Howells et al. (2005) Offender Treatment Options (Edexcel A-Level Psychology): Revision Notes
Howells et al. (2005) Offender Treatment Options
Background
Kevin Howells and colleagues investigated prison and community offender populations in Australia, examining real offenders engaging in actual treatment programmes. The research was conducted during a period when anger management interventions for offenders were increasing, reflecting a shift towards a rehabilitative approach rather than a traditional punishment-focused perspective. The study addresses a key psychological research interest: whether interventions produce genuine behavioural change in offenders.
This study is particularly significant because it examines real-world treatment programmes rather than laboratory-based interventions, providing valuable insights into the practical effectiveness of anger management approaches with actual offender populations.
Aim
The study had two primary aims:
-
To determine whether anger management is more effective than no treatment in producing change in offenders.
-
To investigate whether improvement in treatment can be predicted from pre-treatment offender characteristics, particularly whether offenders are ready to engage in treatment.
The second aim introduces the concept of treatment readiness – a critical factor in determining whether offenders will benefit from intervention programmes. This recognizes that motivation and preparedness significantly influence treatment outcomes.
Participants
418 male offenders participated in the study. All had been referred to prison or community probation equivalent anger management programmes in Australia. Key participant characteristics:
- 86% were from prison-based anger management programmes
- Diverse ethnic backgrounds and offence types (mainly violent offences)
- 73% had not previously completed an anger management programme
- Control sample selected from participants on a waiting list for the programme
The use of a waiting list control group is methodologically important – it provides a comparison group without denying treatment to those who need it, as control participants would eventually receive the intervention.
Procedure
Programme structure
Participants attended programmes lasting approximately 20 hours (over 10 sessions). Programmes in different geographical areas were similar in content and based on a cognitive behavioural approach. The treatment manual was developed in New Zealand and derived from Novaco's framework (1997).
Programme content
Structured exercises focused on:
- Identifying provocations
- Relaxation techniques
- Cognitive restructuring
- Assertion skills
- Relapse prevention
These five components represent the core elements of evidence-based anger management programmes. Each component targets a different aspect of anger control, from recognizing triggers to preventing future incidents.
Assessment measures
Participants completed outcome measures using two internationally recognised questionnaires about anger experiences and triggers. These were administered:
- Before attending the programme
- After completing the programme
- At two-month follow-up
- At six-month follow-up
Support was provided for participants with literacy issues, with questionnaires verbally administered. Staff (either prison or community-based) completed an observational rating scale for each participant.
The multiple time points for assessment are crucial for understanding not just immediate effects, but whether changes persist over time. This addresses a key question in treatment effectiveness: do benefits last?
Completion rates
- 285 participants completed the post-intervention assessment
- 78 participants (93% of prison-based attendees) completed the two-month follow-up
- 21 participants completed the six-month follow-up assessment
Notice the significant drop in participant numbers over time – this sample attrition becomes a key limitation of the study, particularly for drawing conclusions about long-term effectiveness.
Findings
Anger knowledge improvement
Offenders who completed the anger management programme showed greater improvement in anger knowledge than those in the control group:
- Treatment group mean change = 1.80
- Control group mean change = 0.95
There was some improvement among control group members who did not complete any treatment, though substantially less than the treatment group.
The control group improvement raises an interesting point: simply completing questionnaires about anger may have a small therapeutic effect in itself. This is known as a testing effect and suggests that self-reflection on anger issues can prompt some change even without formal intervention.
Anger management skills
Although the treatment group showed relative improvement in ability to manage anger using appropriate skills, this difference was not statistically significant when compared to the control group.
The lack of statistical significance here doesn't necessarily mean the treatment had no effect – it may indicate that measuring actual skill application is more difficult than measuring knowledge, or that skills take longer to develop than knowledge.
Follow-up results
At the two-month follow-up, the initial improvement was observed again and was statistically significant. This demonstrated that the treatment was helping to manage anger and participants were continuing to make progress after the course finished. However, the same effect was not observed at the six-month stage.
Critical Finding: The loss of significant effects at six months raises important questions about the durability of treatment gains. This suggests that:
- Offenders may need booster sessions or ongoing support
- The benefits of short-term interventions may fade without reinforcement
- Follow-up programmes might be necessary for sustained change
Predictive validity
The ability to predict how much change someone will make in treatment (known as predictive validity) was more accurate among those who had the most negative anger symptoms before attending the programme. A positive correlation was found between treatment readiness and progress made after course completion.
Conclusion
The overall impact of the anger management programmes was modest. Completing an anger questionnaire, even without treatment, may provide a small benefit in itself. The improvement shown by those completing programmes is not always maintained several months after the course has finished.
Those who had the most difficulties in anger management made the greatest improvements when completing a programme. Motivated offenders can be predicted to make more positive change than those who are not ready to address their anger problems.
Key Implication: This finding about motivation and readiness suggests that screening offenders for treatment readiness before programme entry might improve overall effectiveness. Resources could be better allocated to those most likely to benefit, while those not ready could receive preparatory interventions first.
Evaluation
Strengths
High ecological validity: The data was obtained within a naturally occurring treatment programme using real offenders. Consequently, the findings have real-life clinical application, making the research more applicable to other treatment programmes of similar content.
Ecological Validity Explained
Unlike laboratory studies that might use artificial scenarios or non-offender populations, this research examined actual treatment programmes in real prison and community settings. This means the findings are directly relevant to practitioners working with offender populations.
Randomised design: The study was a randomised trial, with offenders allocated to either receive treatment or be placed on a waiting list for treatment. This provides a control group against which comparisons can be made, minimising allocation bias by the researcher. This makes the findings more reliable as it is possible to identify the extent to which the treatment influenced behaviour, such as whether the treatment programme actually worked.
Evidence-based approach: The programmes measured used an evidence-based model and are applicable to many anger management programmes undertaken with offenders, as they all follow similar content. This makes the research generalisable to other offenders undertaking treatment of this type.
Control of extraneous variables: The study minimises some potential confounding factors through its randomised allocation and standardised programme content across different locations.
Weaknesses
Unaccounted extraneous variables: Not all variables were accounted for within the research, including individual differences in levels of understanding of the programmes that may account for the amount of change some participants made. Therefore, it is possible that such extraneous variables may influence the outcome of the research.
Confounding Variables Concern
Factors such as:
- Intelligence and cognitive ability
- Previous therapy experience
- Support networks outside the programme
- Severity of anger problems at baseline
- Motivation for change
These were not fully controlled for and could have influenced treatment outcomes independently of the programme itself.
Sample attrition: As the sample size reduced substantially by the six-month stage, it is difficult to determine with certainty whether change is or is not maintained at this time. Further research will be required before it is possible to determine the longer-term effectiveness of anger management programmes.
The dramatic reduction from 418 initial participants to only 21 at the six-month follow-up (approximately 5% retention) severely limits the conclusions that can be drawn about long-term treatment effects. Those who remained may represent a biased sample of particularly motivated or successful participants.
Self-report bias: Most of the questionnaires were self-report and the offenders may not have been honest in their responses, possibly due to concerns about the impact of being honest among their peers. This introduces an element of response bias, which reduces the reliability of the findings.
Social Desirability in Offender Populations
Offenders might be particularly prone to providing socially desirable responses rather than truthful ones, especially if they believe:
- Their responses might affect parole decisions
- Prison staff might view their answers
- Other inmates might judge them negatively
- Appearing improved could benefit their situation
Observer limitations: Staff could only rate behaviours they observed and they may not have seen all behaviours, making their ratings limited in scope. This further affects the reliability of the behavioural measures obtained.
Generalisability limitations: The study focused exclusively on male offenders in Australia, which may limit the generalisability of findings to female offenders or offenders in different cultural contexts.
Key terms
Predictive validity: The extent to which results from a test or study can predict future behaviour. In this study, treatment readiness scores were used to predict how much improvement offenders would show after completing the programme.
Cognitive behavioural approach: A therapeutic approach that addresses dysfunctional emotions and behaviours through structured, goal-oriented procedures focusing on changing thinking patterns. This approach assumes that changing how people think about situations will change how they feel and behave.
Rehabilitative approach: An approach to managing offenders that focuses on changing behaviour and reducing reoffending, rather than purely punishing. This contrasts with retributive justice, which emphasizes punishment for past actions.
Response bias: The tendency of participants to respond inaccurately or falsely to questions, which can affect the validity of research findings. This can occur for many reasons, including social desirability, fear of consequences, or misunderstanding questions.
Remember!
Key Points to Remember:
-
Howells et al. (2005) found that anger management programmes produced modest improvements in offenders' anger knowledge, with treatment group showing mean change of 1.80 compared to 0.95 in the control group.
-
The study demonstrated predictive validity – motivated offenders who showed treatment readiness before the programme made the most positive changes afterwards.
-
Improvements were maintained at the two-month follow-up but not at six months, suggesting the need for continued support or refresher interventions.
-
The research has high ecological validity because it examined real offenders in naturally occurring treatment programmes, making findings applicable to actual clinical practice.
-
Key weaknesses include sample attrition (only 21 participants completed six-month follow-up), self-report bias in questionnaires, and unaccounted extraneous variables such as individual differences in programme understanding.
-
The finding that treatment readiness predicts success has important practical implications – screening for motivation and readiness could improve programme effectiveness by ensuring participants are prepared to engage with treatment.