Strengths of the top-down approach to offender profiling:

• Supporting evidence: Gregg McCrary and Ed Grant used the top-down approach to develop a profile of the man responsible for the murders of several prostitutes in the late 1980s. This profile ended up being a close match for the offender, Arthur Shawcross, who was eventually found guilty. This suggests the top-down approach works.
• Based on evidence: The organised and disorganised offender profiles were developed by the FBI from interview and data from 36 US murderers, including Ted Bundy and Charles Manson.

Weaknesses of the top-down approach to offender profiling:

• Conflicting evidence: Pinizzotto and Finkel (1990) compared the accuracy of profiles created by 6 US-trained profilers (using a top-down approach) against profiles created by control groups. In a task that required participants to create a profile for a sexual assault case, the profiles created by the trained profilers were no more accurate than profiles created by the control groups.
• Limited applications: The top-down approach only works for some types of crime. For example, rapists and murderers typically have a particular modus operandi (i.e. way of committing the crime) that reveal information about their character and so for these crimes the top-down approach may be useful. However, crimes such as burglary do not typically reveal any idiosyncratic details about the criminal and so bottom-up approaches (e.g. geographical profiling) are likely to be more useful.
• Overly simplistic: The binary of organised offender and disorganised offender may be too simplistic as it’s likely many offenders won’t fit neatly into either category. For example, it’s possible that a high IQ person could commit a spontaneous and unplanned murder in a fit of rage. However, high IQ is a characteristic of the organised profile, but a spontaneous crime is characteristic of the disorganised profile and so sticking too rigidly to these offender profiles could lead to inaccurate profiling.

Strengths of the bottom-up approach to offender profiling:

• Supporting evidence:
  • Investigative psychology: David Canter used his bottom-up profiling methods to develop a profile of the ‘Railway Rapist’ – responsible for several rapes and murders of women near railway stations in south-east England in the 1980s. Canter developed the profile of a man in his mid-late 20s, with a criminal record, working in a semi-skilled job, who had a poor relationship with women, knowledge of the railways, and lived near the crime scenes. This profile turned out to closely match the details of the offender, John Duffy, who was found guilty of these crimes.
  • Geographical profiling: In addition to Canter and Larkin (1993) above, several studies support the validity of geographical profiling. For example, Canter and Lundrigan (2001) plotted the locations where 120 serial killers disposed of bodies and analysed this data using a statistical technique called smallest space analysis. The researchers found the killers’ homes tended to be in the centre of the plotted area where they disposed of bodies, which supports the validity of geographical profiling.
• Wider range of applications: Whereas the top-down approach only really works for crimes with a particular modus operandi, such as rape and murder, the bottom-up approach can be applied to a much wider range of crimes. For example, most burglaries are similar in method and so a top-down approach won’t reveal much about the offender’s profile. However, geographical profiling only requires the locations of the crimes, and so this bottom-up approach can be applied to basically every type of crime.
• More scientific: The bottom-up approach relies on objective and measurable data (e.g. plotting geographic locations of crimes), uses mathematical tools (e.g. statistical analysis), and is often based on psychological theory (e.g. interpersonal coherence). Because of this, Canter argues the bottom-up approach is more scientific than the top-down approach, which relies more on intuition and gut-feeling.

Weaknesses of the bottom-up approach to offender profiling:

• Conflicting evidence: Copson (1995) surveyed 184 UK police officers (UK = bottom-up approach) on the use of offender profiles created by trained profilers. Although 83% of the police surveyed said the profiles were ‘useful’, just 3% of profiles created by trained profilers resulted in identification of the offender. This suggests bottom-up profiles are not particularly useful in practice.

Strengths of the atavistic form explanation of criminal behaviour:

• Historical influence: Even though Lombroso’s atavistic form explanation is widely dismissed by modern-day scientists, it played an important role in advancing criminology and scientific explanations of criminal behaviour. Prior to Lombroso, explanations of criminal behaviour tended to be religious (e.g. bad spirits, Satan, etc.) or moralistic (e.g. weak-mindedness). In appealing to evolutionary reasoning, Lombroso’s approach shifted the discussion towards more scientific explanations. This paved the way for more scientific explanations of criminal behaviour, such as genetic factors.

Weaknesses of the atavistic form explanation of criminal behaviour:

• Methodological concerns: A key methodological flaw in Lombroso’s approach was that he did not use a non-criminal control group to compare his measurements of criminals against. Without comparing the features of criminals against the features of non-criminals, it is impossible to say whether criminals do actually have distinctive features that differentiate them from non-criminals as Lombroso claimed.
• Alternative explanations: Even if it were true that criminals do have distinctive facial structures, there might be other causal reasons for this besides having evolutionarily primitive (atavistic) biology. For example, if someone is ‘ugly’, this could cause other people to treat them badly and this bad treatment could cause a person to engage in criminal behaviour. In this case, the link between physical appearance and criminal behaviour would be the result of environmental factors rather than biological factors (i.e. atavistic form).
• Ethical concerns: Lombroso’s atatvistic form explanation is socially sensitive because it could lead to stereotyping and discrimination based on the way a person looks.

Strengths of genetic explanations of criminal behaviour:

• Supporting evidence: See studies above – in particular the twin and adoption studies.

Weaknesses of genetic explanations of criminal behaviour:

• Other factors: If criminal behaviour was entirely determined by genetics, the concordance rate would be 100% among identical twins. However, both Christiansen (1977) and Lange (1931) found the concordance rates for criminal behaviour among identical twins to be less than 100%, which demonstrates that other factors (e.g. different experiences and psychological factors) are needed for a complete explanation of criminal behaviour.
• Methodological issues:
  • Twin studies: The usual methodological concerns with twin studies apply to Christiansen (1977) and Lange (1931). For example, it’s possible that identical twins are more likely to be treated similarly than non-identical twins due to their identical appearance. This similarity in environment may be (at least partly) responsible for the higher concordance rates for criminality among identical twins than non-identical twins.
  • Adoption studies: There are also methodological issues with adoption studies like Mednick et al (1984), too: Adoptees are often raised by their biological parents for a long time before being adopted, so the correlation between having a biological parent who is a criminal and becoming a criminal could be due (at least in part) to environmental factors rather than genetics.
• Deterministic: Genetic explanations are deterministic because they say that a person’s genes are what cause criminal behaviour, not free will. However, this raises legal and moral issues: It is hard to hold someone morally responsible for criminal actions if they didn’t choose them, and so it may seem unfair to punish them.

Strengths of neural explanations of criminal behaviour:

• Supporting evidence: See studies above.

Weaknesses of neural explanations of criminal behaviour:

• Small sample sizes: Studies linking criminal behaviour with neural factors often use small sample sizes. One reason for this is that the population of serious criminals (e.g. murderers) are hard to gain scientific access to. This effect is compounded by the high costs of brain-scanning tools, such as fMRI, which makes it even more difficult to carry out this research on large numbers of people. Because of these small sample sizes, it may be invalid to generalise the findings from these studies to the wider population.
• Ethical issues: An implication of Raine’s research is that brain scans in childhood could be used to identify potentially violent criminals of the future. This policy, if enacted, could potentially reduce crime but is socially sensitive because it could lead to discrimination against people with these brain structures.
• Deterministic: Neural explanations are biologically deterministic because they say that neural factors are what cause criminal behaviour, not free will. However, this raises legal and moral issues: If someone doesn’t choose their criminal actions, it may seem unfair to hold them morally or legally responsible for them.

Strengths of Eysenck’s theory of criminal personality:

• Supporting evidence: Several studies support a link between the personality traits Eysenck describes above and criminality – in particular psychoticism. For example, Furnham (1984) tested 210 subjects using Eysenck’s personality theory, a social skills test, and a test of anomie (which basically means moral values). Of these 3 tests, the results of Eysenck’s personality theory were the most accurate predictors of criminal behaviour, which supports the criminal personality theory explanation of criminal behaviour.

Weaknesses of Eysenck’s theory of criminal personality:

• Relative importance of different factors: Although there is some evidence supporting a link between Eysenck’s 3 personality traits and criminal behaviour, the links are stronger for some personality traits more than others. For example, the correlation between being highly psychotic and criminal behaviour is much stronger than the correlation between being highly neurotic and criminal behaviour.
• Methodological concerns: Much of the research linking criminality with personality type determines personality type using self-report methods such as the Eysenck Personality Questionnaire. However, self-report techniques are often unreliable as the answer the participant gives varies depending on their mood. Self-report techniques are also often invalid as participants lie (e.g. to make themselves sound better).
• Other factors: There are many people who score highly for neuroticism, extroversion, and even psychoticism who do not engage in criminal behaviour. Conversely, there are many stable, introverted, and non-psychotic people who do engage in criminal behaviour. This suggests that other factors (e.g. biology or learning approaches) are needed for a complete explanation of criminal behaviour.

Strengths of level of moral reasoning as an explanation of criminal behaviour:

• Supporting evidence: Several studies suggest that criminal offenders have less developed moral reasoning than non-criminal controls. For example, Palmer and Hollin (1998) used questionnaires to compare the moral reasoning abilities of 126 male convicted offenders with 122 male non-offenders. The researchers found that the male offenders had significantly poorer moral reasoning compared to the male non-offenders, which supports the idea that underdeveloped moral reasoning contributes to criminal behaviour.
• Explanatory power: Differences in moral reasoning can explain why one person might commit a crime when another person may not. A person who commits a crime does not have the belief that what they are doing is wrong, so this means they are free to do so. In contrast, if someone has a belief that a criminal action is morally wrong, this is likely to prevent them from doing it. These differences in moral reasoning explain their different actions.

Weaknesses of level of moral reasoning as an explanation of criminal behaviour:

• Limited applications: Level of moral reasoning is able to explain some types of criminal behaviour better than others. For example, Thornton and Reid (1982) found criminals convicted of robbery were more likely to be at the pre-conventional morality stage than criminals convicted of assault.
• Methodological concerns: Kohlberg’s stages of moral development was based entirely on data from boys, with no girls included in his research. As such, the theory may suffer from gender bias (beta bias) and the results may not be valid when applied to female criminal behaviour.
• Other factors: There are examples of criminals of all stages of moral reasoning, which suggests that other factors besides pre-conventional moral reasoning (e.g. biological factors or learning approaches) are needed for a complete explanation of criminal behaviour.

Strengths of cognitive distortions as an explanation of criminal behaviour:

• Supporting evidence: Several studies support a link between criminal behaviour and cognitive distortions. For example:
  • Hostile attribution bias: Schönenberg and Jusyte (2013) showed pictures of ’emotionally ambiguous’ faces to 55 violent offenders and 55 matched control subjects. The researchers found that the violent offenders were more likely to interpret these ambiguous stimuli as aggressive than the controls were, which supports the explanation that hostile attribution bias may explain (violent) criminal behaviour.
  • Minimalisation: Pollock and Hashmall (1991) conducted a content analysis of the clinical records of 86 child molesters. They found the records often contained attempts to minimise the seriousness of their crimes Excuses including denying their actions were sexual (35%) and arguing that the actions were consensual (36%).
• Practical applications: Understanding the cognitive distortions that cause criminality could yield effective ways to reduce criminal behaviour and prevent recidivism. For example, cognitive behavioural therapy could reduce minimalisation by helping offenders understand how serious their actions are, or reduce hostile attribution bias by challenging the thought processes that lead offenders to misinterpret innocent actions as hostile.

Weaknesses of cognitive distortions as an explanation of criminal behaviour:

• Limited applications: There are many types of crime that cognitive distortions don’t explain well. For example, it’s easy to see how hostile attribution bias could lead to impulsive violence (e.g. the offender misinterprets an innocent look from someone and assaults them) but it’s hard to see how this bias could explain planned violence (e.g. a pre-meditated murder). This suggests that cognitive biases may only explain some types of crime and not others.
• Other factors: There are examples of criminals who don’t exhibit either hostile attribution bias or minimalisation, which suggests that other factors (e.g. biology or learning approaches) are needed for a complete explanation of criminal behaviour.

Strengths of differential association theory as an explanation of criminal behaviour:

• Supporting evidence: Alarid et al (2000) applied differential association theory to 1,153 newly-imprisoned criminals and found that it could accurately explain their criminal behaviour (e.g. which crimes they committed) in many cases. Because of this, the authors conclude that differential association theory is a good ‘general theory’ of crime and that it can be used to predict the likelihood that a person will engage in criminal behaviour.
• Explanatory power: Differential association theory is able to explain why rates of recidivism (reoffending after prison) are so high: People who are sent to prison will be surrounded by people who have positive attitudes towards crime and who will have knowledge of the methods involved in crime. These differential associations will further teach and encourage criminal behaviour, making a person sent to prison highly likely to reoffend.

Weaknesses of differential association theory as an explanation of criminal behaviour:

• Exceptions: There are many examples of people raised around criminals (i.e. people with favourable differential associations for crime) who do not go on to commit a crime. Similarly, there are examples of people raised in environments with very limited or no exposure to criminals who go on to commit crimes. These examples weaken support for the theory and suggest that other factors (e.g. biological factors) are also needed for a complete explanation of criminal behaviour.
• Limited applications: Although differential association may explain certain types of crime (e.g. burglary or drug dealing), there are many criminal acts that are individualistic in nature and so don’t fit with the theory. For example, serial killers are not typically raised in environments among other serial killers, so differential association theory is not able to explain such criminal behaviour.

Strengths of psychodynamic explanations of criminal behaviour:

• Supporting evidence: Bowlby (1944) demonstrates a correlation between maternal deprivation and criminal behaviour, which supports this particular psychodynamic explanation of criminal behaviour.

Weaknesses of psychodynamic explanations of criminal behaviour:

• Methodological concerns: Bowlby’s (1944) study took the form of interviews with the subjects. However, this method has been criticised for researcher bias as Bowlby’s own theories and pre-conceived ideas may have influenced the questions he asked and his interpretations of the answers. For example, Bowlby’s expectation that maternal deprivation could cause a child to become a psychopath could have biased him to diagnose children who were separated from their mothers as such. Because of this, Bowlby’s maternal deprivation hypothesis may not be valid.
• Conflicting evidence: According to Freud’s theory, women have weaker superegos than men (because a boy’s fear of being castrated is stronger than a girl’s fear of losing her mother’s love). If this were true, though, this would predict that women commit crimes than men because women would be more likely to have an underdeveloped superego. However, the opposite is true: Men are (on average) more likely to commit crimes than women, which suggests that psychodynamic explanations of criminal behaviour are inaccurate.
• Alternative explanations: The usual correlation does not guarantee causation argument can be made here.
  • Freudian explanations of criminal behaviour: Even if there is a correlation between, for example, having an absent same-sex parent and criminality, this does not necessarily prove that an underdeveloped superego is the cause, as Blackburn (1993) suggests.
  • Bowlby’s maternal deprivation hypothesis: Even if there is a correlation between maternal deprivation as an infant and criminal behaviour, this does not necessarily prove that maternal deprivation causes criminal behaviour. For example, it could be that genetics cause both the mother to abandon her child and criminal behaviour in the child. Alternatively, differential associations could explain the correlation between maternal deprivation and criminality if children raised in orphanages are more likely to be surrounded by criminal influences than children raised by their mothers.
• Unscientific: Science is about what can be observed and measured, but psychodynamic concepts are often impossible to test. For example, Freud’s theories claim that many of the psychological concepts that explain behaviour are below conscious level, which means that even the individual themself can’t observe them! The fact that these concepts can’t be measured or observed means the psychodynamic explanations of criminal behaviour based on these concepts may be unfalsifiable and unscientific.
• Other factors: There are many examples of people who experienced maternal deprivation or had an absent same-sex parent who do not go on to become criminals. This suggests that other considerations besides psychodynamic factors (e.g. biology) are needed to fully explain criminal behaviour.

Strengths of custodial sentencing:

• Positive outcomes upon release: Rehabilitation schemes in prison may result in the offender improving themself and changing their life for the better. For example, an offender who used crime to support themself financially may learn skills that enable them to make an honest living upon release. This results in a more positive contribution to society and gives the offender a second chance.
• Complements other approaches: Time in prison can be combined with other ways of dealing with criminal behaviour, such as behaviour modification or anger management, to improve the outcomes of custodial sentencing. For example, violent offenders can undergo anger management programs while in prison, which may accelerate their rehabilitation.

Weaknesses of custodial sentencing:

• Negative psychological effects: The many negative psychological effects of custodial sentencing (see above) can be said to go beyond retribution. For example, Fazel et al (2011) found that rates of suicide in prison were roughly 3 times higher than in the general population.
• Negative effects on criminal behaviour: The actual effects of custodial sentencing are often the complete opposite of the aims of custodial sentencing. For example, institutionalised prisoners may be more likely to commit crimes when released, which is the opposite of deterrence. Similarly, prison may reinforce criminal behaviour (e.g. differential associations may reinforce criminal values and teach better methods of crime), which is the opposite of rehabilitation. These criticisms are supported by the high rates of recidivism described above.

Strengths of behaviour modification:

• Supporting evidence: Some studies suggest that token economies can successfully change offenders’ behaviour. For example, Milan and McKee (1976) found that token economies were more effective at improving prisoner behaviour (defined and operationalised by measures such as getting out of bed on time, keeping a clean cell, partaking in educational activities, and completing tasks) compared to other methods, such as direct orders or praise for good behaviour.
• Low costs: Token economies are easy to implement and do not require specialist psychological training (unlike e.g. anger management).

Weaknesses of behaviour modification:

• Not effective long-term: In a study of youth offenders, Kirigin et al (1982) found that although token economies improved behaviour compared to controls during the program, this effect disappeared after the program ended. This suggests that token economies are not effective tools for long-term rehabilitation.
• Ethical concerns: Some argue that token economies are unethical for various reasons. Firstly, it may be seen as unethical and dehumanising to manipulate prisoners and restrict their ability to make their own choices via token economies. Secondly, negative reinforcement and taking away privileges (e.g. phone calls with loved ones) by taking away tokens may be seen as unfair or cruel.
• Alternative approaches to dealing with criminal behaviour: Other approaches to dealing with criminal behaviour may be better at achieving the aims of sentencing than token economies. For example, anger management is likely to be a more effective for rehabilitating violent behaviour, and restorative justice is likely to be more effective at achieving retribution.

Strengths of anger management:

• Supporting evidence: Several studies suggest that anger management is effective. For example, Ireland (2004) compared anger levels (as measured and operationalised by a self-report anger questionnaire and assessments by prison officers) of 50 prisoners who underwent anger management therapy with 37 prisoners in a control group who did not undergo anger management therapy. In the anger management group, mean anger ratings (as measured by the questionnaire) fell from 42.7 before therapy to 27.3 after therapy. In the control group mean anger ratings did not change significantly (39.7 to 36.6). This study suggests that anger management is an effective way to reduce anger.

Weaknesses of anger management:

• Conflicting evidence: Howells et al (2005) studied anger levels in Australian offenders before and after an anger management program. Although the offenders receiving anger management therapy demonstrated reduced anger levels after therapy compared to a control group who did not receive anger management, the difference was not statistically significant. This suggests that anger management programs do not significantly reduce anger levels and so are unlikely to prevent recidivism.
• Not effective long-term: Blackburn (1993) argues that, although anger management therapy demonstrates benefits in the short-term, there is little evidence to suggest anger management reduces recidivism in the long-term.
• Limited applications: Anger management will only be an effective way to deal with criminal behaviours caused by anger, such as assault. For criminal behaviours that are not caused by anger (e.g. burglary, drug dealing, etc.), anger management is likely to have little rehabilitative benefit.

Strengths of restorative justice:

• Supporting evidence: A meta-analysis by Latimer et al (2005) found restorative justice programs resulted in greater victim satisfaction, greater offender satisfaction, and lower rates of recidivism compared to traditional criminal justice responses such as custodial sentencing and probation.
• Lower costs: Restorative justice is typically much less expensive than alternative methods of dealing with criminal behaviour, such as custodial sentencing or anger management.

Weaknesses of restorative justice:

• Methodological concerns: Restorative justice programs are voluntary, so there will always be a self-selection bias for participants who are motivated to change their ways (as Latimer et al (2005) acknowledge). In other words, the kinds of offenders who volunteer for restorative justice programs may be less likely to reoffend anyway. This makes it difficult to say whether restorative justice programs actually cause lower rates of recidivism, or whether they are just correlated.
• Limited applications: The voluntary nature of restorative justice programs also means their applications are limited. For example, restorative justice can’t work if either the offender or the victim doesn’t consent to take part. Similarly, if the offender doesn’t take responsibility for their actions, restorative justice isn’t an option. This limits the potential applications of restorative justice.
• Soft option: There is a common perception that restorative justice is a soft option in that it doesn’t sufficiently punish offenders for their crimes. As such, there is often greater support for harsher approaches to dealing with crime, such as custodial sentencing.