The role of release planning in the reintegration experiences of high-risk offenders

Sophie R. Dickson & Devon L. L. Polaschek

Victoria University of Wellington

Author biographies:

Sophie Dickson is a PhD and clinical psychology student in the School of Psychology at Victoria University of Wellington, New Zealand. Her research interests include the rehabilitation and reintegration of high-risk offenders (with a particular focus on release planning), desistance, and risk assessment.

Devon Polaschek, PhD DipClinPsyc is a Professor of psychology at Victoria University of Wellington, New Zealand. Her research interests include theory, intervention, and intervention evaluation with serious violent and sexual offenders, psychopathy, desistance, reintegration, parole and experimental approaches to offender assessment.


The challenges of reintegration

Internationally, the literature on reintegration has demonstrated that high-risk offenders face a number of challenges when being released from prison. Stable accommodation is crucial to the successful reintegration of prisoners back into the community, yet it is difficult for many of them to obtain. Most prisoners return to live with their families after release, but these living arrangements are often only temporary, and are problematic for offenders from criminogenic families (Solomon, Visher, La Vigne, & Osborne, 2006). High-risk offenders also struggle to find employment; in a study of ex-prisoners’ adjustment to life in the community, it was reported that unemployment was the norm (Shinkfield & Graffam, 2009). Similarly, offenders have difficulty in finding positive social supports, in part due to the difficulty of maintaining positive attachments over long periods of imprisonment, and instead turn to antisocial peers for support (Lynch & Sabol, 2001). And finally, high-risk offenders tend to be released to socially disadvantaged neighbourhoods with high levels of crime and poverty (Hipp, Turner, & Jannetta, 2010). The cumulative effect of all of these barriers makes it difficult to resume an ordinary life; the more of these challenges an offender faces after release, the higher his risk of re-offending.

Release planning

One approach to aiding the transition from prison into the community is release planning. Release planning involves examining an offender’s plans for life after release, and helping him to improve weak plans by, for example, facilitating access to resources that could ease his transition back into the community. At the time this research was conducted, release planning was a component of the Special Treatment Unit Rehabilitation Programmes (STURPs) for both high-risk offenders and child sex offenders. Willis and Grace (2008, 2009) looked at the quality of release plans in two samples of child sex offenders who had attended Kia Marama or Te Piriti. Overall, they found that better quality release plans (i.e., plans that were more specific, confirmed, and more prosocial) were linked to lower rates of re-offending after release from prison. Dickson, Polaschek and Casey (2013) found a similar pattern of results with high-risk offenders from a STURP: better quality plans predicted a reduced likelihood of recidivism. Thus it appears that good quality release plans are related to a reduced rate of re-offending after release; however, so far no research has addressed the question of how release plans aid the reintegration process. What is the mechanism underlying the efficacy of release plans?

Research Question One

How do release plans work?

The current research explored two possible explanations for the efficacy of release planning. The first is the external pathway, which assumes that good quality release plans simply translate into good quality experiences in the community (e.g., good accommodation plans lead to somewhere reasonable to live on release). Theoretically, this idea is supported by the work of Sampson and Laub (1993), who tested out the assumption that crime is stable across an offender’s lifespan. They examined the lifespans of 1000 males and found that, generally, delinquent boys turned into antisocial men, and well-behaved boys turned into prosocial men. However, Sampson and Laub discovered a group who were delinquent as children but stopped offending and became prosocial adults. Upon closer inspection they found that these juvenile delinquents had experienced ‘turning points’ in their lives, which counteracted the continuity of their delinquency. These turning points were external events such as getting a stable job or marrying the ‘right’ woman. Sampson and Laub said that any psychological change was unnecessary and irrelevant: badly behaved people simply stumbled across good life events that turned things around for them. So our first hypothesis was that making good plans leads to offenders having more positive external experiences after release, interrupting their delinquency, and making them less likely to commit another crime.

The second explanation was the internal pathway (referring to internal, psychological processes). Giordano, Cernkovich, and Rudolph (2002) argued that Sampson and Laub provided an incomplete picture of the desistance process, because they ignored the work the offender does to move toward, and then sustain, a new way of life. They argued that offenders develop psychologically so that they turn towards environments that include positive external events (e.g., employment or marriage), rather than simply chancing upon external turning points. If they didn’t change psychologically, they would not recognise turning points, or would be unable to take advantage of them (e.g., when an offender gets a good job offer but then loses the job because of his drinking). Therefore it is important to focus on the role of factors internal to the offender. In order to operationalise the psychological changes that the desistance literature suggests, this study focused on three internal factors: Motivation to desist (the desire to give up crime), Self-Efficacy (the belief in being capable of giving up crime), and Prosocial Identity (seeing oneself as a prosocial individual). So the second hypothesis was that good quality release plans create higher levels of these internal factors, which in turn promote desistance from crime.


To test the above hypotheses, data from three samples of high-risk offenders were collected. The samples were comprised predominantly of Maori men (e.g., 55-66 percent M?ori, 22-29 percent NZ European, 6-19 percent Pasifika, and 2-3 percent other) with RoC*RoIs of 0.7 or higher who had been sentenced to at least two years in prison and who were therefore eligible for parole. The quality of offenders’ release plans was coded from file data, including psychological reports, and reports to the parole board, as well as other file information. See Box 1 for Release Plan Quality items.

Box 1.

Release Plan Quality Items:



Prosocial support

Plans to avoid antisocial associates

Risk level of the release environment

Plans to manage risks in the release environment

Total score

Box 2.

Parole Experience Quality Items:



Prosocial support

Amount of contact with antisocial associates

Number of risks in the release environment

Total score

Both the quality of offenders’ external experiences and the three internal factors were assessed at two months after release from prison. The quality of offenders’ external experiences was measured from their probation service notes. See Box 2 for Parole Experience Quality items. The three internal factors (i.e., Motivation, Self-Efficacy, and Identity) were assessed through interviews with the offenders. They were asked questions such as "How much do you want to go straight?", "How confident are you that you’ll be able to go straight?" and "How much do you see yourself as a criminal?" Finally, recidivism data were extracted for the first six months after release. The first study was retrospective and file-based, and focussed solely on Te Whare Manaakitanga participants. The second two studies were prospective, included both STURP participants and untreated high-risk offenders, and were part of the Victoria University Parole Project. The Parole Project included both offenders who had graduated from one of the four high-risk STURP rehabilitation programmes, and a comparison sample of similarly high-risk offenders serving sentences of a similar length but who had not gone through a STURP.

Summary of results

Study one

Study one explored the external pathway, asking: "Do release plans simply translate into better quality experiences on parole?" As expected, overall release plan quality was a significant predictor of overall parole experience quality. But unexpectedly there was only one direct relationship between an item in the release plan scale and the corresponding item in the parole experiences scale: better plans for employment led to better experiences of employment. All of the other relationships between individual items were indirect. For example, making better quality plans for avoiding risk (i.e., release environment and antisocial associate items) led to poorer quality experiences of accommodation or employment.

Parole experiences were found to explain a significant amount of the relationship between release plans and recidivism, indicating that good quality release plans help to reduce recidivism by improving experiences on parole. However, parole experiences did not fully explain the link between release plans and re-offending, indicating that the effect of release plans on re-offending is not due simply to their impact on the external experiences on parole that were measured in this study.

Study two

Study two explored the internal pathway with a slightly different sample, asking: "Do release plans influence factors internal to the offender?" The results demonstrated that good quality release plans led to increased levels of motivation to desist in the community, which in turn led to decreased rates of recidivism. Neither self-efficacy nor identity helped to explain the relationship between release plans and re-offending in this sample. Thus, the results indicated that one specific internal factor, motivation to desist, connects release plans to re-offending. However, motivation to desist did not fully explain this relationship, as there was still a significant relationship between release plans and reconviction when motivation to desist was taken into account.

Study three

Study three aimed first to explore the relative contributions of, and interplay between, the internal and external factors in the prediction of re-offending. Building on the results of studies one and two, it was hypothesised that the relationship between release planning and re-offending would be explained by both the internal and external experiences in the community and that these two domains would have a positive impact on one another, meaning that both motivation to desist in the community and parole experiences would positively predict each other. The results revealed that when the internal and external pathways were both included in a model (see Figure 1), release plan quality’s relationship to re-offending was explained by the external pathway, but not by the internal pathway. These results suggest that release planning helps offenders to have better quality experiences in the community, which reduce their risk of re-offending. When examining the interplay of the internal and external variables (see Figure 2), the results showed that better parole experiences led to increased levels of motivation to desist but motivation had no effect on parole experiences.


Overall, these results demonstrated that, as expected, good quality release plans predicted reduced rates of re-offending. The relationship between release plans and re-offending appears to be explained by the impact of release plans on experiences external to the offender (e.g., good quality plans for employment lead to good experiences of employment, which then decrease the likelihood of re-offending). These positive experiences then help the offender to feel more motivated to desist, but this motivation level does not on its own impact on recidivism.

The importance of the external experiences over the internal experiences is consistent with Maslow’s (1943) theory of human motivation. Maslow theorised that our needs fall into five different categories (in order of prepotency: physiological, safety, love, esteem, and self-actualisation needs) and that only when our most basic needs are satisfied, do our ‘higher’ needs emerge, or warrant attention. In Maslow’s terms, the external experiences of accommodation, social support, and employment meet needs that sit lower in the hierarchy than internal experiences of motivation to desist from crime (which may reflect a need for esteem or self-actualisation). During the initial few months of life in the community, these external experiences reflect needs that are likely still being met. Put simply, a homeless man will be thinking solely of finding food and shelter. Only after these needs are met will he contemplate satisfying his higher-order need of living a prosocial life1.

Box 3.
The relative contributions of the internal and external pathways
Figure 1.

The relative contributions of the internal and external pathways

Figure 1 shows the standardised regression coefficients for the theorised model in which release plan quality’s relationship to reimprisonment is mediated by both motivation to desist and parole experiences. The bigger the regression coefficient, the stronger the relationship. If a number has an asterisk next to it, it means the relationship is statistically significant. In this model release plan quality was a predictor of parole experiences. Parole experiences, in turn, were predictive of reimprisonment. The only significant indirect effect was the relationship between release plans and reimprisonment, mediated by parole experiences (coefficient of .041, p<.05).

Interplay between the internal and external pathways

Figure 2.

Interplay between the internal and external pathways

Figure 2 shows the same model as Figure 1 with the nonsignificant relationships removed and relationships between motivation and parole experiences added. The results showed that release plan quality predicted parole experiences, which in turn predicted reimprisonment. Parole experiences were also a positive predictor of motivation to desist, but motivation did not predict parole experiences. The indirect effect from release plan quality to reimprisonment (mediated by parole experiences) was significant with a standardised indirect coefficient of – .040, p<.05.

Research Question Two

Why do some offenders make better release plans than others?

The final major area of interest in this research was how offenders make good quality plans, with a focus on the contextual variables associated with better quality release plans. Deci and Ryan developed Self-Determination Theory (SDT) (Deci & Ryan, 2000) when they saw that, at their best, people can be inspired and exert a great deal of effort, yet at other times, people can also be passive and uninspired. This variation in behaviour represents variation in motivation. Deci and Ryan proposed that behaviours fall on a continuum from extrinsic through to intrinsic motivation. Research on SDT has demonstrated that the theory is able to predict persistent positive behaviour change across a variety of domains. When people are more intrinsically motivated to perform an action, they tend to perform the action better and persist with the action longer than if they are relatively more extrinsically motivated. They also tend to have more interest in the action, more confidence, more creativity, less stress, better self-esteem, and better general well-being (Ryan & Deci, 2000).

Deci and Ryan proposed that we have three basic psychological needs that need to be satisfied as we pursue a goal. The first is autonomy, which is the desire to be the source of our own actions. The second is competence, or one’s ability to interact effectively with our environment. The third is relatedness, or the feeling of being connected to significant others. These three factors facilitate intrinsic motivation and the internalization of extrinsic motivation. In other words, when our basic needs are met we begin to value and internalise goals that we may not have done before, and become more intrinsically motivated. The main hypothesis for this study was that offenders who experience higher levels of autonomy, competence, and relatedness during the planning process would become more intrinsically motivated to create release plans, and would then create better quality release plans.


The sample used for this study was the same as the final sample described in the above method section. We developed two measures to examine the variables of interest; one measure to assess the offenders’ levels of the three basic psychological needs, and another to measure the degree to which an offender feels intrinsically motivated to create a release plan. Offenders completed these measures prior to their release from prison. The quality of offenders’ release plans just prior to release was once again coded from file data, such as psychological reports, and reports to the parole board.

Summary of results

We examined how the three basic psychological needs contributed to the level of intrinsic motivation an offender reported to make release plans, and whether the level of intrinsic motivation then predicted release plan quality. The results demonstrated that autonomy was a significant predictor of the level of intrinsic motivation, which in turn predicted release plan quality: the more an offender felt he got to decide what his plans were, the more intrinsically motivated he felt to create good quality release plans, and the better his plans were.

Lastly, the research explored the role of STURP attendance in release planning. We expected that treated offenders would report greater levels of relatedness because they have more assistance in creating release plans, and that this relatedness would lead to them feeling more intrinsically motivated to create release plans. It was also hypothesised that untreated offenders would report higher levels of autonomy, compared to treated offenders because they created plans with less assistance (and therefore may have had more say over what their plans were). Treated offenders did report significantly higher levels of relatedness than untreated offenders, but there was no difference in their levels of autonomy. Treated offenders were also significantly more intrinsically motivated to create release plans. Finally, for the treated group, competence was the only significant predictor of the level of intrinsic motivation, whereas for the untreated group autonomy was the only significant predictor of the level of intrinsic motivation.


The results from this study were encouraging for a couple of reasons. First, because they indicated that offenders became more intrinsically motivated to make release plans (and in turn made better plans) when they had higher levels of autonomy. The results showed that offenders in the sample reported high levels of autonomy in general, meaning they had the necessary basic need met to make good release plans.

Second, because they suggest that offenders who participate in treatment in the STURPs perceive that they receive more support with their release planning than untreated offenders but this additional support does not make them feel that they are any less in control of their release plans. Importantly, treated offenders also had significantly better quality release plans than untreated offenders, suggesting that treatment is helpful in improving release plans.

It was interesting to note that treated offenders reported being motivated by feeling competent, whereas untreated offenders reported feeling motivated by autonomy. This result may reflect the pre-existing differences between treated and untreated offenders. Offenders who refuse treatment may be people who personally value autonomy highly and prefer to do things on their own, whereas offenders who accept treatment may be people who value competence and building personal skills. Alternatively, participating in treatment may lead to offenders feeling more competent at making plans and then valuing that competence highly. We cannot be sure which of these explanations is correct because we did not measure these variables prior to treatment.

Summary and implications

Offenders face a number of barriers to reintegrating into the community, such as unstable accommodation, unemployment, and little prosocial support. The current research suggests that offenders with better quality plans for life after release face fewer of these barriers and are then less likely to re-offend. These more positive experiences in the community then help offenders to feel more motivated to desist from crime. Their basic needs are being met so they can start to think about living a more prosocial life. It’s still early days for them though, so getting their basic needs met is more important than their level of motivation to desist in determining whether they re-offend or not. Finally, in order to help offenders to make good plans, it is important that they feel they get to decide what their plans are, and they feel capable of developing good plans. At this stage, we see that the STURPs are creating the kind of environment that helps offenders to make good quality plans.

A few implications arise from these results. The first is that men who come through the STURPs onto parole have better quality release plans than those who do not. Since the time this research was conducted, increased assistance has been provided for untreated high-risk offenders: a change this research suggests is worthwhile. Next, is that if offenders are educated about how good plans can help to keep them out of prison, it may encourage them to put more effort into creating good quality plans. Likewise, educating staff about the importance of promoting autonomy and competence in the release planning context could help men to make even better plans.

And finally, motivation to desist appeared to be unimportant in avoiding re-offending in this research. Does that mean that desistance theory is wrong? Probably not. Most offenders were motivated to desist at release, meaning that what makes the difference among those who are motivated is whether or not basic needs are met, and this is more likely if they have a good plan.


The authors would like to thank the Department of Corrections for funding the Parole Project and Victoria University of Wellington for the first author’s PhD Scholarship.


  • Deci, E. L., & Ryan, R. M. (2000). The "what" and "why" of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11, 227-268.
  • Dickson, S. R, Polaschek, D. L. L., & Casey, A. R. (2013). Can the quality of high-risk prisoners’ release plans predict recidivism following intensive rehabilitation? A comparison with risk assessment instruments. Psychology, Crime & Law, 19, 371-389.
  • Giordano, P. C., Cernkovich, S. A., & Rudolph, J. L. (2002). Gender, crime, and desistance: Toward a theory of cognitive transformation. American Journal of Sociology, 107, 990-1064.
  • Hipp, J. R., Turner, S., & Jannetta, J. (2010). Are sex offenders moving into social disorganization? Analyzing the residential mobility of California parolees. Journal of Research in Crime and Delinquency, 47, 558-590.
  • Lynch, J. P., & Sabol, W. J. (2001). Prisoner reentry in perspective. Crime Policy Report, Urban Institute: Justice Policy Centre.
  • Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50, 370-396.
  • Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68-78.
  • Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways and turning points through life. Cambridge, MA, Harvard University Press.
  • Shinkfield, A. J., & Graffam, J. (2009). Community reintegration of ex-prisoners: Type and degree of change in variables influencing successful reintegration. International Journal of Offender Therapy and Comparative Criminology, 53, 29-42.
  • Solomon, A. L., Visher, C., La Vigne, N. G., & Osborne, J. (2006). Understanding the challenges of prisoner reentry: Research findings from the Urban Institute’s prisoner reentry portfolio. Washington DC, Urban Institute Justice Policy Centre.
  • Willis, G. M., & Grace, R. C. (2008). The quality of community reintegration planning for child molesters: Effects on sexual recidivism. Sexual Abuse: Journal of Research and Treatment, 20, 218-240.
  • Willis, G. M., & Grace, R. C. (2009). Assessment of community reintegration planning for sex offenders: Poor planning predicts recidivism. Criminal Justice and Behavior, 36, 494-512.

1.  However, it should also be noted that the measures of internal experiences in this study were limited to a single item for each type of experience. Future research using better measures of internal experiences should be conducted before we reach the conclusion that internal experiences are not relevant to success on parole.