types of causal processes, contextual
factors, interact to generate symptoms of
DRF (i.e., feelings of fear or loneliness are
common in emotional congruence) which
make them composite constructs that are
markers of cause rather than causal
mechanisms themselves.
DRFs as instigators of inquiry (markers of
causal processes etc.) rather than as
endpoints of inquiry because they require
additional theory to tease out causal
mechanisms and targets for treatment.
Protective Factors:
Experience the same problem of DRFs!
Original Definition:
Criminal Justice Literature:
Three Main Shifts in the Meaning of PFs:
Protective Factors:
Experience the same problem of DRFs!
Original Definition:
o Originated in child maltreatment
literature.
o “A protective factor may influence,
modify, ameliorate, or alter how a
person responds to the adversity that
places them at risk for maladaptive
outcomes.”
o Thought to buffer or protect child from
onset of psychological and social
problems when experience adversity.
Criminal Justice Literature:
o “factors that can compensate for a
person’s risk factors…characteristics of
an offender, or alternatively, his
environment or situation, that reduce
the risk of future violent behaviour.”
o Vague!!!
Three Main Shifts in the Meaning of PFs:
o The protected object (i.e., from the
child/offender to society).
o The source of the potential harm (i.e.,
external ACEs to internal sources of the
offender).
o The time frame concerned (i.e., focus
on past events influencing current
behaviour to current risks predicting
future behaviour).
Moving Forward:
Exclusive focus on risk-related dimensions
provides a description of vulnerability but
cannot explain the purpose of behaviour, a
necessary requirement for both
explanation and intervention.
o RFs need theory to be useful for treatment
and in their current state they only predict
offending and not explain why it occurs.
Timely to provide agency-based model of
depicting the proximate mechanisms those
that are temporally and causally linked to
behaviour resulting in offending.
o They should be deconstructed, and theory
should be used to guide the explanation of
causal mechanisms across multiple levels
to build treatment targets and plans.
Agency Model of Risk (AMR)
Agency Model of Risk (AMR)
Provides a new way to think about risk
Agency (offender who has the capacity to make a decision and acts in the world):
Personal Identity:
vulnerable or misunderstood.
Social Roles:
teacher
System-Levels
their social role gives rise to opportunities
to offend.
Surrounded by:
Goals:
Intimacy with a child.
Strategies:
Groom child, gifts, share feelings.
Implementation:
approach child, strike up a conversation,
refine goals and plans if necessary.
Context:
Social:
Access to children.
Cultural Norms:
Teacher role means its normal to have one
on one interactions with the child.
Physical (offending):
proximity
(6) Conclusions:
Preoccupation with risk prediction and
dynamic risk factors left the field of
criminal psychology theoretically stagnant.
o Reliance on DRFs to predict and explain
crime has led to theoretical illiteracy and
dead ends in literature.
Conflation of predication and explanation.
o DRFs believed to have a dual purpose
when they were developed for prediction
not explanation.
o Risk assessment and Risk Reduction are
different tasks (i.e., predicting future
events or explain past events increase risk
of offenders).
o We cannot expect DRFs to do both tasks.
Confuse risk assessment with risk
reduction.
o Assumes that just because these factors
can predict offending doesn’t mean we
are able to explain why or how this
happens (i.e., correlation does not mean
causation).
o Risk Reduction relies on theory to build
explanations on causal mechanisms of
crime and Risk Assessment relies on
prediction.
AMR: conceptual bridge between
etiological theory (i.e., explanation), risk
prediction and treatment.
o Example of how deconstructing DRFs to
build causal mechanisms theories to
identify causal mechanisms that increase
risk of offending and provide treatment
targets.
Risk assessment reasonably successful
because risk factors refer to problems that
are correlated with offending behavior.
o i.e., adds incremental value of DRFs than
Static RFs but correlation does not mean
causation.
Fine for prediction but does not explain
offending behavior or to guide treatment.
Bradford’s (9) criteria for inferring causal mechanism status:
- There is a strong statistical association
between the risk predictor and the
outcome (i.e., pattern is present). - The relationship is consistent across time,
place, circumstances and observers.
*RFs-Crime association is consistently
found across samples are not due to
chance, bias or confounds (1-2). - The relationship between risk predictor
and outcomes is specific to particular
groups.
*use system level analysis or comparing to
prototypical offender to address the lack
of specificity in DRFs and identify the
causal mechanism. - The right temporal relationship is present
(i.e., risk precedes outcome).
*experimental or longitudinal analysis to
identify temporal relationship
between RF-Crime. - A reasonable biological gradient is
present (i.e., the effect/outcome is smaller
if the causal factor is weaker).
*define the mechanism and its
consecutive parts to be able to identify
said gradient. - The hypothesized causal mechanism is
plausible in light of current scientific
theories and research.
*DRFs lack plausibility because they’re
incoherent, lack specificity, reference
make it hard to pull out causal
mechanisms: various RF types, normative
nature of offense categories makes them
poor explanatory targets: defining them
and linking them to existing theory
should increase their plausibility (6, 7 &
9). - Hypothesized causal mechanism is
congruent with existing theory about this
RF and outcomes. - There are well designed experiments that
support causal relationship.
*manipulate risk to see if it produces
predicted outcome or compare groups
with known deficits to see if their
performance supports predicted
outcome. - Analogous knowledge for similar types of
causal mechanisms support the findings.
Two conceptual frameworks for identifying causal elements in DRFs:
o Deconstructing them into contextual and
causal factors which interact to produce
symptoms/state factors to build causal
mechanism theories.
o Using DRFs as indicators of clinical
phenomena and shift the foci of inquiry
from legal categories to psychological
categories.
Causal mechanisms:
o Are multi-level structures with maladaptive
and adaptive components.
o A structure that it and its consecutive parts
fulfil its function through coordinated
action and are responsible for one or more
phenomena.
o Dynamic and integrated.
o Hierarchal nested multilevel structure
which is sensitive to local contextual
factors and consists of parts and
processes that relies on interrelatedness
of the components to adequately perform
their function.
Integrative Pluralism:
o Constructing multi-level explanations by
incorporating divergent levels of analysis
in one theory because single-level
analysis leads to partial answers (i.e., use
top-down and bottom-up analysis).
It’s important to be clear about the type of
explanation being offered by the theory:
o Etiological Explanation: where the
components and causal mechanisms
result in a down-stream effect (RFs leads
to crime).
o Constitutional Explanation: where the aim
is to describe the components of a single
mechanism and interactions responsible
for the phenomena (single-factor theory).
o Need to clearly define the constructs
being explained.
Deconstruction of DRFs:
o Can differentiate explanatory targets (i.e.,
contextual and symptoms) from potential
causes.
o Can help develop multi-level, integrated,
explanatory models that specify the
different DRFs interact to cause offending.
Are GLM and RNR principles prone to the same conceptual weaknesses as DRFs?
Yes.
