Risk and Protective Factors for Criminal Behavior
Author: Ukrainian Psychological Hub · Published: October 11, 2026 · Editorial Policy
Risk and protective factors for criminal behavior are characteristics of people, relationships, situations, institutions, and environments that are associated with a higher or lower probability of offending. They help researchers understand patterns of crime and help prevention and rehabilitation services identify potentially changeable conditions. Their meaning is probabilistic: a risk factor does not make a person a criminal, and a protective factor does not guarantee that offending will never occur.
Research supports a multilevel account. Previous offending, substance-use patterns, antisocial attitudes, peer networks, family experiences, school engagement, opportunities, and neighborhood conditions may all matter, but their influence varies by age, outcome, setting, and measurement. An umbrella review of youth-crime research synthesized 58 examined factors and rated the evidence for only 11 as highly suggestive or suggestive. That distinction between commonly discussed factors and better-supported associations is the foundation of an evidence-based explanation.
This article examines how risk and protection operate in research on criminal behavior, what can and cannot be inferred about an individual, and how findings inform prevention, rehabilitation, and forensic decision-making. General psychological constructs such as trauma, addiction, personality, and aggression have their own literatures; the focus here is their specific relationship to criminal offending.
What are risk and protective factors in criminal behavior?
A risk factor is a measurable characteristic, exposure, or condition associated with an increased probability of a specified outcome, such as self-reported offending, an arrest, a conviction, or a new offense after release. A protective factor is associated with a reduced probability of that outcome, or, in a stricter theoretical use, buffers the adverse influence of a risk exposure. Research papers sometimes use “protective” broadly for any favorable association; the meta-analysis by Gubbels and colleagues explicitly acknowledges this definitional variation.
A psychological construct is a concept such as self-control, attachment, impulsivity, or prosocial attitudes. Researchers operationalize constructs through questionnaires, interviews, records, observations, or behavioral tasks. A clinical diagnosis is a separately established health classification. Neither a construct nor a diagnosis is equivalent to a criminal act or a judicial determination.
A behavioral observation describes something recorded or witnessed, such as a threat, a school suspension, a prior fight, or compliance with supervision. A risk factor is a statistical relationship established in an identified population. A screening result flags a need for further inquiry. A psychometric score summarizes responses or assessed characteristics according to an instrument's rules. A forensic opinion integrates relevant information within a defined referral question. A legal finding is reached by an authorized decision-maker using applicable legal standards. These categories must remain distinct.
The CDC's youth-violence framework organizes risk and protective factors across individual, relationship, and community levels and emphasizes that risk factors may or may not be direct causes. Youth violence is one outcome within the larger category of criminal behavior, so its findings should be identified as youth-violence evidence rather than generalized to every offense.
Risk factor, cause, marker, and mechanism
An observed association can arise because the factor contributes causally to behavior, because another variable affects both, because the outcome changes the measured factor, or because the research design creates selection or measurement bias. Prior arrests, for example, may statistically predict further arrests while also reflecting differences in surveillance, enforcement, access to counsel, and exposure to criminal-justice institutions. A prediction-oriented marker can be useful without being a proven causal intervention target.
A mechanism is a proposed process connecting conditions to conduct. Social learning, impaired access to legitimate opportunities, situational incentives, emotional regulation, peer reinforcement, and changing social bonds are among the processes examined in criminology. Each proposed pathway requires evidence that tests timing, alternatives, and intervening processes. A correlation alone establishes none of them.
Static factors and dynamic factors
Static factors include events that have already happened, such as age at a previous offense or documented criminal history. They cannot be undone, although their predictive relevance may change with time and circumstances. Dynamic factors can vary, including substance use, current peer associations, housing stability, coping skills, or immediate access to an escalating conflict. Their apparent changeability does not establish that altering them will necessarily reduce crime; intervention evidence is needed.
Some dynamic factors are relatively stable patterns, whereas others fluctuate over hours or days. A threat assessment concerned with an imminent event therefore needs a different time frame and evidence base from a study of ten-year reoffending.
What does the strongest evidence show?
The evidence base varies greatly by population. Youth delinquency and broad antisocial behavior have several substantial syntheses. General adult crime, specific offenses, and people selected into forensic psychiatric services require distinct research. Combining them into a single list of universal “causes of criminality” would obscure important differences.
Findings from the 2024 youth-crime umbrella review
Ayano and colleagues' umbrella review reported highly suggestive associations for substance use, history of crime, moral development measures, psychopathology measures, adverse childhood experiences, poor parental supervision, maltreatment or neglect, attachment measures, and school bullying. Peer pressure and supportive school environments met a suggestive level of evidence. The review reported an odds ratio of 2.29 for substance use, 2.03 for previous crime, and 0.56 for supportive school environments.
Those odds ratios compare odds in groups under the analyses used by the underlying studies. They are not the probability that a particular young person will commit a crime. Measures of “psychopathology” differed across studies and cannot be translated into a claim that people with a mental health diagnosis are dangerous. Some constructs, including attachment and moral development, have measurement and coding complexities; the review's labels require reading in their original study context.
The synthesis is strongest as a map of youth-crime associations. It does not establish causal effects of removing any one factor, and it should not be repurposed as a validated adult sentencing score or as a prediction of a specific person's behavior.
Findings from the protective-factors meta-analysis
Gubbels, Assink, and van der Put synthesized 305 studies of protective factors for antisocial behavior in youth and young adults, examining 1,850 candidate factors grouped into 77 domains. Fifty domains had statistically significant negative associations with antisocial behavior. Notable domains included prosocial peers and values, relationship quality, social skills, resilience, life satisfaction, self-reflection, school self-esteem, and parental control.
The authors' outcome was broad antisocial behavior, encompassing activities from rule-breaking and bullying to delinquency and violence. Their data cannot be treated as a direct meta-analysis of criminal convictions alone. The included designs were also heterogeneous, including cross-sectional and retrospective studies, so associations are not automatically evidence that programs increasing a score will prevent an offense.
The study found moderators involving age, gender, and behavioral severity for some domains. This is an additional reason to avoid applying a youth sample's average effect to an adult, a different offense type, or a population with different resources and constraints.
Individual and behavioral risk factors
Prior offending and behavioral history
A documented history of offending is frequently associated with later offending. In the youth umbrella review, previous crime was among the factors with more substantial evidence of association. A meta-analysis of persistent delinquency likewise found comparatively large associations for criminal history, aggressive behavior, and alcohol or drug abuse domains.
History provides information about past patterns and contexts, yet “criminal record” is not a neutral or complete measure of past conduct. It depends on whether behavior was detected, recorded, prosecuted, and adjudicated. A previous allegation, an arrest, a conviction, and self-reported conduct are different variables. For a given case, the type, recency, seriousness, opportunity, and circumstances of earlier behavior may be more informative than a simple count.
Substance use and substance-related environments
Alcohol and drug use can be associated with offending through several routes: intoxication-related disinhibition in some circumstances, illegal market exposure, acquisitive pressures, social networks, shared socioeconomic stressors, and preexisting behavioral difficulties. The relevant process differs between substance possession, impaired driving, interpersonal violence, and property offenses.
The 2024 umbrella review found a notable aggregate association for substance use in youth-crime research. That evidence does not support a claim that use of any substance produces a fixed criminal disposition. Clinical substance-use disorder, an episode of intoxication, and participation in an illicit market are separate constructs with different implications.
Impulsivity, self-regulation, and antisocial attitudes
Difficulties delaying gratification, controlling behavior under stress, evaluating consequences, or disengaging from conflict have been investigated in offending research. Antisocial attitudes and beliefs can also shape how someone interprets harmful conduct and available choices. These are dimensional characteristics that vary with development and circumstances.
An impulsive response is not synonymous with a crime. Many people with attention or self-regulation difficulties never offend, and many offenses are deliberate, organized, or embedded in situational incentives. The CDC's youth-violence inventory lists several behavioral and cognitive correlates, but that inventory is an evidence-informed prevention framework, not an individual diagnostic test.
Adverse experiences, maltreatment, and exposure to violence
Childhood adversity, neglect, maltreatment, and exposure to violence have been associated with later antisocial and offending outcomes in many studies. They may affect stress responses, trust, access to support, school trajectories, behavioral learning, and opportunities over time. Adversity also co-occurs with poverty, unstable housing, and unequal access to services, complicating causal interpretation.
The youth umbrella review identifies adverse childhood experiences and maltreatment among factors with highly suggestive evidence of association. The same evidence requires a humane interpretation: a person who experienced abuse remains a person harmed by an adverse condition, not a future offender designated by that history. Most exposed people cannot be classified from exposure alone, and the possible pathways depend on protective relationships, resources, development, and later circumstances.
For detailed information on the consequences of witnessing violence, see Childhood Exposure to Domestic Violence: Effects, Risk, and Resilience. That clinical-developmental topic should not be confused with a forensic prediction.
Mental health and psychological symptoms
Psychological distress, symptoms, and diagnoses may appear in some offending studies, but a diagnosis is not a criminal-behavior mechanism and cannot substitute for an examination of conduct and context. Associations vary widely by condition, co-occurring substance use, exposure to victimization, deprivation, and whether the outcome is violence, general offending, or police contact.
Research terms such as “psychopathology,” “externalizing symptoms,” and “mental disorder” are not interchangeable. A screening score does not establish a diagnosis; a diagnosis does not establish dangerousness or criminal responsibility; and criminal responsibility is a legal determination governed by jurisdiction-specific criteria. The APA's Specialty Guidelines for Forensic Psychology emphasize legally relevant functioning, suitable methods, corroboration, and limits of clinical inferences in forensic work.
Developmental and neurobiological correlates
Attention, learning, neurocognitive development, stress physiology, and executive functioning can be studied as possible contributors to behavioral trajectories. Their meaning depends on age and environment. The National Institute of Justice's review of biosocial factors describes a developing research agenda in which neuropsychological functioning and stress-system response are considered alongside social conditions.
Biological or genetic associations do not identify an inevitable behavioral destiny. No brain image, gene, or isolated physiological marker establishes who committed an offense. Claims about intervention benefits need stronger evidence than the observation of a group difference.
Family, peer, school, and community factors
Family relationships and supervision
In studies involving children and adolescents, inconsistent supervision, exposure to conflict, neglect, and harmful disciplinary environments frequently appear among risk correlates. Supportive communication, predictable boundaries, reliable caregiving, and constructive adult involvement appear among possible protective conditions.
The term “poor parental supervision” is a population-level research variable, not a finding that a particular parent caused a crime. Parents' opportunities to provide supervision can be constrained by employment, housing instability, health, discrimination, and availability of safe childcare. Risk explanations become inaccurate when structural constraints are reduced to individual blame.
The Office of Juvenile Justice and Delinquency Prevention's child-delinquency bulletin describes how accumulating risk domains can be associated with earlier offending, while also emphasizing prevention opportunities.
Peer networks, reinforcement, and belonging
Peer association can affect opportunities to offend, social rewards for particular behavior, definitions of acceptable conduct, and the practical likelihood of participation in group activity. Repeated contact with peers engaged in delinquency is often measured as a risk correlate, whereas prosocial peers can support alternative norms and opportunities.
Peer effects are difficult to interpret because people also select friends with whom they share attitudes and circumstances. Observed similarity may reflect social influence, selection, shared context, or all three. The youth umbrella review rated peer pressure as suggestive evidence rather than proof that contact alone causes crime.
School connection and achievement
School can provide supportive relationships, routine, skills, credible future opportunities, and access to services. School exclusion, disengagement, and poor achievement may correlate with offending, but the direction of influence can run both ways. A student struggling with behavior can be removed from school, while exclusion can itself interrupt protective connections.
In Ayano and colleagues' synthesis, supportive school environments showed an association in the protective direction. Academic performance is a useful descriptive measure in some samples, but success at school is shaped by disability accommodations, instructional quality, language, family resources, and institutional practices. The prevention question is how to build inclusion and support, not how to attach a criminal label to students with low grades.
Housing, employment, neighborhood conditions, and opportunity
Stable housing, fair access to education and employment, community organizations, accessible health care, and safe places to spend time can influence behavior through routine, opportunity, security, and the alternatives available to people. Concentrated deprivation, instability, and exposure to violence may coincide with greater offending or criminal-justice contact.
Community conditions should be interpreted with particular attention to what was measured. A neighborhood arrest rate also reflects policing and enforcement decisions, not solely the underlying prevalence of criminal acts. Environmental correlations cannot justify treating residents of an area as presumptively risky.
How risk factors combine over time
Risk factors rarely occur one at a time. An adolescent might experience unstable housing, conflict at home, disrupted schooling, and peers who reward rule-breaking, while simultaneously having a trusted adult, supportive teacher, and future plans. The same profile can change substantially within a year.
Cumulative-risk approaches examine whether multiple exposures are associated with increased outcome probabilities. The OJJDP review describes this accumulation. It does not follow that all factors have equal weight, that their effects add linearly, or that a simple count accurately predicts a particular offense.
Timing matters. Early childhood, adolescence, emerging adulthood, and later adulthood have different developmental tasks and opportunities. Risk can also be reciprocal: school disengagement may precede offending, while arrest or incarceration may subsequently worsen educational and employment access. A rigorous explanation asks when the exposure occurred, what alternatives were considered, and how outcomes were measured.
Protective factors may have direct associations with lower offending, may buffer particular exposures, or may provide practical routes to change. The distinction is empirically important. Gubbels and colleagues aggregated broad favorable associations and cautioned that terminology varies; they did not demonstrate that every included domain statistically moderated the impact of risk.
Protective factors that may reduce offending
Prosocial relationships and trusted adults
Reliable and respectful relationships can provide practical support, expectations, belonging, accountability, and help navigating difficult circumstances. In youth research, connections with prosocial peers and adults recur among protective associations. The strength of those connections depends on their quality and safety, not on whether someone is formally described as part of a household.
Social skills, reflection, and self-regulation
The ability to recognize conflict escalation, consider others' perspectives, plan alternatives, and seek help may create more options at moments of stress. Meta-analytic findings concerning social skills, self-reflection, prosocial values, and resilience support their relevance in youth samples. They also show that broad constructs are assessed in many different ways, so a single exercise or course cannot be assumed effective merely because its label matches a protective factor.
Supportive schools and meaningful opportunities
Educational inclusion, employment pathways, accessible training, constructive leisure activities, and organizations offering stable support may strengthen community participation. These conditions often operate together. A program that teaches decision-making without addressing an inaccessible school, unsafe housing, or a lack of legitimate income opportunities may leave major drivers of risk untouched.
Treatment, stable housing, and reentry support
When a person has a treatable health condition, receives support that fits their needs, or gains stable housing and employment after incarceration, these changes can matter for well-being and participation. Whether a particular intervention reduces recorded recidivism is an additional empirical question requiring follow-up and comparison groups.
The National Institute of Justice's desistance synthesis emphasizes the roles of development, relationships, institutions, and opportunities in the process of reducing and ending criminal activity. Desistance is a process measured across time, not merely the absence of an arrest over a short period.
Why protective factors are not simply the opposite of risk
Low school engagement and high school engagement can sometimes be represented as opposite ends of one dimension. Other protective resources, such as a committed mentor or stable housing, may exist alongside serious risk exposures. The same protective factor may have different associations at different ages or in different groups.
Researchers distinguish a promotive effect, in which a favorable factor is associated with a better outcome generally, from a buffering or moderating effect, in which a factor changes the strength of a particular risk-outcome association. A study showing that social support correlates with lower offending has not necessarily shown that support neutralizes the effect of childhood maltreatment.
This distinction changes prevention design. A strategy can reasonably aim to build broad protective resources while remaining open about which mechanisms and risk interactions have actually been tested.
Do risk factors cause crime?
Some risk-related pathways are plausible causal targets, but causation must be established with stronger designs than cross-sectional correlations. Longitudinal studies improve temporal ordering; natural experiments and well-designed intervention studies can strengthen causal inference; randomized trials can test specific programs under defined conditions. Each design still has limitations, including attrition, measurement changes, implementation variability, and the possibility that an intervention affects several mechanisms at once.
Even a causal contributor rarely explains an entire offense. Human actions occur within situations that include immediate opportunities, constraints, beliefs, relationships, and choices. Different kinds of crime involve different pathways. A model developed for adolescent general delinquency should not be treated as an established account of financial fraud, stalking, terrorism, homicide, or every form of violence.
A scientifically responsible answer to “what causes criminal behavior?” therefore identifies multiple possible pathways and specifies which claims concern prediction, which concern association, and which have intervention-based causal evidence.
Do risk factors predict which individual will commit a crime?
Prediction studies ask whether a set of variables helps rank or estimate future outcomes in a specific population over a specified time. Their results are conditional on who was included, the outcome definition, the time horizon, and how the instrument was used.
Discrimination measures whether a model tends to give higher estimated risk to people who later experience the measured outcome than to people who do not. Calibration asks whether observed outcome rates correspond to predicted probabilities. Sensitivity is the proportion of actual outcomes correctly flagged; specificity is the proportion of non-outcomes correctly identified. Positive predictive value is the proportion of people flagged who go on to have the measured outcome. These statistics answer different questions.
A simple example of false positives
Imagine a hypothetical group of 1,000 people in which 10% have the defined outcome during the follow-up period. If a screening rule has 70% sensitivity and 80% specificity, it identifies 70 of 100 people with the outcome, misses 30, correctly identifies 720 of the 900 without the outcome, and flags 180 people who do not have it.
Of the 250 people flagged, 70 have the outcome. The positive predictive value is 28%. This is an arithmetic illustration, not a measured result for a particular criminal-justice tool. It shows why apparently respectable sensitivity and specificity can still produce many false positives when the outcome is uncommon.
A major systematic review in BMJ found that violence-risk tools often had substantially lower positive than negative predictive values and cautioned against their use as sole determinants of detention, sentencing, or release. The review's samples and outcomes were diverse and its reported values must not be transplanted into unrelated settings.
A 2023 Lancet Psychiatry systematic review of post-discharge violence-risk instruments in forensic psychiatric populations examined 50 studies covering 10,460 people. Most studies emphasized discrimination while omitting other essential metrics, and 49 of 50 were rated at high risk of bias. That specialized population is not representative of all people with mental disorders or the general public.
Recidivism is an outcome definition, not an inherent trait
Reoffending can be operationalized as a new arrest, conviction, imprisonment, violation, or self-reported behavior. These measures have different sensitivity to criminal activity and institutional decision-making. Follow-up length, release conditions, local law, policing, data linkage, and reporting practices all influence the measured rate.
A “low-risk” score does not confer immunity from future behavior. A “high-risk” score does not establish that any crime will occur. A risk-factor inventory is therefore useful primarily as one component of appropriately validated assessment and risk management rather than as proof of guilt or a fixed identity.
Fairness, subgroup performance, and cultural limits
Evidence from one group may not transport reliably to another. Age, sex, gender, disability, language, cultural context, migration, socioeconomic conditions, and jurisdictional practices can affect both the measurement of predictors and the recording of outcomes. A model trained mainly on men in one justice system may require independent validation before any use with women, adolescents, or another jurisdiction.
Fairness requires examining missing data, measurement invariance, subgroup discrimination, calibration, false-positive and false-negative rates, and the consequences of decisions. Some statistical fairness goals can conflict, especially when observed base rates differ. The relevant ethical and legal judgment cannot be supplied by an AUC alone.
The APA forensic guidelines emphasize population-appropriate assessment, cultural and language competence, multiple information sources, and explicit communication of limitations. Such safeguards are particularly important when assessments affect liberty, supervision, treatment access, or family circumstances.
How are risk and protective factors used in forensic assessment?
An appropriate forensic assessment begins with a defined question: for example, the likelihood of a specified behavior in a particular period and setting, the circumstances relevant to managing a threat, or an identified rehabilitation need. The evaluator then considers records, interviews, collateral information, verified behavior, protective resources, and the validity of any selected measures.
A structured instrument may organize information, but the choice of tool must match the population, outcome, and purpose for which evidence exists. Reported scores require interpretation within their uncertainty and base rates. Historical events, current needs, situational triggers, and protective resources should be distinguished in the explanation.
In a legal case, the assessment's research conclusions cannot replace the applicable legal test. Competency, criminal responsibility, evidentiary admissibility, sentencing factors, detention conditions, and legal capacity are governed by the law of the jurisdiction and by the relevant decision-maker. The same psychological fact can have different legal significance in different jurisdictions.
The professional task is to make the limits of inference clear, not to translate psychological probability into an assertion of guilt, an insanity determination, or a future-crime certainty.
How can prevention and rehabilitation address these factors?
Prevention can operate at several levels. Universal approaches improve the conditions of schools, families, and communities without assigning children a criminal identity. Selective approaches provide additional resources where particular exposures or unmet needs are identified. Individualized interventions for justice-involved people can address behaviorally relevant needs while strengthening relationships, housing, education, health, and opportunities for participation.
The risk-need-responsivity framework organizes correctional intervention around the level of risk, criminogenic needs, and characteristics affecting a person's ability to benefit from services. It is influential in corrections, but an intervention's effectiveness still needs to be established empirically rather than inferred from adherence to a model.
A 2023 meta-analysis of community correctional programs found a favorable association with full program participation, while reporting substantial heterogeneity and low study quality. Programs fully adhering to risk-need-responsivity principles did not clearly outperform others in that synthesis. Participation itself may also reflect selection and motivation, so its association is not automatically a causal estimate.
Successful practice therefore needs both evidence on the target factor and evidence on the proposed intervention, supported by reliable implementation and appropriate follow-up. Monitoring should include harms, access, completion, participant experience, and meaningful outcomes rather than solely administrative recidivism counts.
Criminal behavior, juvenile delinquency, violence, and desistance: different questions
General criminal behavior is a broad category shaped by legal definitions and enforcement practices. Juvenile delinquency refers to youth conduct under the applicable juvenile-justice framework. Violence is a particular behavioral outcome and cannot be assumed to share every correlate of nonviolent offending. Recidivism measures new outcomes after an earlier justice-system event. Desistance studies the process of moving away from offending over time.
These distinctions define the boundary of this article. Developmental pathways to offending, juvenile-specific risk, violence-risk instruments, and correctional treatment warrant their own dedicated analyses. The present synthesis explains the shared risk-and-protection framework while preserving those separate questions for their respective forensic topics.
Common myths and misconceptions
“A difficult childhood makes someone a criminal”
Adversity can be associated with increased risk, but a history of trauma does not determine criminal behavior. Protective circumstances, development, social institutions, individual differences, and later opportunities matter. Retrospectively identifying trauma in one person who offended does not show that trauma caused the offense.
“A diagnosis explains why a crime happened”
A clinical diagnosis describes a health-related classification. Establishing what happened, why a behavior occurred, whether someone met a legal mental-state standard, and what intervention may help are different questions requiring different evidence.
“One factor is enough to predict future crime”
A single factor rarely supports reliable individualized prediction. Even multivariable tools need calibration, validation, relevant base rates, and attention to false positives and false negatives.
“The absence of risk equals protection”
Some protective factors are independently meaningful resources. Their presence and effects require evidence, and favorable correlations cannot automatically be interpreted as causal protection.
“A risk score proves that a person will offend”
A score is a summary of information under a measurement system. It does not establish a future event, a past offense, criminal responsibility, or the legal propriety of detention.
“More punishment always reduces risk”
Short-term incapacitation, deterrence, rehabilitation, social disruption, and longer-term desistance are different possible effects. Their balance depends on the policy, population, and outcome. The NIJ desistance review highlights why long-term reintegration and social conditions matter.
Frequently asked questions
What is the biggest risk factor for criminal behavior?
There is no universally largest factor across all offenses, ages, and legal systems. Prior offending and certain behavioral histories often have strong predictive associations in selected samples, while the evidence for youth additionally highlights substance use and contextual exposures. A factor's rank depends on its definition, outcome, follow-up, and the study population.
Are risk factors the same as causes of crime?
No. Risk factors are variables associated with changes in outcome probability. A causal explanation needs additional evidence about mechanisms, timing, alternative explanations, and ideally intervention or strong quasi-experimental findings.
What is the most important protective factor against delinquency?
The literature identifies several favorable domains, including prosocial peers, supportive relationships, self-regulation, and school connections. Their relative associations differ across age and outcome. The 2024 protective-factors meta-analysis found multiple domains rather than one universally dominant protective factor.
Can criminal behavior be predicted in children?
Youth research can identify population-level associations, but it cannot reliably identify an inevitable future offender from childhood characteristics. Development is variable; conditions and opportunities change. Prevention should support children and families without labeling children as prospective criminals.
Does mental illness increase the risk of crime?
That question is too broad for a single group comparison. Specific symptoms, conditions, co-occurring substance use, exposure to adversity, outcome definitions, and contexts need to be separated. Clinical diagnosis alone cannot establish individual dangerousness, a criminal mechanism, or legal culpability.
What are dynamic risk factors?
Dynamic risk factors are variables that can change, such as current substance use, peer association, housing stability, conflict situations, or certain attitudes and behaviors. Changeability makes them plausible intervention targets, but proof that modifying them reduces reoffending depends on intervention evidence.
Are protective factors relevant after someone has offended?
Yes. Safe relationships, education, employment, housing, health care, behavioral change, and opportunities for prosocial participation may support rehabilitation and desistance. Their specific effectiveness requires study rather than assumption.
How do courts use criminal risk factors?
Courts and agencies may consider risk information in some jurisdictions and decision contexts, but legal admissibility, required procedures, authorized uses, and the weight of such evidence vary. A forensic evaluator supplies a qualified scientific opinion; a court applies the governing law. Neither diagnosis nor risk score alone settles a legal finding.
Do crime statistics show the true amount of offending?
No single data source captures all criminal behavior. Arrests and convictions are administrative outcomes, whereas victimization surveys and self-reports capture different dimensions. Differences in reporting, enforcement, definitions, and observation periods affect conclusions about risk.
Related articles
For an in-depth account of how violence exposure affects development without turning trauma into a forensic label, read Childhood Exposure to Domestic Violence: Effects, Risk, and Resilience. For clinical context on long-term experiences, see Childhood Emotional Abuse: What It Is, Examples, and Long-Term Effects. For evidence on the role of interpersonal support, read Social Support and Stress: When Relationships Buffer Strain.
Related Articles
References
American Psychological Association. (2011). Specialty guidelines for forensic psychology. APA.
Ayano, G., et al. (2024). Risk and protective factors of youth crime: An umbrella review of systematic reviews and meta-analyses. Clinical Psychology Review, 113, 102479. https://doi.org/10.1016/j.cpr.2024.102479.
Centers for Disease Control and Prevention. (2024). Risk and protective factors: Youth violence prevention. CDC.
Duan, W., Wang, Z., Yang, C., & Ke, S. (2023). Are risk-need-responsivity principles golden? A meta-analysis of randomized controlled trials of community correction programs. Journal of Experimental Criminology. https://doi.org/10.1007/s11292-022-09550-w.
Fazel, S., Singh, J. P., Doll, H., & Grann, M. (2012). Use of risk assessment instruments to predict violence and antisocial behaviour in 73 samples involving 24,827 people: Systematic review and meta-analysis. BMJ, 345, e4692. https://doi.org/10.1136/bmj.e4692.
Gubbels, J., Assink, M., & van der Put, C. E. (2024). Protective factors for antisocial behavior in youth: What is the meta-analytic evidence? Journal of Youth and Adolescence, 53, 233–257. https://doi.org/10.1007/s10964-023-01878-4.
National Institute of Justice. (2021). Biosocial factors and their influence on desistance. NIJ.
National Institute of Justice. (2021). Desistance from crime: Implications for research, policy, and practice. NIJ.
Office of Juvenile Justice and Delinquency Prevention. (2003). Risk and protective factors of child delinquency. OJJDP.
Office of Juvenile Justice and Delinquency Prevention. (2015). Model Programs Guide literature review: Protective factors against delinquency. OJJDP.
Office of Juvenile Justice and Delinquency Prevention. (2015). Model Programs Guide literature review: Risk factors for delinquency. OJJDP.
Ogonah, M. G. T., Seyedsalehi, A., Whiting, D., & Fazel, S. (2023). Violence risk assessment instruments in forensic psychiatric populations: A systematic review and meta-analysis. The Lancet Psychiatry, 10, 780–789. https://doi.org/10.1016/S2215-0366(23)00256-0.