ArticlePDF Available

The effect of digital technology on prisoner behavior and reoffending: a natural stepped-wedge design

Authors:

Abstract and Figures

Objectives Although prisons aspire to rehabilitate offenders, they fail to prepare prisoners for release into our modern digitally sophisticated society. The objectives of the current study were to assess the impact of digital technology on the culture of prisons, and on prisoners’ ability to self-manage their behavior and reoffending. Method Using a natural stepped-wedge design, 13 prisons in the UK were examined that had installed self-service technology over a period of 7 years. A longitudinal multi-level model was used to analyze frequencies of disciplinary proceedings within and between the prisons before and after installation. Reoffending was examined in comparison with a control sample. Quantitative results were supported by a prisoner survey and usage data. Results Prison disciplinary offenses were significantly reduced over a two-year period, and reoffending in the first year after release was reduced by 5.36% compared to a 0.78% reduction in comparison prisons. The prisoner survey and usage data suggested that prisoners felt much more in control of their lives in prison and much more confident in coping with technology in the outside world. Conclusions The changes created by the introduction of digital technology offer the opportunity to make prisons more efficient for staff, and places of improved learning and rehabilitation for prisoners, contributing to a safer society. This study offers an important contribution to the field of corrections, providing the first quantitative assessment of the effect of prisoner self-service technology on prisoner behavior and reoffending.
This content is subject to copyright. Terms and conditions apply.
The effect of digital technology on prisoner behavior
and reoffending: a natural stepped-wedge design
Cynthia McDougall
1
&Dominic A. S. Pearson
2
&
David J. Torgerson
1
&Maria Garcia-Reyes
1
#Springer Science+Business Media B.V. 2017
Abstract
Objectives Although prisons aspire to rehabilitate offenders, they fail to prepare pris-
oners for release into our modern digitally sophisticated society. The objectives of the
current study were to assess the impact of digital technology on the culture of prisons,
and on prisonersability to self-manage their behavior and reoffending.
Method Using a natural stepped-wedge design, 13 prisons in the UK were examined
that had installed self-service technology over a period of 7 years. A longitudinal multi-
level model was used to analyze frequencies of disciplinary proceedings within and
between the prisons before and after installation. Reoffending was examined in com-
parison with a control sample. Quantitative results were supported by a prisoner survey
and usage data.
Results Prison disciplinary offenses were significantly reduced over a two-year period,
and reoffending in the first year after release was reduced by 5.36% compared to a
0.78% reduction in comparison prisons. The prisoner survey and usage data suggested
that prisoners felt much more in control of their lives in prison and much more
confident in coping with technology in the outside world.
J Exp Criminol
DOI 10.1007/s11292-017-9303-5
*Cynthia McDougall
cynthia.mcdougall@york.ac.uk
Dominic A. S. Pearson
dominic.pearson@port.ac.uk
David J. Torgerson
david.torgerson@york.ac.uk
Maria Garcia-Reyes
mariaelena.garciareyes@york.ac.uk
1
University of York, Heslington, York YO10 5DD, UK
2
University of Portsmouth, King Henry Building, King Henry I Street, Portsmouth PO1 2DY, UK
Conclusions The changes created by the introduction of digital technology offer the
opportunity to make prisons more efficient for staff, and places of improved learning
and rehabilitation for prisoners, contributing to a safer society. This study offers an
important contribution to the field of corrections, providing the first quantitative
assessment of the effect of prisoner self-service technology on prisoner behavior and
reoffending.
Keywords Digital exclusion .Offender rehabilitation .Prison misconduct .Prisoner
reentry.Recidivism .Digital technology.Stepped-wedge design
Introduction
We have learned much in recent decades about how to rehabilitate offenders. We have
discovered that offending behavior programs targeting Brisk^and Bneed^with cognitive
behavioral interventions can significantly reduce reoffending of prisoners and offenders
in the community (Andrews et al. 1990;BontaandAndrews2017;Lipseyetal.2007;
Lipsey and Wilson 1998). These principles have subsequently been incorporated into
structured community supervision with successful outcomes in reducing reoffending in
comparison with traditional supervision methods (Bonta et al. 2008; Lowenkamp et al.
2014; Pearson et al. 2011;Robinsonetal.2012; Smith et al. 2012; Taxman 2008). More
recently, attempts have been made to enhance these approaches with methods to
encourage desistance from crime through building on the positive characteristics of
offenders (McNeill and Weaver 2010; Ward 2010). However, so far, these approaches
have not demonstrated an improvement on the reduced levels of reoffending already
achieved (Andrews et al. 2011; Netto et al. 2014),particularly among those serving short
prison sentences (Ministry of Justice 2011).
There are, however, opportunities to improve rehabilitation in prisons by using
digital teaching techniques to expand our capacity to deliver evidence-based interven-
tions. Our school children are experts in digital technology, while our prisoners
Bconstitute one of the most impoverished groups in the digital age^(Jewkes and
Reisdorf 2016, p. 1). Their digital exclusion is exacerbated by refusing them the
opportunity to acquire and use the basic skills they need to function successfully in
modern society. Although not all prisoners lack digital skills, it is argued that there is a
growing Bdigital divide^in society, with many prisoners finding themselves at the
extreme end of this divide (Champion and Edgar 2013, p. iii). In this paper, we evaluate
the use of digital technology in prisons to assess whether access to this technology can
improve prisoner behavior inside and outside of prison. We argue that the technology
can support offender rehabilitation by encouraging self-responsibility and preparedness
for release, thereby reducing reoffending.
Reforming prisons to assist offender rehabilitation
Recently, governments with high rates of imprisonment have recognized that prisons
are an expensive way not to rehabilitate people, and that more use should be made
while prisoners are in our care to rehabilitate persistent offenders. In a number of US
C. McDougall et al.
jurisdictions, it has become clear that the political climate is moving away from
punishment-based policies. The US state of California, for example, has taken decisive
action in reducing its prison population by one-third, with radical plans for offenders to
be managed in their local communities (Petersilia 2016). Similarly, in England and
Wales, ministers have spoken emphatically about the need for change so that prisons
may become places of learning and rehabilitation, rather than outdated human ware-
houses. Prison governors will be given Bunprecedented freedom^to introduce reforms
and operational autonomy to introduce opportunities to promote rehabilitation, and the
Prime Minister announced that six reform prisons would be created to pilot this
increased autonomy (Cabinet Office 2016).
Prisoner rehabilitation and reentry to the community plays a large part in plans to
make communities safer in both the US and the UK, with attention being given to
encouraging activities that will better prepare offenders for reentry and improve links
between prison and the community before release (Cabinet Office 2016;Officeofthe
Attorney General 2016). However, those who have been detained for many years will
not recognize the modern world they will be released into, having a complete lack of
familiarity with the electronic devices they will need to understand in order to function.
Although those more recently imprisoned will be more familiar with digital commu-
nication, we are, nevertheless, aggravating their disadvantage by depriving them of
technology as a means of preparing for their release. According to Her MajestysChief
Inspector of Prisons in 2013, Bprisons [are] in a pre-internet dark age: inefficient,
wasteful and leaving prisoners woefully prepared for the real world they will face on
release. I have not met one prison professional who does not think drastic change is
needed^(Champion and Edgar 2013,p.iii).
The introduction of digital technology to prisons can assist rehabilitation in two
main ways discussed below: by changing the prison culture to be more supportive of
rehabilitation and by providing prisoners with the skills and means to prepare effec-
tively for their release.
Changing the prison culture
To really achieve effective rehabilitation, efforts made must be supported by the prison
culture. Officers are known to suffer stress if there is a clash between the culture and
how they are being asked to perform their roles (Lambert et al. 2011; Viglione et al.
2017). Incongruence may also impact on prisoners undertaking rehabilitation programs
in a classroom, but then returning to a hostile and inhospitable prison environment at
the end of the session (Mann et al. 2013). Mann and colleagues found that prisoners
thought attending treatment would harm their social status in the prison.
A growing body of research supports the view that prisons, rather than being an
effective deterrent to crime, are often in themselves criminogenic (Agnew 2006;Cid
2009; Listwan et al. 2013; Mazerolle and Piquero 1998; Nagin et al. 2009; Sherman
1993). Listwan et al.sresearchtestsAgnews(2006) general strain theory (GST),
which proposes that events and conditions that are physically or psychologically
distressing to individuals increase the likelihood of criminal behavior. Events
included in categories of strain are similar to stressors prevalent in most prisons, such
as perceived unfairness and lack of control. These are said to lead to negative emotional
states, such as anger and frustration, which can lead to violent and criminal behavior.
The effect of digital technology on prisoner behavior and...
Listwan et al. (2013) propose that, not only do the pains of imprisonment promote a
negative response to current events in prison, but also that certain types of strain appear
to increase the likelihood of reimprisonment. The kinds of strain predicted to have an
association with reoffending were: Bnegative impact by other inmates, particularly
verbal and physical abuse; negative treatment by staff; negative prison environment,
notably the perceived level of violence in the prison; and anticipated strains after
release, especially economic and housing strains^(Listwan et al. 2013, pp. 150151).
Their results showed that a negative prison environment and negative relations with
other inmates were associated with a higher probability of reimprisonment.
Surprisingly, negative relationships with correctional officers did not appear to affect
reoffending, although impact within the prison was not considered, as Listwan et al.s
study was based on community adjustment data and self-report from ex-prisoners. The
GST might suggest, however, that both a negative prison environment and negative
treatment by staff would impact on misconducts.
Prisons are also inefficient, as attested to by HM Inspector of Prisons (see above),
which can also lead to frustration. In most UK prisons today, prisoners, if they have a
request, have to speak to a prison officer or write the request on a piece of paper. This
piece of paper is then transferred to the relevant department, dealt with by hand, and
returned to the prisoner with an answer (C. McDougall and D.A.S. Pearson, 2014,
Process evaluation: The prisoner Custodial Management System (CMS), Unpublished).
As one prisoner stated to the authors, BWere treated like children. Everything is done
for us. Come onwere adults.^These paper-based transactions could take days, and
depend on the quantity of paperwork the officer is required to handle that day. In the
case of buying items from the prison shop, this could amount to more than a thousand
pieces of paper being handled by officers during one week on this one function. This
process is inevitably subject to delay and human error, resulting in much frustration.
Prisoners and officers both recognize such incidents as a major source of dispute
between them; often, prisoners equate the speed of response with the degree to which
they are treated with respect (Hulley et al. 2012). This can become a major source of
tension and is far removed from preparing the prisoners to take responsibility for their
lives on release. The impact of these processes are in line with the main strains of
perceived lack of fairness and lack of control, identified in the GST (Agnew 2006).
Disputes over paper-based systems can be a major contributor to adjudications, the main
disciplinary procedure in prisons for misconducts. Adjudications have been recognized as a
major predictor of reoffending (Cochran et al. 2014; French and Gendreau 2006;Heiletal.
2009), so the ability to reduce adjudications by removing its causes could be a valuable
contributor to rehabilitation. Giving prisoners responsibility for managing basic tasks for
themselves would decrease the blame culture that exists, and allow prisoners to feel more in
control of their lives, so removing one of the sources of perceived unfairness.
There is increasing interest in the US, the UK, Europe, and in Australia in normal-
izing the prison environment by bringing prisons more in line with modern society
through digital technology and use of tablet computers in prisons (Thompson 2014;
Tran 2014). Conferences are also being organized to extend knowledge of what might
be achieved through introducing digital self-service devices so that prisoners have more
control over their lives (International Corrections and Prisons Association, ICPA 2017).
In the UK, some prisons have installed Bself-service kiosks^on prison wings,
similar to those found in superstores, where prisoners can independently take
C. McDougall et al.
responsibility for many of their own requirements. The prisoner can, for example, order
toiletries and small items from the prison shop by simply choosing from a list on a
kiosk; subsequently, the cost of purchases are automatically calculated, deducted from
the prisoners private account, and a receipt is provided and the products are delivered.
Other functions are described below (see Theory of change for prisoners). These
functions are similar to the kinds of digital self-service in the community, and offer
the opportunity to practice the skills required to prepare for release. Prisoners can,
therefore, gain experience of using the technology in the security of the prison
environment, while at the same time increasing their autonomy and personal skills.
The benefits for officers, as well as prisoners, are also apparent. These include relieving
officers from mundane administrative tasks to allow them to become involved in rehabil-
itation activities and maintaining safe and secure prisons (C. McDougall and D.A.S.
Pearson, 2014, Process evaluation: The prisoner Custodial Management System (CMS),
Unpublished).
The use of prisoner self-service (PSS) technology is not widespread across the UK Prison
Service. Although intuitively the benefits appear evident, we have heard strongly held views
that handing over such control to prisoners will have a deleterious effect on the relationships
between officers and prisoners. This reluctance to yield control may be an artifact of Bamore
militaristic approach^to dealing with prisoners, as observed, for example, in a comparison
of Dutch and English prisons (Dirkzwager and Kruttschnitt 2012, p. 409). The alternative
view from prisoners and officers is that the use of kiosks actually improves the relationship
between them. Researchers and policy-makers should, therefore, not be deterred from
embracing new technology because of fears of changing officer/prisoner relationships, as
this could provide an opportunity for genuine culture change in prisons.
Encouraging prisoner self-directed rehabilitation
Wolff e t a l. ( 2012) examined readiness for release in prisons and concluded that those
who had been in prison the longest were the least prepared for release, although this
conclusion may be open to dispute (Tiedt and Sabol 2015). Wolff et al. recognized that
prisoners require education, financial assistance, job training, employment assistance,
and community living skills, together with the self-management skills associated with
their offense-related problems. Whilst acknowledging the importance of these training
needs, few studies so far have identified the importance of acquiring skills in the use of
digital technology to achieve these goals, nor in attempting to disconnect the relation-
ship between digital and social exclusion (Helsper and Eynon 2013).
The disadvantage to prisoners created by the lack of digital facilities and, hence,
limited acquisition of skills, has been recognized (Jewkes and Reisdorf 2016;Knight
2015). However, there has, so far, been no published empirical evaluation of the impact
on prisoners of teaching digital skills and using them to facilitate prisoner rehabilitation.
A number of prison studies have examined whether information technology has been
an aid to education but, often, the research designs have been weak. One rigorous
experimental study evaluated the BEfficacy of a computer-assisted instruction program
in a prison setting^(Batchelder and Rachal 2000). The authors found no statistically
significant additional impact from computer assistance over that of traditional tuition.
This may be because many prisoners have had discouraging early experiences of
classroom-based tuition, which dissuades them from taking part in adult education.
The effect of digital technology on prisoner behavior and...
However, if the use of the computer is in the prisoners interests, such as completing a
daily activity on their own behalf, or organizing a visit with their family, then there may
be greater motivation to learn the skills.
This proposition appears to be supported in research by King et al. (2017), who
tested, in a randomized controlled trial, the difference between prisoners completing
psychological assessments on tablet computers and those using paper and pencil.
Although there was no difference in content between the two methods of response,
King et al. found that prisoners had a preference for using the tablet computer and,
subsequently, adopted a more constructive attitude to the correctional institution after
using the technology, compared to the attitude of those using the paper and pencil
method. If a similar response was found using technology for managing onesown
basic requirements in prison, this would be a valuable indicator as to how we may
motivate prisoners to learn new skills related to rehabilitation. A wider participant pool
for rehabilitation might, therefore, be accessible through the use of digital technology.
Many of the skills identified by Wolff et al. (2012), and Andrews and Bonta (1995)
in the Level of Service InventoryRevised, as being important criminogenic needs to
be addressed in rehabilitation could be assisted by the use of technology in prison. This
could expand the range of prisoners who can be encouraged to take advantage of the
riskneedresponsivity (RNR) principles. In the current study, we do not expect digital
technology in itself to be the active agent of change in addressing criminogenic needs,
but could be the instrument that aids the translation of a well-designed intervention
from that of a demonstration project to being part of regular practice reaching large
numbers of offenders (Bourgon et al. 2009). This approach might also give more
flexibility to take account of participant Bresponsivity^, which has sometimes received
less attention than other elements in the RNR model (Goggin and Gendreau 2006). This
has been highlighted as a potential problem in large-scale roll-out studies which depend
on a fixed program being delivered with little scope for adapting to individual learning
needs/styles (Polaschek 2012).
Theory of change for prisoners
Our proposition can be illustrated in a Btheory of change for prisoners^following
implementation of PSS technology (Fig. 1). PSS has the advantage, once installed, of
having to be used by all prisoners in order to access basic routine requirements, such as
ordering food, buying items from the prison shop, checking financial account balance,
and booking visits. There are additional functions, such as applying for education, a
change of employment, healthcare appointments, and to join a rehabilitation program.
These PSS functions may facilitate plans for release and allow the prisoner to show
interest in rehabilitation. Although the theoretical model has a standard format, it offers
flexibility to adapt to the individual, depending on the assessed type and level of risk
and need. King et al. (2017) have demonstrated that psychometric assessments, such as
the Risk Need Perception Survey developed to appraise self- and evaluator-perceived
criminogenic needs (King 2016), based on Bonta and Andrews(2017) RNR model,
can be successfully administered by the use of tablet computers. Though these were not
available in the current study, variations on the technology provided may be considered
in the future.
C. McDougall et al.
In Fig. 1, the first stage shows implementation of the kiosks and in the second stage,
the prisoner learns the skills and begins to use the system. The third stage presents a
selection of tasks that prisoners will be able perform for themselves:
Direct contact with the offender supervisor Direct contact with the offender supervi-
sor (OS) will allow the prisoner to initiate written contact in order to make appointments,
discuss rehabilitation plans, offense-related programs, employment prospects, accommo-
dation, or family issues. This allows a focus on purposeful activity. Mann et al. (2013)
have noted that many offenders are deterred from seeking help when they have to make
the request through a prison officer; hence, direct contact may help to overcome this.
Access to programs/education/support Access to programs will facilitate changes in
criminogenic needs, as identified by Bonta and Andrews (2017).
Self-responsibility/reduced dependency on officers The value of being in control of
basic activities is supported in the GST (Agnew 2006) as alleviating some of the strains
on prisoners, which can lead to criminal activity. The theoretical model shows a
pathway towards reducing adjudications.
Easier contact with family and friends This has been emphasized by Lösel et al.
(2012) as being extremely important to prisoners and their families in a longitudinal
study of imprisoned fathers and their families. Positive resettlement, which included
desistance from crime, was associated with a high quality of family relationships, good
communication between father and family during his imprisonment, and high frequen-
cy of contact in prison.
Fig. 1 Theory of change for prisoners
The effect of digital technology on prisoner behavior and...
Improved attitudes and prospects for accommodation and employment
Stage 4 shows the potential positive impact on prisoner thinking, feeling, and behavior,
following the opportunities presented in stage 3. Consistent with the GST (Agnew
2006), negative attitudes dissipate due to reduced feelings of injustice and ill-treatment.
This would aid prisoners in making contact with an external supervisor, finding
employment on release, finding accommodation, and improving relationships with
family, thereby reducing strains and improving prisoner well-being and mental health
(Listwan et al. 2013).
Final outcome measure
Stages 5 and 6 are the outcome measures linked to reducing dynamic risk. At the final
stage, 7, it is anticipated that the changed behavior in the prison and reductions in risk
of some offenders may impact on the actual rate of reoffending following release
(French and Gendreau 2006).
It is proposed, therefore, that the implementation of the technology, interacting
with the opportunities it presents for more purposeful activities, is likely to impact
on prisonersattitudes and behavior. The improvement in prison behavior will
impact on adjudications, which is a positive predictor of reduced reoffending
(Cochran et al. 2014; French and Gendreau 2006;Heiletal.2009). At the same
time, there is likely to be a positive impact on rehabilitation, with improved family
relationships (Lösel et al. 2012) and an increase in successful offending behavior
program completions (Bonta and Andrews 2017). The model allows for individual
tailoring depending on the individual prisoners risk, criminogenic needs, and
treatment responsivity. It is anticipated that officers will be relieved of many of
the mundane daily paper-based tasks involved in meeting the basic requirements
of prisoners, and, hence, will have more time to assist in encouraging individual
prisoners in developing their problem-solving skills (Champion and Edgar 2013;
C. McDougall and D.A.S. Pearson, 2014, Process evaluation: The prisoner
Custodial Management System (CMS), Unpublished).
The current study
Only a small number of mainly private prisons have, so far, installed PSS technology.
The current study was set up to rigorously examine the psychological, social, and
rehabilitative impact of PSS in those prisons that had adopted it.
The main study was preceded by a process evaluation in a prison that was about to
install PSS and this contributed to the theory of change described above (C. McDougall
and D.A.S. Pearson, 2014, Process evaluation: The prisoner Custodial Management
System (CMS), Unpublished). The process evaluation facilitated the identification of
suitable measures to empirically test the impact of PSS. For statistical reasons, we
selected three representative performance measures to examine impact on the main
prison priorities of providing Bsafe, secure, and decent prisons^,Bprisoner
rehabilitation^,andaBreduction in reoffending^. The measures would be supported
by qualitative information and usage data.
C. McDougall et al.
In discussion groups, we discovered that both officers and prisoners thought
that the outdated paper-based systems actually contributed to hostility between
them. Some prisoners thought that officers did not take their applications
seriously, and officers were aware that they were often accused of being
responsible for delays in the system, or even deliberately failing to process
an application. In addition, PSS encouraged prisoners to take responsibility for
their own activities in prison. Prison adjudications are the main disciplinary
proceedings in UK prisons for a range of behaviors, from violence to
disobeying an order. It was hypothesized that:
A. Adjudications would be statistically significantly reduced as a consequence of
removing sources of tension between prison officers and prisoners, and giving
prisoners more responsibility for their actions.
Since prisoners were now no longer dependent on officers to apply for atten-
dance at education and rehabilitation programs and to make contact with their OS,
it was considered that they would be more likely to be motivated to enroll for
these programs (Mann et al. 2013). It was, therefore, hypothesized that:
B. The completion rates of offending behavior programs would significantly increase.
According to research evidence (Bonta and Andrews 2017), if attitudes and behavior
in prison improved (Hypothesis A), and more prisoners were addressing the dynamic
risk factors associated with their offense, it was hypothesized that:
C. Reoffending after release would be reduced among prisoners who had experienced
PSS when compared to those who had not.
Method
Data sample
PSS kiosks, located on prison landings, exist in a small proportion of prisons in
England, Scotland, and Wales, and 13 prisons were chosen for the study as
being those that had introduced PSS prior to 2014. This date allowed at least
6 months prisoner experience of the kiosk system in all prisons and at least
12 months follow-up of proved reoffending (reconvictions in a court) in the
community after release. This sample was drawn from establishments covering
a range of categories of prison, including public and private sector, local,
training, and high security. Hence, these represented differences in jurisdiction,
prison types, and commercial operating models. The identities of prison sites
are anonymized in this paper as agreed in the ethical approval process and in
accordance with non-disclosure agreements.
Comparison groups of similar category prisons without PSS were used to
compare trends in the rate of reoffending nationally. The comparison groups were
based on Bfamily^groupings used by the then National Offender Management
The effect of digital technology on prisoner behavior and...
Service (NOMS) to categorize similar types of prisons in terms of population,
size, and risk (see Ministry of Justice 2015).
1
To avoid sample overlap, prisons
identified as PSS prisons were removed from the comparison group data.
Design
As most of the prisons that had installed kiosks were private prisons, it was not
possible to use public prisons without kiosks as a comparison group for interim
prison outcome measures. There are numerous cultural and operational differ-
ences between public and private prisons, which may make comparisons be-
tween them inappropriate (Hulley et al. 2012). We, therefore, adapted the
randomized controlled Bstepped-wedge^design of Hussey and Hughes (2007),
to enable the most rigorous method available to examine retrospective data,
equating to an observational quasi-experiment. A stepped-wedge randomized
controlled trial (RCT) is a type of cross-over design that allows prisons to
begin as no-intervention controls, and at random intervals to cross over from
the control group to the intervention group. Thus, at the end of the trial, all
prisons will have crossed over to the intervention group.
The stepped-wedge design in Fig. 2shows no interventions in 2007, with a
systematic introduction of the technology to prisons in subsequent years,
resulting in all prisons receiving the technology by 2014. Hence, the design
was able to take account of the dates when a prison received PSS and examine
monthly data from the prison during the time periods before and after the PSS
technology had been installed. As most of the prisons installed kiosks at
different times across a time period from 2007 to 2014 in a pseudo-random
fashion, if PSS was having an impact, one would expect to see a change in the
value of the relevant measure associated with the installation date in each
prison, but no change in the other prisons at those times. This would, therefore,
isolate an effect in each prison as being attributable to the PSS installation (a
within-prison comparison) when no change was occurring in the other prisons
(a between-prison comparison), so minimizing the possibility that the outcome
was due to some change other than the installation of PSS. This approach is
particularly powerful if one can compare the time series of those who are and
those who are not exposed to an intervention (Cleary et al. 2012).
Measures
Selected outcome measures
Although the theory of change for prisoners (Fig. 1) identified a number of variables
that might impact on prison behavior, from a statistical point of view, we chose to select
a primary measure for each main prison priority of providing Bsafe, secure, and decent
prisons^,Bprisoner rehabilitation^,andaBreduction in reoffending^(see Table 1).
1
Available from https://www.gov.uk/government/statistics/proven-reoffending-statistics-july-2012-to-june-
2013.
C. McDougall et al.
Adjudications
It was anticipated in the research design (Hypothesis A) that tensions between prisoners
and staff would be reduced following the introduction of PSS. We, therefore, selected
adjudications as the most appropriate outcome measure for this change of process.
Adjudications are the main disciplinary procedures used in prisons by governors/
directors to consider cases of breaches of prison rules and to impose sanctions on
behaviors ranging from minor acts such as disobeying an order to serious violent
assaults. This outcome measure was recorded across all of the PSS prisons for a period
of seven years on a monthly basis, regardless of the severity of the misconduct.
Analysis of individual prisons was conducted using monthly data points collected over
the full time periods before and after the installation of PSS. The data for each prison
installing PSS were compared in the analysis with data over the same time period in the
other prisons when PSS was not being installed. All prison data were divided by the
size of the population in that prison at the end of each month to control for changes in
the population. It was proposed that, if a meaningful reduction in adjudications could
be associated with the installation of PSS in each prison, this would be evidence that
Fig. 2 Overview of time period of prisoner self-service (PSS) implementation by site
Tab l e 1 Outcome measures corresponding to prison priorities in England and Wales
Prison priorities Outcome measure
Interim Longer term
Safe, secure, and decent prisons Adjudications Proved reoffending
Prisoner rehabilitation OBP completions
Supporting evidence Prison survey
Usage of functions
Note: OBP = offending behavior program
The effect of digital technology on prisoner behavior and...
PSS was likely to be creating that impact, if no other major reason for the change was
evident.
A sensitivity analysis was conducted post hoc on the five prisons with the longest
time periods pre- and post-PSS to examine whether those prisons with shorter time
series might be over-influencing the outcome.
Offending behavior program completions
Offending behavior program (OBP) completions were selected as a measure of
rehabilitation. These were restricted to accredited living skills or thinking skills
programs, as these tend to be run in most prisons and they are general offending
programs and not offense-specific. The number of prisoners completing programs
(OBP completions) was calculated as a proportion of the number of prisoners
starting programs. This measure was selected as a proxy for prisoner commitment
to rehabilitation, and OBP completions are recognized as an indicator of reduction
in reoffending (Lipsey et al. 2007)(HypothesisB).
Longer-term outcome measure: proved reoffending (reconviction in court)
Hypothesis C was based on the assumption that, if Hypotheses A and B were
supported, as predicted by research, this would assist in dynamic risk reduction, leading
to reduced reoffending (Bonta and Andrews 2017; French and Gendreau 2006).
A proved reoffense is defined as any offense committed in a specific follow-up
period that leads to a court conviction or caution (including a further 6-month waiting
period to allow the offense to be proved in court). Proved reoffending data for the PSS
prisons were provided by Justice Statistics Analytical Services (JSAS), Ministry of
Justice, UK.
A 6-month period of releases before and after PSS installation was selected. In order
to be sure that a genuine effect of the technology was measured, a period of 6 months
was allowed without measurement immediately before and after installation. This was
to ensure that there was no contamination of results caused by activities in preparation
for PSS installation and, after PSS, to allow time for equipment to work efficiently and
prisoners to learn how to use the equipment. Therefore, the samples of releases were
from 12 months to 6 months (pre-PSS) and +6 months to +12 months (post-PSS).
Prisoners released during the pre- and post-PSS test phases were followed up for 1 year
(plus 6 months to allow for court conviction) and their proved reoffending recorded.
Supporting information from a process evaluation
Prisoner survey A prisoner survey was conducted in a prison with newly installed PSS
technology. The survey used the newly introduced wing kiosks as a means of
completing the survey 4 weeks after installation. When a prisoner logged on to the
kiosk, the purpose of the survey and the confidentiality of responses were explained,
and each prisoner was invited to complete the questionnaire. The questionnaire
included questions relating to the impact of the technology on prisonerslives, their
relationship with prison officers and family and friends, and the likely effect on
prisonerslives after release.
C. McDougall et al.
Prisoner training to use the kiosks had been conducted by staff, with the aid of
written instructions. Staff also trained prisoners already engaged in peer mentoring
programs in the prison to assist those needing additional help in using the PSS kiosks.
These peer mentors included Blisteners^trained to support prisoners at risk of self-harm
and suicide, Bbuddies^designated to help older prisoners, and educationally trained
mentors who assist prisoners with reading difficulties. This staff/prisoner training
process was thought likely to help improve relationships between staff and prisoners.
Usage of self-service technology Usage data were collected from a main wing in the
process evaluation prison (N
prisoners
= 76) for 1 week at two time periods: (i) 1 week
after and (ii) 4 weeks after the installation of PSS. These data were collected remotely
by the contractor.
Data analysis
Given the nesting of time points of response within prisons and the need to account for
variability at the prison level, the proposed statistical analysis of the interim outcome
measures was longitudinal multi-level modeling (LMM) (Singer and Willett 2003).
This Bmixed effects^method allows assessment of the immediate impact on, and
change over time in, the selected outcome variables associated with the implementation
of an intervention. We were, therefore, able to estimate the impact of PSS on within-
prison change in adjudications over time, controlling for individual prison populations,
and trends over time using monthly data points and installation dates of PSS.
A key decision in the specification of mixed effects models is which explanatory
variables are considered fixed or random effects. Specifying a random effect assumes
that there is variation in the impact of that variable on the outcome, typically around a
central value. It refers to the randomness in the probability model for the cluster-level
coefficients. In our case, the prisons were the clusters and we have assumed that these
prisons represent a sample of the population of prisons that we could have observed.
However, if the impact of a given variable is interesting in itself, say comparing a
specific prison to another prison, then these are referred to as fixed effects (to explore
this distinction in depth, see McCulloch et al. 2008 and Gelman and Hill 2007). Here,
we used a fixed effect for the PSS variable because we were interested in the impact of
technology, so comparing before installation to after installation.
Unfortunately, LMM was not suitable for use in analyzing OBP completions, due to
missing data, which reduced the sample size. In this case, a non-parametric method,
Wilcoxon signed-rank, was used.
The longer-term analysis of reoffending (reconviction in court) also required a
different method of analysis, as only two measures were collected for each prison,
pre- and post-PSS, that is, proportion of releases who reoffended. A related samples
design was used, identifying a cohort of prisoners released 6 months prior to and after
self-service technology implementation in each of seven prison sites,
2
with subsequent
comparison of their one-year reoffending rates.
3
The actual reoffending data were
2
One year reoffending data for seven of the prisons were provided by Justice Statistics Analytical Services
(JSAS); some prisons were excluded due to missing data.
3
This analysis differs from that used by JSAS, who compare full calendar year data against a selected baseline
year, whereas data in the current study were taken from a selected subset to fit the stepped-wedge design,
centered on PSS installation.
The effect of digital technology on prisoner behavior and...
adjusted to take account of the level of risk of prisoners based on historical data (i.e.,
predicted rate of reoffending in each prison). The predicted reoffending rates were
based on scores on the Offender Group Reconviction Scale (OGRS; Howard et al.
2009)
4
averaged at a prison level. To generate a proportional difference in reoffending
between actual and predicted, we used the formula: [actual predicted]/actual. A non-
parametric related samples test was used to compare the proportional differences in
reoffending pre- and post-PSS implementation at the prison level. To compare with the
national trend, we applied the JSAS formula in this case, which compares OGRS
adjusted data with a baseline figure.
5
In the current study, we have used as baseline
figures the prisoner data in the 6 months before the installation of PSS and prisoner data
in the same time periods in the control prisons.
Results
Adjudications
Tab le 2gives summary statistics for adjudications by prison site pre- and post-
implementation of PSS averaged over the available time points. However, this does
not reflect the slope of change across time data points and accounted for in the
statistical analysis. Figure 3shows the aggregated impact of PSS over time, with time
being centered on the date of implementation of PSS in each prison.
The LMM analyzed population-adjusted adjudications from ten prisons (due
to missing data in three prisons), at available time points before and after PSS.
This controlled for prison variability, time point data, and PSS installation date,
and allowed for within- and between-prison analysis. As shown in Table 3,we
found a statistically significant reduction in the level of adjudications following
the installation of kiosks (estimates of fixed and random effects: γ=0.49
[95% CI: 0.75, 0.24], df = 137.65, t=3.79, p< 0.001). We can,
therefore, be confident that this change in adjudications at the prison level
was associated with installation of the technology, as no other reason for the
change affecting all prisons was evident. This reduction in adjudications con-
tinued for 2 years, with frequency very slowly returning towards pre-installation
levels at a rate of 4% per year (0.0082 0.0048 = 0.0034 per month).
As some prisons had a smaller amount of either pre- or post-technology data, a
sensitivity analysis was conducted on adjudications in five prisons which had the
longest time periods of pre- and post-data. The analysis in this subsample appeared
to confirm that the significant reduction in adjudications shown in ten prisons (esti-
mates of fixed and random effects: γ=0.54 [95% CI: 0.82, 0.26], df = 89.13,
t=3.84, p< 0.001) was not over-represented by those with fewer pre- and post-PSS
time data points.
4
The Offender Group Reconviction Scale (OGRS) is a predictor of reoffending based on static risk factors:
age, gender, and criminal history.
5
r
x
1
=r
x
(g
x
G), where r
x
1
= adjusted reoffending rate in period x, r
x
= original reoffending rate in period
x, g
x
= OGRS in period x, and G = OGRS in the baseline year (2011).
C. McDougall et al.
Tab l e 2 Site statistics for proportion of adjudications by prisoner self-service (PSS) implementation
Site PSS implementation
Pre-PSS Post-PSS
Proportion of adjudications Proportion of adjudications
Mean SD SEM Time points Mean SD SEM Time points
1–––– 0.064 0.014 0.002 57
2
0.089 0.010 0.002 21 0.081 0.026 0.003 75
3–––– 0.080 0.026 0.003 57
4
0.605 0.091 0.019 23 0.372 0.231 0.027 72
5 0.064 0.039 0.020 4 0.052 0.020 0.003 53
6 0.094 0.016 0.007 5 0.087 0.025 0.004 52
7–––– 0.109 0.021 0.004 33
8 0.081 0.012 0.005 5 0.085 0.027 0.004 52
9
0.203 0.070 0.011 38 0.131 0.032 0.008 15
10
0.082 0.024 0.004 40 0.088 0.025 0.004 32
11
0.028 0.008 0.002 26 0.035 0.008 0.001 31
12 0.108 0.029 0.004 52 0.126 0.029 0.013 5
13 0.104 0.030 0.003 94 ––––
Overall 0.146 0.167 0.053 31 0.109 0.087 0.050 45
Note: Adjudications as a proportion of the prison population at the site; SD = standard deviation; SEM = stan-
dard error of mean; time points are number of months with data
Sites in post hoc sensitivity analysis subsample
Time period of measurement (years)
3210-1-2-3
Mean Proportion of adjudications
.17
.16
.15
.14
.13
.12
.11
1
0
PSS
implementation
Fig. 3 Population-adjusted proportion of adjudications among sites with pre- and post-PSS means (N= 10)
The effect of digital technology on prisoner behavior and...
Offending behavior program completions
A large amount of program completion information was made available by NOMS from
April 2010. However, information on program Bstarts^was only available from April 2009,
which limited the analysis of the impact of installations that occurred up to 2009. It was,
therefore, not possible to apply the full multi-level model analysis to these data, so a simple
non-parametric analysis was applied to investigate trends. The mean proportion of com-
pleters to starters pre-PSS was 88.25% (SD = 4.55), while post-PSS, the mean was 93.67%
(SD = 4.57). The difference in means was not statistically significant (t=1.96, df = 4,
p= 0.121). This is likely to be due to the small sample size; however, the average completion
rate of 93.67% was near the ceiling of possible performance. Although the sample size limits
our confidence in the results, Fig. 4shows that three out of five of the prisons with complete
data showed a sizeable increase in completions after PSS, which is encouraging and positive.
A larger sample would have allowed for a full analysis, which would have accounted for the
impact of PSS, the different types of prison, and the system-wide changes over time affecting
prisons with and without PSS.
Reoffending (reconviction in court)
Data were available for seven of the 13 prisons to calculate the rate of reoffending. The
missing data were due to some prisons having insufficient releases in the time period in
question or insufficient time to allow an 18-month follow-up after release. A Wilcoxon
signed-rank test indicated that, before PSS in each of the prisons in the stepped-wedge
design, the proportion of actual to predicted offenses (Mdn = 0.09) was significantly higher
than that after PSS installation (Mdn = 0.06). The actual rate was closer to or better than
predicted by OGRS in six of the seven sites and the adjusted rate was lower after PSS than
before PSS (z = 2.03, p=0.04,r= 0.54). The effect size of 0.54 shows that this represents a
large change in outcomes between the pre- and post-PSS cohorts (Cohen 1992).
Tabl e 3 Fixed and random parts
of the model predicting propor-
tion of adjudications (population
adjusted)
***p<0.001;**p<0.01;
*p<0.05
Parameter Estimate (standard error)
Fixed effects
Intercept 2.006 (0.2607)***
Time po int 0.0048 (0.0025)
PSS 0.4938 (0.1303)***
Time point * PSS 0.0082 (0.0024)**
Random effects
(covariance param eters)
Repeated measures
AR1 diagonal 0.1251 (0.0080)***
AR1 rho 0.3807 (0.0372)***
Intercept + time point
[subject = prison site]
Var (intercept) 0.7436 (0.3176)*
Cov (time point, intercept) 0.0014 (0.0018)
Var (time point) 1.89E-05 (1.72E-05)
C. McDougall et al.
Comparison with the national trend in reduced proved reoffending
As we were aware that general prison proved reoffending was decreasing over the 7-
year time period of our study, JSAS also kindly provided us with control group data
from the same Bfamily group^of prisons, that is, similar in terms of size, security
category, and type of offender, to those in our sample, so that we could compare the
trend in prisons in general with those prisons with PSS.
We, therefore, measured the PSS prison proved reoffending data and the control
prison proved reoffending data over the same time periods, adjusted for OGRS scores
(see Fig. 5). The difference between the baseline and post-PSS proved reoffending for
the comparison prisons was 0.78% and for the PSS prisons it was 5.36%,
demonstrating a greater reduction in the PSS prisons than in similarly categorized
prisons in the same time period.
It is evident that PSS prisons started at a higher actual rate of reoffending pre-PSS
than the control prisons. However, the pre-PSS predicted rate (52.00%) was not so
much higher than that of the controls (48.83%). This resulted in a larger decrease post-
PSS, with the PSS prisons ending with a smaller proportional difference between their
Post-InstallationPre-Installation
11
10
9
7
5
Site
Mean (%)
100.00
95.00
90.00
85.00
80.00
Fig. 4 Proportion of completers relative to starters before and after PSS installation
The effect of digital technology on prisoner behavior and...
actual and predicted rates. This is in line with the risk principle that higher risk
offenders have more scope for improvement (Bonta and Andrews 2017).
Prisoner survey
In one prison where PSS had been installed 4 weeks previously, we conducted a
prisoner survey. Out of a possible 1389 prisoners, 743 (53%) responded to the
questionnaire. Although only one-half of the population responded, this is more than
double what can be expected and is considered to be an exceptionally good response
rate in a prisoner survey.
6
Below, we divide the prisoner survey questions and responses
by two relevant prison priority performance monitoring areas (Bsafe, decent, and secure
prisons^and Bprisoner rehabilitation^).
Prisoner survey results
Of the 743 respondents, 93% thought the kiosks were Beasy^or Bvery easy^to use,
even though very little formal training was offered, and 7% thought the kiosks were
Bdifficult^or Bvery difficult^to use. When asked BDid you get enough training/help to
use the kiosks?^, 10% thought the training/help was about right, 80% of respondents
6
An example of a typical prison survey response rate is 4% to 25% across eight prisons, e.g., published in
Third Sector Research Centre Working Paper 61, 2011, a report on BOffender engagement with third sector
organisations: a national prison-based survey^.
Cohort Period
Post-Installation (Adjusted for OGRS)Baseline
Prisoner Self-Service
Prisons
Comparator Prisons
Reoffending (%)
57.0
55.0
53.0
51.0
Group
Fig. 5 Summary of change in reoffending compared to the national trend
C. McDougall et al.
said they had had no training/help or not much training/help, and 10% said they had
Bquite a bit^or Bvery much^training/help to use the kiosks.
Safe, decent, and secure prisons
Many of the staff, including the director of the prison where the survey was conducted,
thought that the kiosks would give the prisoners more responsibility and control over
their lives in prison. When asked BHave the kiosks given you more control over your
life in prison?^, 55% thought the kiosks had given them Bmore^or Bmuch more^
control over their lives in prison. This was the highest affirmatory response in the
survey, which suggests that the technology was having this impact. Meanwhile, 36%
thought the kiosks had made no difference, while 8% thought the kiosks had given
them Bless^or Bmuch less^control over their lives.
When prisoners were asked if PSS had affected their relationships with prison
officers, 32% thought relationships were Bbetter^or Bmuch better^after PSS, 58%
thought PSS had made no difference to relationships with officers, and 10% thought
relationships were Bworse^or Bmuch worse^.
Prisoner rehabilitation
Following the introduction of the kiosks, 37% thought that relationships with family
and friends were Bbetter^or Bmuch better^, 53% thought that PSS had made no
difference to relationships with family and friends, and 10% thought that relationships
were Bworse^or Bmuch worse^.
When asked if using the kiosks would give them more confidence to deal with
information technology-enabled services in the outside world, 43% said PSS had given
them Bmore^or Bmuch more^confidence, 50% said PSS had made no difference to
their confidence, and 7% said that PSS had made them Bless^or Bmuch less^confident.
Usage of the self-service functions
There was a strong correlation in usage between Time 1 (after 1 week) and Time 2
(after 4 weeks) (r=0.92,p< 0.001), with the most popular functions being personal
account balance, prison shop, phone top-up, and personal timetable (Table 4). Usage of
all of these increased over the two time points. The only function to decrease was the
noticeboard, which went down by 59%. By far the most used function was personal
account balance, which accounted for 1604 of the total number of log-ons of 1785. The
difference between Time 1 and Time 2 usage was not statistically significant. However,
when the negative noticeboard usage was removed, there was a statistically significant
result (z=2.524, p=0.012,r= 0.89) with a large effect size.
Discussion
This was the first quantitative evaluation of PSS technology in UK prisons, and, to the
authorsknowledge, the first internationally. The evaluation has shown a statistically
significant reduction in adjudications, the main disciplinary measure in UK prisons,
The effect of digital technology on prisoner behavior and...
between pre- and post-PSS installation, and a statistically significant reduction in proved
reoffending (reconviction in court) with a large effect size between groups of prisoners
released before and after PSS installation. These results support Hypotheses A and C.
The study adopted a rigorous natural stepped-wedge design, adapted from the
Hussey and Hughes (2007) randomized controlled stepped-wedge design.
Adjudications were analyzed using a within-prison and between-prison longitudinal
multi-level analysis.
The reoffending analysis controlled for the risk level of the offenders in the
establishment using OGRS, a leading international risk measure in terms of its predic-
tive validity (Yang et al. 2010). The analysis used the proportion of difference between
actual reoffending and predicted reoffending as a measure of reduction in reoffending
pre- and post-PSS installation. These findings could not be accounted for by national
reductions in reoffending over the 7-year study period, since a comparison with Bfamily
groups^of prisons, considered to be of the same level of security, type of prisoner, and
risk as the PSS prisons, over the same time periods, showed a reduction in reoffending
of 0.78%, as compared to the PSS prisonsreduction of 5.36%.
These results imply that behavior in prison improved with the introduction of PSS, and
this was reflected in improved behavior after release. This finding is in keeping with other
research which has shown that adjudications by adult inmates are a reliable predictor of
future reoffending(Cochranetal.2014; French and Gendreau 2006;Heiletal.2009).
OBP completions showed a non-statistically significant increase, thus failing to
support Hypothesis B. This may be attributable to the amount of missing data for this
variable and should be measured more reliably in future prospective research. Although
we were unable to demonstrate that the completion rate of OBPs increased significantly
after PSS was introduced, this measure did increase from 88.25% to 93.67%, which
takes us near the ceiling of possible performance. The perceived stigmas experienced
by prisoners in undertaking treatment are thought to affect their personal change
process and can lead to treatment refusal (Mann et al. 2013). This improvement in
OBP completions, although non-significant, indicates that the ability of prisoners to
take personal responsibility for their applications to attend OBPs may overcome some
of the resistance Mann et al. observed and motivate prisoners to take advantage of these
program opportunities. This needs to be tested in future research.
Tabl e 4 Frequency of PSS
functions usage and log-ons per
week
Function Main wing (N
prisoners
=76)
Time 1 ( n) Time 2 (n) Change (%)
Account balance 1405 1604 14.16
Prison shop 409 511 24.94
Phone top-up 409 511 24.94
FAQs 8 121 1412.50
Messages 56 77 37.50
Noticebo ard 695 284 59.14
Surveys 13 21 61.54
Timetable 202 286 41.58
Log-on total 1699 1785 5.06
C. McDougall et al.
Changing the prison culture
It is important to reflect on the mechanism that achieved the statistically significant results
in relation to adjudications and reoffending. Giving prisoners more control of their lives in
prison by means of self-service technology may have made a contribution to changing the
culture of the prison environment. We know that some sources of tension between prison
officers and prisoners were removed, as prisoners were in control of their own requests
rather than being dependent on officers. This development is not expected to fundamen-
tally change the relationships between officers and prisoners, which is a complex process,
but it does remove a number of sources of frustration and disputes from the environment.
This may also take away sources of perceived unfairness attributed as a major cause of
anger and frustration for prisoners in GST (Agnew 2006). It is, therefore, proposed that
introduction of the self-service may have contributed to this reduction in perceived
unfairness, hence significantly reducing the level of adjudications.
Evidence that taking control was happening came from the process evaluation con-
ducted in one prison prior to the main evaluation (C. McDougall and D.A.S. Pearson,
2014, Process evaluation: The prisoner Custodial Management System (CMS),
Unpublished). When we examined the data on the usage of functions among prisoners,
we found that the most frequently used functions were those that demonstrated taking
responsibility for managing finances, buying products from the shop, topping up their pay
phones, and consulting their own individual timetable, that is, taking control of their lives
in prison. These are all functions that would have been slow and error-prone under the old
paper-based system. The function for arranging visits from an availability timetable was
not yet in operation at the time of the process evaluation, but it is anticipated that this
would be another popular measure of Btaking control^in the prisons in the outcome study.
The least used functions were those that were less instrumental, such as reading the
noticeboard and messages, frequently asked questions, and completing surveys.
A second source of evidence came from the prisoner survey. Despite the lack of
enthusiasm for surveys coming from the usage data, the highest positive response was
that 55% of prisoners who responded thought the kiosks gave them more control over their
lives in prison and 43% thought that the kiosks had given them much more confidence in
dealing with information technology-enabled services in the modern world.
Although having more control over ones life in prison was not specifically tested in
the Listwan et al. (2013) study on strain and inmate recidivism, Bnegative prison
environment^was found to be associated with rearrest and reincarceration, and was
linked to being in control, as one of the main hypothesized overarching contributors to
the strains of imprisonment, together with perceived injustice. This lends support to the
view that the change in culture of the prison associated with PSS may have impacted
positively upon prisoner attitudes to the prison and after release.
It is difficult to interpret why the positive effect on adjudications was only achieved over
2 years before beginning to rise again. The answer may lie in studies which have examined
the duration of treatment effects over time. Prendergast et al. (2004), who conducted a 5-year
follow-up in an RCT of drug offenders in a therapeutic community, found that the
intervention effect size was reduced between year 1 and year 3 at follow-up (Cohens
d= 0.31 reduced to 0.13), indicating a diminished impact. It was found, however, that,
where there was further involvement in aftercare in the intervention group, this was a
significant predictor of lower levels of return to prison. It may, therefore, be that aftercare is
The effect of digital technology on prisoner behavior and...
particularly important to maintain the initial changes due to the opportunities in accessing
services afforded by prisoner self-service. However, a contrary view is presented by Jolliffe
et al. (2013). A 10-year follow-up of a high intensity training regime for young offenders
with added educational training initially showed a statistically significant reduction in
reoffending at 2 years, but the superiority over the control group diminished over time (at
4 years). However, follow-up at 10 years showed that the cumulative number of convictions
saved was 3.35 at 10 years compared to 1.35 at 2 years, and the benefit:cost ratio increased.
Therefore, a deterioration of impact in the short term does not necessarily mean that the
benefits will not be experienced over a longer time period. This feature of benefits
maintenance should be examined in future research.
Improving rehabilitation
A prison culture supportive of rehabilitation has also been recognized as contributing to
effective behavior change (Lambert et al. 2011;Mannetal.2013; Viglione et al. 2017) and,
ultimately, reduction in reoffending (Listwan et al. 2013). The introduction of PSS may also
have had a direct impact on activities relating to rehabilitation. As a result of PSS, prisoners
are becoming familiar with modern digital technology, while improving access to their own
basic requirements. These skills will become useful when preparing for reentry to society.
Although it is recognized that there is a wide range of ability among prisoners in the use of
digital technology, there are those who may never have been able to have access and learn to
use the technology without a great deal of assistance. This is now made available, with
regular daily practice, in the prison environment. In the prisoner survey, 43% said they had
increased confidence in dealing with digitally enabled services on release, which must be of
assistance on reentry.
It is also possible that the effect of giving control to prisoners over some aspects of
their lives in prison may have an impact on dynamic risk, providing them with more
skills in coping in a non-criminal manner, resulting in a reduction in adjudications and
reoffending (e.g., Bonta and Andrews 2017) (Hypothesis C).
The results from this study support the theory of change (Fig. 1), which also anticipated
that the introduction of PSS could impact on adjudications and, ultimately, reoffending. The
pathways from installation of PSS to the positive outcome measures were also, to some
extent, supported. Prisoners were given the opportunity for more self-responsibility, and
were no longer dependent on prison officers. Hence, they were able to make direct contact
with offender supervisors and have direct access to education and offending behavior
programs. Contact with families was made easier by the ability to organize their own visits
from an availability timetable. Although we did not test directly the impact of these abilities
on attitudes and well-being, we did ask for opinions via the survey and there did appear to be
fewer frustrations, as illustrated in reduced adjudications. In the prisoner survey, 31% of
prisoners thought relationships with officers were Bbetter^or Bmuch better^, and 37% of the
sample thought that relationships with family and friends were either^better^or Bmuch
better^after PSS. It seems likely that this improvement in relationships in general and more
confidence in life skills due to the use of technology could have contributed to the outcome
of non-criminal coping and reduced reoffending.
There appears to be an important link between a supportive prison culture and attempts
to achieve rehabilitation with programs and supervision. In the case of technology, it
appears that the opportunity to self-manage within a secure and dependable environment
C. McDougall et al.
instills a sense of autonomy in prisoners that is generally lacking when they are dependent
on officers for the most menial daily tasks (Dirkzwager and Kruttschnitt 2012). If a
prisons culture and fitness for purpose are able to combine to form a rehabilitative
climate, this may motivate and encourage the vital Bresponsivity^factor (Birgden
2004), which is an important element of the RNR model.
Of course, no initiative will help all offenders and there will be a minority that do not
make the most of self-service or who attempt to use it for anti-social purposes. This study,
however, provides optimism for the effects of this technology in supporting rehabilitation.
Limitations of the data
The current study was not without its limitations. We did not have data to track individual
prisoners, and, so, the focus of the study was on the effect of PSS at the prison level. The
promising results of the current study raise the importance of testing the technology on
individual prisonersbehavior. Second, collecting historical data over a period of 7 years is
subject to errors and missing data. Adjudications were, however, one of the most robust
measures and, hence, many of the prisons were included. The number of prisons in the
reoffending analysis was reduced to seven, as exclusions had to be made where small
numbers of offenders were released and where there was insufficient time to follow up
reoffending for a year. Despite the small numbers, the effect size was large, due to the
consistent positive impact, which we attribute to the prisoners making use of the self-
service technology and taking more control over their lives in prison.
The measure of prisoner attitudes via the survey of prisoners in one establishment also
lacked data, as all possible participants did not complete the questionnaire. The percentage
that did complete the questionnaire was, however, substantial for a prisoner population.
There is no means of knowing if those that declined to complete the survey had similar
views to participants. The level of responding, 53%, is, nevertheless, high in a prisoner
context, and may be attributable to the ease of access, completion, and return, due to the
PSS kiosks. This is encouraging for future surveys in prisons with self-service technology.
Implications
The introduction of digital technology to prisons is in its infancy, and there is much
potential for development. The technology is likely to bring improved efficiencies by
automating time-consuming administrative tasks and releasing officers to be more
involved in making prisons safer by improved visibility on the wings and aiding
prisoner rehabilitation. This is likely to have a positive impact on staff well-being.
Prisoners have adapted well to using self-service technology. Although it has proved
beneficial, the range of applications is limited due to provision via wing-based kiosks, so
restricting privacy and usage. Consideration is being given to the use of tablet computers,
which would allow more scope for educational features, personal assessments, for example,
of risk, need, and responsivity, such as the Risk Need Perception Survey (King 2016), and
appropriate interactive interventions. The current study also indicates the importance of self-
service technology operating within a rehabilitative prison climate and linking with external
probation supervision where prisonersmotivation to change is supported with aftercare.
Given the importance highlighted in research of addressing the needs of prisoners on
release (Bonta and Andrews 2017; Wolff et al. 2012), self-service technology presents
The effect of digital technology on prisoner behavior and...
an opportunity for facilitating reentry preparation and promoting offender rehabilitation
on a prison-wide scale. With the aid of self-service technology, this process could reach
a wider participant pool, not just in selected group programs but across the prison in
many aspects of prison life. The digital environment can assist the real-time acquisition
of life skills, such as finance management, managing relationships within the prison,
developing skills to aid employment on release, and strengthening family ties. In the
process evaluation (McDougall and Pearson 2014), officers had noticed a distinctly
more positive problem-solving approach from prisoners when they had more access to
information and control of outcomes.
The programs and activities in prison still need to adhere to the RNR principles by
matching interventions to levels of risk, identifying and addressing criminogenic needs,
and responding to the learning style of the prisoner. However, within this framework,
PSS provides the opportunity to develop self-responsibility in coping more effectively
with prison life and maintaining purposeful pro-social activities and relationships in
preparation for release. This approach could be facilitated by ensuring that the time
saved by relieving prison officers of the laborious basic tasks of looking after prisoners
is channeled into supporting prisoners in taking advantage of the opportunities for self-
responsibility for their own rehabilitation.
Conclusion
The present study offers an important contribution to the field of corrections in
providing the first assessment of the effect of prison digital technology on
prison behavior and reoffending. Although there were some effects of missing
data, this study can be regarded as a positive indication of the likely effect of
technology on the lives of prisoners, officers,andthecommunityintowhich
prisoners will be released. Our study demonstrated that prison behavior was
significantly improved and reoffending in the community was significantly
reduced. Supporting information from usage data and a prisoner survey showed
that the interaction with the technology produced a feeling of worth and
personal control. This suggests that, by introducing prisoners to modern tech-
nology, it can transform their lives from dependency to self-responsibility,
where they can learn new ways of behaving in a supportive rather than a
punitive environment. As for staff, they are released from mundane administra-
tive tasks, offering the opportunity to undertake a more fulfilling role of
assisting prisoners to be involved in purposeful activities. It is rare for a change
inaprisonregimetobewelcomedbybothprisonersandprisonofficers,
resulting in a positive outcome. Prisoners have shown themselves to be recep-
tive to taking responsibility using modern technology when they recognize the
benefits to themselves. We should take the opportunity to channel this enthu-
siasm into reform and rehabilitation when it is presented.
Acknowledgements The authors would like to commend Francis Toye, Director of Unilink, who, having
funded the project, deliberately abstained from trying to influence the research and the outcomes, and Roger
Holding, whose original idea it was to seek a rigorous independent evaluation. Neither of the above was
known to the authors prior to the commission.
C. McDougall et al.
Thanks are due to HM National Offender Management Service (NOMS) National Research Com-
mittee, who critically appraised the research application for quality of design and ethical issues, and
made helpful suggestions to improve the research.
Thanks are also due to NOMS Statistical Departments, who advised on availability and provided
data to assist the evaluation, and to the Ministry of Justice, Justice Statistics Analytical Services, who
provided us with selected samples of proven reoffending data and helpfully commented on the
analysis.
We are grateful to the unnamed prisons in England, Scotland, and Wales, and to their executive
directors, who willingly helped with our research and entrusted us with commercially sensitive data to
assist the evaluation.
Finally, we also thank Dr. Mona Kanaan, statistician in the Health Sciences Department at the University of
York, for her valuable comments on the statistical analyses.
Funding This work was funded via the universities of York and Portsmouth by Unilink Software Ltd. The
sponsor specifically requested an independent university evaluation and made no interventions in the research
design, analysis, interpretation, or conclusions.
References
Agnew, R. (2006). Pressured into crime: An overview of general strain theory. New York: Oxford University
Press.
Andrews, D. A., & Bonta, J. (1995). The Level of Service InventoryRevised. Toronto: Multi-Health Systems.
Andrews, D. A., Bonta, J., & Hoge, R. D. (1990). Classification for effective rehabilitation: Rediscovering
psychology. Criminal Justice and Behavior,17(1), 1952. https://doi.org/10.1177
/0093854890017001004.
Andrews, D. A., Bonta, J., & Wormith, J. S. (2011). The risk-need-responsivity (RNR) model: Does adding
the good lives model contribute to effective crime prevention? Criminal Justice and Behavior,38(7),
735755. https://doi.org/10.1177/0093854811406356.
Batchelder, J. S., & Rachal, J. R. (2000). Efficacy of a computer-assisted instruction program in a prison
setting: An experimental study. Adult Education Quarterly,50(2), 120133. https://doi.org/10.1177
/07417130022086946.
Birgden, A. (2004). Therapeutic jurisprudence and responsivity: Finding the will and the way in offender
rehabilitation. Psychology, Crime & Law,10(3), 283295.
Bonta, J., & Andrews, D. A. (2017). The psychology of criminal conduct (6th ed.). New York: Routledge.
ISBN 9781138935778.
Bonta, J., Rugge, T., Scott, T. L., Bourgon, G., & Yessine, A. K. (2008). Exploring the black box of
community supervision. Journal of Offender Rehabilitation,47(3), 248270.
Bourgon, G., Bonta, J., Rugge, T., Scott, T. L., & Yessine, A. K. (2009). Translating what worksinto
sustainable everyday practice: Program design, implementation and evaluation. 2009-05. Public Safety
Canada. http://www.publicsafety.gc.ca/cnt/rsrcs/pblctns/2009-05-pd/index-en.aspx. Accessed 1
July 2017.
Cabinet Office. (2016). Queens Speech 2016. London: Cabinet Office.
Champion, N., & Edgar, K. (2013). Through the gateway: How computerscan transform rehabilitation. Prison
Reform Trust and Prisoners Education Trust. https://fbclientprisoners.s3.amazonaws.
com/Documents/CQ%20through%20the%20gateway%20WEB1.pdf. Accessed 1 July 2017.
Cid, J. (2009). Is imprisonment criminogenic? A comparative study of recidivism rates between
prison and suspended prison sanctions. European Journal of Criminology,6,459480.
Cleary, P. D., Gross, C. P., Zaslavsky, A. M., & Taplin, S. H. (2012). Multilevel interventions: Study design
and analysis issues. Journal of the National Cancer Institute Monographs,2012(44), 4955. https://doi.
org/10.1093/jncimonographs/lgs010.
Cochran, J. C., Mears, D. P., Bales, W. D., & Stewart, E. A. (2014). Does inmate behavior affect
post-release offending? Investigating the misconduct-recidivism relationship among youth and
adults. Justice Quarterly,31(6), 10441073. https://doi.org/10.1080/07418825.2012.736526.
Cohen, J. (1992). A power primer. Psychological Bulletin,112(1), 155159. https://doi.org/10.1037/0033-
2909.112.1.155.
The effect of digital technology on prisoner behavior and...
Dirkzwager, A. J. E., & Kruttschnitt, C. (2012). Prisonersperceptions of correctional officersbehavior in
English and Dutch prisons. Journal of Criminal Justice,40(5), 404412. https://doi.org/10.1016/j.
jcrimjus.2012.06.004.
French, S. A., & Gendreau, P. (2006). Reducing prison misconducts: What works! Criminal Justice and
Behavior,33(2), 185218. https://doi.org/10.1177/0093854805284406.
Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical models (analytical
methods for social research). Cambridge: Cambridge University Press.
Goggin, C., & Gendreau, P. (2006). The implementation and maintenance of quality services in offender
rehabilitation programmes. In C. R. Hollin & E. J. Palmer (Eds.), Offending behaviour programmes:
Development, application, and controversies (pp. 209246). Chichester: John Wiley & Sons.
Heil, P., Harrison, L., English, K., & Ahlmeyer, S. (2009). Is prison sexual offending indicative of community
risk? Criminal Justice and Behavior,36(9), 892908. https://doi.org/10.1177/0093854809338989.
Helsper, E. J., & Eynon, R. (2013). Distinct skill pathways to digital engagement. European Journal of
Communication,28(6), 696713.
Howard, P., Francis, B., Soothill, K., & Humphreys, L. (2009). OGRS 3: The revised Offender Group
Reconviction Scale (research summary 7/09). London: Ministry of Justice.
Hulley, S., Liebling, A., & Crewe, B. (2012). Respect in prisons: Prisonersexperiences of respect in public
and private sector prisons. Criminology & Criminal Justice,12(1), 323. https://doi.org/10.1177
/1748895811423088.
Hussey, M. A., & Hughes, J. P. (2007). Design and analysis of stepped wedge cluster randomized trials.
Contemporary Clinical Trials,28(2), 182191. https://doi.org/10.1016/j.cct.2006.05.007.
International Corrections and Prisons Association (ICPA) (2017). Innovation in rehabilitation: Building better
futures. 2227 October 2017, London. https://icpa.ca/london2017/. Accessed 1 July 2017.
Jewkes, Y., & Reisdorf, B. C. (2016). A brave new world: The problems and opportunities presented by new
media technologies in prisons. Criminology & Criminal Justice,16(5), 534551. https://doi.org/10.1177
/174889581665495.
Jolliffe, D., Farrington, D. P., & Howard, P. (2013). How long did it last? A 10-year reconviction follow-up
study of highintensity training for young offenders. Journal of Experimental Criminology,9(4), 515531.
King, C. M. (2016). The prediction of criminal recidivism using self- and evaluator appraised risk and needs.
Doctoral dissertation. http://search.proquest.com/docview/1790102519. Accessed 1 July 2017.
King, C. M., Heilbrun, K., Kim, N. Y., McWilliams, K., Phillips, S., Barbera, J., & Fretz, R. (2017). Tablet
computers and forensic and correctional psychological assessment: A randomized controlled study. Law
and Human Behavior.https://doi.org/10.1037/lhb0000245.
Knight, V. (2015). Some observations on the digital landscape of prisons today. Prison Service Journal,229,39.
Lambert, E. G., Altheimer, I., Hogan, N. L., & Barton-Bellessa, S. M. (2011). Correlates of correctional
orientation in a treatment-oriented prison: A partial test of personenvironment fit theory. Criminal Justice
and Behavior,38(5), 453470. https://doi.org/10.1177/0093854811400716.
Lipsey, M. W., & Wilson, D. B. (1998). Effective intervention for serious juvenile offenders. In R. Loeber &
D. P. Farrington (Eds.), Serious and violent juvenile offenders: Risk factors and successful interventions
(pp. 313345). Thousand Oaks, CA: Sage.
Lipsey, M., Landenberger, N. A., & Wilson, S. J. (2007). Effects of cognitive-behavioral programs for criminal
offenders: A systematic review. The Campbell Collaboration Library. http://www.campbellcollaboration.
org/lib/project/29/. Accessed 1 July 2017.
Listwan, S. J., Sullivan, C. J., Agnew, R., Cullen, F. T., & Colvin, M. (2013). The pains of imprisonment
revisited: The impact of strain on inmate recidivism. Justice Quarterly,30(1), 144168. https://doi.
org/10.1080/07418825.2011.597772.
Lösel, F., Pugh, G., Markson, L., Souza, K. A., & Lanskey, C. (2012). Risk and protective factors in the
resettlement of imprisoned fathers with their families. Ormiston Childrens and Families Trust, Milton.
Accessed 1 July 2017 from Google Scholar.
Lowenkamp, C. T., Holsinger, A., Robinson, C. R., & Alexander, M. (2014). Diminishing or durable
treatment effects of STARR? A research note on 24-month re-arrest rates. Journal of Crime and
Justice,37(2), 275283. https://doi.org/10.1080/0735648X.2012.753849.
Mann, R. E., Webster, S. D., Wakeling, H. C., & Keylock, H. (2013). Why do sexual offenders refuse treatment?
Journal of Sexual Aggression,19(2), 191206. https://doi.org/10.1080/13552600.2012.703701.
Mazerolle, P., & Piquero, A. (1998). Linking exposure to strain with anger: an investigation of deviant
adaptations. Journal of Criminal Justice,26(3), 195211.
C. McDougall et al.
McCulloch, C. E., Searle, S. R., & Neuhaus, J. M. (2008). Generalized, linear, and mixed models. In Wiley
series in probability and statistics. Applied probability and statistics. New York: John Wiley & Sons.
McNeill, F., & Weaver, B. (2010). Changing lives? Desistance research and offender management. The
Scottish Centre for Crime & Justice Research (SCCJR), project report no. 03/2010. http://www.sccjr.ac.
uk/publications/changing-lives-desistance-research-and-offender-management. Accessed 1 July 2017.
Ministry of Justice. (2011). 2011 Compendium of re-offending statistics and analysis. London, England:
Ministry of Justice.
Ministry of Justice. (2015). Proven re-offending statistics quarterly bulletin: July 2012 to June 2013, England
and Wales. London, England: Ministry of Justice. http://www.gov.uk/government/statistics/proven-
reoffending-statistics-july-2012-to-june-2013. Accessed 1 July 2017.
Nagin, D. S., Cullen, F. T., & Jonson, C. L. (2009). Imprisonment and reoffending. In M. Tonry (Ed.), Crime
and justice: A review of research (vol. 38, pp. 115200). Chicago: University of Chicago Press.
Netto, N. R., Carter, J. M., & Bonell, C. (2014). A systematic review of interventions that adopt the BGood
Lives^approach to offender rehabilitation. Journal of Offender Rehabilitation,53(6), 403432.
https://doi.org/10.1080/10509674.2014.931746.
Office of the Attorney General. (2016). Justice Department announces reforms at bureau of prisons to reduce
recidivism and promote inmate rehabilitation. https://www.justice.gov/opa/pr/justice-department-
announces-reforms-bureau-prisons-reduce-recidivism-and-promote-inmate.Accessed1July2017.
Pearson, D. A. S., McDougall, C., Kanaan, M., Bowles, R. A., & Torgerson, D. J. (2011). Reducing criminal
recidivism: evaluation of Citizenship, an evidence-based probation supervision process. Journal of
Experimental Criminology,7(1), 73102.
Petersilia, J. (2016). Realigning corrections, California style. The Annals of the American Academy of Political
and Social Science,664(1), 813. https://doi.org/10.1177/0002716215599932.
Polaschek, D. L. (2012). An appraisal of the riskneedresponsivity (RNR) model of offender rehabilitation
and its application in correctional treatment. Legal and Criminological Psychology,17,117. https://doi.
org/10.1111/j .2044- 8333.2011.02038.x.
Prendergast, M. L., Hall, E. A., Wexler, H. K., Melnick, G., & Cao, Y. (2004). Amity prison-based therapeutic
community: 5-year outcomes. The Prison Journal,84(1), 3660.
Robinson, C. R., Lowenkamp, C. T., Holsinger, A. M., VanBenschoten, S., Alexander, M., & Oleson, J. C.
(2012). A random study of Staff Training Aimed at Reducing Re-arrest (STARR): Using core correctional
practices in probation interactions. Journal of Crime and Justice,35(2), 167188. https://doi.org/10.1080
/0735648X.2012.674823.
Sherman, L. W. (1993). Defiance, deterrence, and irrelevance: A theory of the criminal sanction. Journal of
Research in Crime and Delinquency,30,445473.
Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and event
occurrence. New York: Oxford University Press.
Smith, P., Schweitzer, M., Labrecque, R. M., & Latessa, E. J. (2012). Improving probation officers
supervision skills: An evaluation of the EPICS model. Journal of Crime and Justice,35(2), 189199.
Taxman, F. S. (2008). No illusions: Offender and organizational change in Marylands proactive community
supervision efforts. Criminology & Public Policy,7, 275302. https://doi.org/10.1111/j.1745-
9133.2008.00508.x.
Thompson, C. (2014). Pennsylvania prison inmates may soon be allowed a version of electronic tablets.
PennLive, 24 February 2014.
Tiedt, A. D., & Sabol, W. J. (2015). Sentence length and recidivism among prisoners released across 30 states in 2005:
Accounting for individual histories and state clustering effects. Justice Research and Policy,16(1), 5064.
Tran, M. P. (2014). Pennsylvania DOC launches interactive map to connect individuals to reentry services.
Pennsylvania: Justice Center. https://csgjusticecenter.org/reentry/posts/pennsylvania-doc-launches-
interactive-map-to-connect-individuals-to-reentry-service. Accessed 1 July 2017.
Viglione, J., Lerch, J., Rudes, D. S., & Taxman, F. S. (2017). Big stick management: Misconducts as discipline within
a correctional reentry facility. Criminal Justice and Behaviour,44(2), 163183. https://doi.org/10.1177
/0093854816668917.
Ward, T. (2010). The good lives model of offender rehabilitation: Basic assumptions, etiological commit-
ments, and practice implications. In F. McNeill, P. Raynor, & C. Trotter (Eds.), Offender supervision: New
directions in theory, research and practice (pp. 4164). New York, NY: Routledge.
Wolff, N., Shi, J., & Schumann, B. E. (2012). Reentry preparedness among soon-to-be-released inmates and the role
of time served. Journal of Criminal Justice,40(5), 379385. https://doi.org/10.1016/j.jcrimjus.2012.06.008.
Yang, M., Wong, S. C., & Coid, J. (2010). The efficacy of violence prediction: A meta-analytic comparison of
nine risk assessment tools. Psychological Bulletin,136(5), 740767. https://doi.org/10.1037/a0020473.
The effect of digital technology on prisoner behavior and...
Cynthia McDougall PhD, is a professor of forensic psychology at the University of York, UK. Formerly she
was Head of Psychology for Prisons and Probation in England and Wales. She has worked as a consultant to
prison and probation services on changing offending behaviour and evaluating effectiveness using randomized
controlled methodologies. She is currently advising on evaluation of the impact of digital technology on prison
regimes, staff and prisoners. Cynthia is a member of the Campbell Collaboration and a Fellow of the Academy
of Experimental Criminology.
Dominic Pearson PhD, is a senior lecturer in the International Centre for Research in Forensic Psychology at
the University of Portsmouth, UK. He is a chartered and registered forensic psychologist, with many years of
experience with Her Majestys Prison and Probation Services. He is the lead investigator on a randomized
controlled trial evaluation of a domestic violence programme in Portsmouth (England), and is a member of the
research team advising on evaluation of the impact of digital technology on prison regimes, staff and prisoners.
Dominic is a Fellow of the British Psychological Society.
David Torgerson PhD, is director of the University of York Trials Unit. He has a methodological interest in
randomised trials in both the social sciences and health sciences and has over 200 peer reviewed publications.
He is co-author of Designing Randomized Trials in Health, Education, and the Social Sciences (Palgrave
Macmillan 2008), and has collaborated with the Behavioural Insights Team at HM Cabinet Office, supporting
the design and development of educational and criminal justice trials.
Maria Garcia-Reyes PhD, is an economist and lecturer in the Department of Economics and Related Studies
at the University of York. She has a wide range of research interests such as income polarization, inequality,
poverty and income distribution as well as criminology. She has collaborated with the York Forensic
Psychology team on a number of studies helping to develop cost/benefits analyses of criminal justice
interventions.
C. McDougall et al.
... These technologies provide inmates with access to methods of communicating with family and friends, such as phone or video calling, as well as selfservice administration, including account management, request forms and personal legal information (Krikorian & Coye, 2019;Palmer et al., 2020). Devices in various jurisdictions also provide access to entertainment resources, e-learning and public service information (Krikorian & Coye, 2019;McDougall et al., 2017). ...
... In outlining the aims of self-service technologies, both industry professionals and researchers have focused on a variety of outcomes. These include reducing the administrative and inmate management burden on staff, increasing inmate autonomy, improving the relations between staff and inmates, and de-escalating friction between inmates (Krikorian & Coye, 2019;McDougall et al., 2017;Palmer et al., 2020). Others have highlighted the value of tablets for incentivising good behaviour, arguing that inmates avoid behaviour that could cause them to lose access to the tablets (Coppola, 2017). ...
... Tablets have also been identified as providing an opportunity to increase access to, and dosage of, behavioural interventions through online delivery of programs and education during incarceration (Krikorian & Coye, 2019;McDougall et al., 2017). ...
Technical Report
Full-text available
Aims To examine inmates’ experiences of the implementation of digital tablets in NSW correctional centres. The study examines inmates’ use of the tablets and the issues affecting their use, as well as their views on the impacts of the tablets on their lives and the prison environment. Methods A narrative qualitative approach was used to examine inmates’ reflections on their experiences. Data was collected through semi-structured face-to-face interviews with 20 inmates in two pilot centres where the tablets were introduced. Results The tablets were widely accepted and utilised; however, initial support and training for using the tablets was reportedly limited and inconsistent. Inmates saw the tablets as an important part of the nightly post-lock-in routine, with the phone feature noted as the most popular and heavily used feature. Implementation issues such as unreliable Wi-Fi connectivity and poor-quality headphones were a cause of frustration for some inmates but did not discourage them from using the tablets. Inmates reported that access to the tablets had both direct and indirect positive impacts on their experience of life in prison, including improving their relationships in and out of prison, providing them with a greater sense of autonomy, and enhancing their overall wellbeing. Conclusion The study provided an in-depth understanding of inmates’ experiences of the implementation of tablets and the perceived benefits associated with tablet use. Several identified implementation issues have the potential to affect some of the benefits provided by the tablets. Both the benefits and complexities of implementing digital tablets are likely to increase with the ongoing development of tablet features and content, therefore consideration should be given to similar technical and administrative issues that may arise with a focus on developing best practices for continued implementation of tablets.
... Offenders are among the marginalised groups who are excluded from digital access because they have no power to make decisions on issues that affect their future while incarcerated. Those who have been imprisoned for a long time will not recognize the modern world after being released because they have no experience with the digital tools they will use to function in a digital society (McDougall & Pearson, 2020). While some correctional facilities have implemented ICT for security purposes, convicts' usage of ICT has taken longer to catch on. ...
... Several of the selected literature support that society has become digital and that most of the prison rehabilitation models and policies are still offline and do not cater for the digital realms. Over recent years research in this area agree that corrections rehabilitation models need to include digitalisation so that they prepare offenders for re-entry into a digitally driven society (Cullen, 2001;Gurusami, 2018;McDougall et al., 2017;McDougall & Pearson, 2020;McKay, 2022;Rantanen et al., 2021;Reisdorf & DeCook, 2018;Reisdorf & Jewkes, 2016;Taugerbeck et al., 2019;Toreld et al., 2018;Steene et al., 2017). ...
... All the scholars agree that society has become digital to the extent that most of the essential services such as social conduct, job seeking, health and business are now embedded in ICTs (Bedford et al., 2014;Järveläinen & Rantanen, 2020;Pamungkas, 2020;Rantanen et al., 2021;Seo et al., 2021;The Centre for Social Justice, 2021;Toreld et al., 2018;Willems et al., 2018b). Digital skills should be considered an exit prerequisite, especially when preparing the offender for release since offenders will need to use these digital services to find a job and housing, handle payments etc. Scholars agree that a rehabilitation model that does not align with societal changes is close to nothing (Kerr & Matthew, 2018;McDougall & Pearson, 2020;McDougall et al., 2017;Reisdorf et al., 2021). As a theory of punishment, rehabilitation refers to the process of an individual readapting to society or restoring an individual to a former position, rank or state (Reisdorf & DeCook, 2018). ...
Article
Full-text available
Digital transformation as a universal phenomenon has created a new reality in prison rehabilitation. Digitization promotes incarcerated people’s social skills, self-esteem, rehabilitation, and re-integration into society. Yet, many offenders are digitally incapacitated because they have never utilised digital technology or because they were denied access to any while incarcerated. The purpose of this study is to fill this gap. In this study, we systematically review the extant literature on prison rehabilitation to explain how the inclusion of digital rehabilitation of offenders makes re-entry successful and guarantees enhanced post-prison life in a digitalised society. To the best of our knowledge, this paper—which relies on the Good Lives Model and the Critical Theory Approach—is one of the first systematic literature reviews on digital prison rehabilitation. The Good Lives Model and the Critical Theory Approach are used in combination to investigate the three realms of digital prison rehabilitation: successful re-entry, post-prison life, and the digital society. We found that most of the prison rehabilitation practices and policies are still offline and do not cater for the digital realms. We conclude that the use of digital rehabilitation could facilitate a smooth transition back into society and ensure an improved post-prison quality of life in a technologically advanced society. By combining current digital technologies with augmented and virtual reality researchers in digital prison rehabilitation can create models that foster a new reality of prison rehabilitation.
... Providing inmates with access to digital technologies has the potential to change how inmates spend their time in prison, the extent to which they can access and engage in rehabilitative and educational programs, how official information is shared, and how requests or applications are processed and managed (e.g., Palmer et al., 2020). 3 The aim of self-service technologies is to increase autonomy for inmates and improve staff-inmate relationships, through reducing the administrative burden on staff and de-escalating frustrations for inmates (Krikorian & Coye, 2019;McDougall et al., 2017;Palmer et al., 2020). Giving inmates regular practice with digital devices while in prison has been argued to substantially increase their skills and confidence in dealing with digitally enabled services on release (Blomberg et al., 2021;McDougall et al., 2017;Palmer et al., 2020 • Entertainment (i.e., select games); ...
... 3 The aim of self-service technologies is to increase autonomy for inmates and improve staff-inmate relationships, through reducing the administrative burden on staff and de-escalating frustrations for inmates (Krikorian & Coye, 2019;McDougall et al., 2017;Palmer et al., 2020). Giving inmates regular practice with digital devices while in prison has been argued to substantially increase their skills and confidence in dealing with digitally enabled services on release (Blomberg et al., 2021;McDougall et al., 2017;Palmer et al., 2020 • Entertainment (i.e., select games); ...
... Previous research further identified that inmates who had served long sentences of 15-20 years or more were apprehensive about using digital technology due to a lack of skills and understanding about technology that had not been developed prior to them entering gaol (e.g., . A key goal of providing inmates with access to digital devices in prison is to increase their skills and confidence with technology that they will be required to use following their release (Blomberg et al., 2021;McDougall et al., 2017;Palmer et al., 2020). ...
Technical Report
Full-text available
Aims To develop an understanding of inmates' uptake of digital tablets following their implementation in two NSW correctional centres. The study examines the extent of inmates' tablet use, the features they are most likely to use, and their views of having access to tablets. The study also examines whether both the quantity and quality of inmates' tablet use are associated with their perceptions of social climate, wellbeing and autonomy. Methods A cross-sectional design was used whereby a self-report survey was administered to all inmates with tablet access in two NSW correctional centres. The sample included both male and female inmates (N = 208) who were advised of the survey via an email-style 'facility message' sent directly to the tablet, as well as communication via recruitment fliers posted around the centre. Results Inmates reported frequent use of the tablets and identified the phone calls feature as the most used feature. They felt access to tablets improved their ability to connect with family and friends, and their overall experience of life in gaol. Their experience of the impact of tablet access on their life in gaol was positively related to perceptions of the social climate, while their perception of tablets' impact on their ability to connect with family and friends was positively associated with general wellbeing and a sense of autonomy. Conclusion Findings from the current study indicate digital tablets were widely accepted by inmates. Despite the tablets being in the early stage of rollout with limited features, inmates regularly utilised many of those features. The study provides early indications that providing inmates access to such technology could have positive implications for their relationships with significant others, experience of life in prison, and psychological wellbeing. Additional features planned for the tablets are likely to further enhance inmates' engagement with the tablets and associated outcomes. 2
... With such good intentions of technology, the incarcerated individuals do not have to lose hope as they can still earn because of the technical skills they acquire while confined. McDougall et al. (2017) also researched the influence of technology in the incarcerated individuals, specifically on their behavior and reoffending. McDougall et al. (2017), opined that technology reforms prisons to assist offender rehabilitation. ...
... McDougall et al. (2017) also researched the influence of technology in the incarcerated individuals, specifically on their behavior and reoffending. McDougall et al. (2017), opined that technology reforms prisons to assist offender rehabilitation. They stated that the US government is moving from punishment-based policies to education-oriented policies. ...
... It was observed that the prisoners become motivated as they enjoy services utilized by their peers outside the jails and prisons contexts. With technology, the prisoners embrace self-directed rehabilitation, according to McDougall et al. (2017). There is an assurance of financial assistance, job training, and employment assistance with vast opportunities that tag along with technologies. ...
Article
Full-text available
This article the influence of technology on incarcerated students’ motivation and engagement in classroom-based learning. The juvenile jails and prisons confine many students who depend on education for future prosperity. In the twenty-first century, technology has dominated the education sector, and has been improving education delivery both to the incarcerated students and released students. Concerning the Covid-19 pandemic that resulted in the closure of learning setups, this work explicitly considers the jails and prisons contexts. The goal is to determine the effects of technology, after which the technical field can work towards improving the experience of the incarcerated students in the classrooms. Incarcerated students need modern skills that would enable them to survive in the technology-demanding society. This paper gives a brief review of previous research work, and my present work to determine the direction of classroom-based learning for incarcerated students.
... Video visitation has the added benefit of preventing one of the main ways that drugs and other contraband items enter the prisons (Christian et al., 2006). It has been suggested that some factors may include support, place to live, and jobs reflecting the social capital aspect (Christian et al., 2006), and that the interaction with the technology can affect dependency and lead to self-responsibility and personal control (McDougall et al., 2017). ...
... This is in accordance with previous research findings on better behavior from offenders in treatment with fewer disciplinary infractions in prisons and better outcomes after being released from prison (Crabbe, 2002). Two participants further agreed that video visitation improves self-responsibility and personal control (Mcdougall et al., 2017). Most importantly, all participants strongly felt that the maintenance of strong ties to the families and children is one of the key factors contributing to successful resocialization and post-penal acceptance. ...
Conference Paper
Full-text available
The article provides a review of the developments and characteristics of video visitation in PCI Sremska Mitrovica, with a focus on the contact between offenders and their families and children. The study draws upon data from fieldwork interviews with the prison employees and aims to understand experiences of using video visitation and their importance in the offender behavior as well as the prison system. The study gives voice to field experts with the knowledge and skills to suggest how video visitation impacts the offender behavior and its relevancy in the context of familial relationships. Moreover, it reflects on the nature of digital technologies in prisons and considers how they are embraced and managed in Serbia. Special focus is on the legislative framework in the Republic of Serbia.
... Ex-prisoners were regarded as having low level of technologically sophistication and being one of the most impoverished groups in digital age (Jewkes & Reisdorf, 2016;McDougall et al., 2017). A recent study of post-release technology experience of ex-prisoners highlighted that prisoners experienced "digital disconnection" where their digital skills were not refreshed during the imprisonment and would face substantial barriers to technology upon release (Davis & Ostini, 2019). ...
Article
Full-text available
Background The number of ex-prisoners worldwide has constantly been increasing in recent years. Currently, little is known about post-release daily adaptation, not to mention valid and reliable instruments for post-release daily routines pertinent to mental health. Objective This study aims to develop and validate a self-report instrument, hereafter referred to as Post Release Living Inventory for Ex-prisoners (PORLI-ex). Methods Three separate samples of ex-prisoners were recruited to complete an online survey (N=1,277, age range=17–89 years, 53.2% male, 72% white). Results The final model evidenced acceptable goodness-of-fit and consisted of 45 items on nine dimensions, which loaded on three second-order factors: Consolidation (three dimensions; e.g., Institutional Routines), Replacement (two dimensions; e.g., Maladaptive Behaviors), and Addition (four dimensions; e.g., Socializing with Ex-prisoner Friends) (α=.695–.915). Convergent validity was demonstrated in the positive correlations with IADL, SOLI, MLQ, GSE-6, and MSPSS. Discriminant validity was demonstrated in the weak correlations with the LEC-5 and perceived social and personal cost of punishment. Criterion-related validity was demonstrated in the correlations with psychiatric symptoms and crime-related outcomes and incremental validity in the correlations with these measures independent of the scores on IADL, SOLI, MLQ, GSE-6, and MSPSS. Conclusion This study calls for more resources on fostering psychological strengths and resilience through regularizing basic daily life experiences on top of traditional interventions for risk management among the ex-prisoners.
... The use of ICT in the prison context has long been the subject of studies and research (Champion & Edgar, 2013;Hughes, 2012;Pike & Adams, 2012;Pillera, 2017;2020;Suriano, 2011) that have stressed the importance of their integration into prison treatment and education activities (Torlone & Vryonides, 2016). According to the Report Review of European Prison Education Policy and the Council of Europe Recommendation (89)12 on Education in Prison (KING, 2019), politics must invest in technology and secure Internet services in prisons in order to ensure equity in the access to learning activities, and to promote digital skills -understood as one of the transversal competencies/transferable skills -essential skills in jobs and occupations such as communication or critical thinking, that can be transferred to other contexts (McDougall et al., 2017;Toreld et al., 2018). With this approach, in numerous initiatives (Taugerbeck et al., 2019), training in the so-called soft skills related to new forms of education and training through the use of ICT is provided in addition to digital literacy. ...
Article
Full-text available
The restrictions imposed by Covid-19 provided an opportunity, also in prison, to be confronted with technology. The use of digital technology, however, should not be a transitional solution, but rather it should be maintained and strengthened in the prison system as an integrated and permanent resource within the re-educational treatment and school activities. In the perspective of this consolidation, this research, through an analysis of the relationship between teachers' beliefs and attitudes towards Information Communication Technology (ITIS-Intrapersonal Technology Integration Scale) and the perception of knowledge domains (content, pedagogy and technology) involved in teaching and learning processes in which technology plays a substantial role (TPACK-Technological Pedagogical Content Knowledge), proposes an in-service training model that is an expression of that possible trans-processuality of schooling, education and creativity and that enhances the figure of the teacher in their role as an agent of change. Le restrizioni imposte dal Covid-19 sono state l'occasione, anche in carcere, per confrontarsi con la tecnologia. L'utilizzo del digitale non deve però costituire una soluzione transitoria, al contrario, la sua implementazione nel sistema penitenziario dovrà essere mantenuta e potenziata come risorsa integrata e permanente all'interno del trattamento rieducativo e delle attività scolastiche. Nella prospettiva di tale consolidamento, la presente ricerca, attraverso un'analisi della relazione tra teachers' beliefs and attitudes towards Information Communication Technology (ITIS-Intrapersonal Technology Integration Scale) e la percezione dei domini di conoscenza (contenuto, pedagogia e tecnologia) coinvolti nei processi di insegnamento e apprendimento in cui la tecnologia gioca un ruolo sostanziale (TPACK-Technological Pedagogical Content Knowledge), propone un modello formativo in-service che sia espressione di quella possibile trans-processualità di istruzione, educazione e creatività e che valorizzi la figura dell'insegnante nel suo ruolo di agente di cambiamento.
Chapter
This chapter explores the practice of strip searches in Spanish prisons. First, I briefly describe the regulation of body searches as security measures and the standards set to protect prisoners’ fundamental right to personal privacy, especially during strip searches. Second, I examine how body searches work in practice. Drawing on NPM annual reports, case-law, and previous research, it is shown that strip searches are used more extensively than they should and that certain standards, such as avoiding that the person is completely naked during the strip search, are frequently infringed. In addition, through an analysis of post-search reports from Catalan prisons, I assess if the use of strip searches is sufficiently justified and whether it proves to be effective in terms of security. Third, I go on to reflect on the consequences of strip searches for those involved: prisoners, visitors, and prison officers. Concretely, I use Goffman’s concept of the mortification of the self to discuss the impact that systematic searches and strip searches may have on inmates’ identity; I problematize the fact that visitors can also be subjected to strip searching and that the refusal to be strip-searched entails losing the right to communicate with the prisoner; and I consider how strip searching may influence staff-prisoner relationships. Finally, it is concluded that the enhanced security that strip searches provide is not proportional to the invasion of personal privacy they represent and the negative consequences they have.
Chapter
The one place that sits outside of the education marketplace is prisons. Characterised by dated technology, a lack of connectivity and poor staffing ratios, prisons are generally not renowned as places of educational innovation. When it does happen, it is all the more remarkable. This chapter describes an innovative educational intervention happening in a prison on the South Island of New Zealand. Learners work on cars, diagnose mechanical issues and deal with workplace safety issues all while improving their numeracy and literacy. What makes this learning unique is that it all takes place in a virtual workshop accessed through virtual reality technology. The cars may not be real, but the learning is. Education is becoming increasingly commodified and significant shifts are occurring in response to changes in the way people work and study. As people seek to upskill to increase their earning potential, qualifications are shifting entirely online to be completed when it is convenient to the learner. Simulations are helping to upskill generations of pilots, engineers and nurses. The need for flexibility in learning and an increasingly crowded marketplace is driving innovation in education, particularly in regard to educational technology. This innovation rarely makes it to the carceral environment, but when it does, the results can be life changing.
Article
Full-text available
p>Защита прав осужденных является сложной задачей, основные направления решения которой выработаны ООН. В статье приводятся результаты и перспективы использования положительного опыта тюремных систем Финляндии и России по цифровизации сферы предоставления услуг по реализации прав осужденных к лишению свободы (минимальных правил ООН).</p
Article
Full-text available
Mobile computing technology presents various possibilities and challenges for psychological assessment. Within forensic and correctional psychology, assessment of justice-involved persons facilitated by such technology has not been empirically examined. Accordingly, this randomized controlled experiment involved administering questionnaires about risk—needs, treatment readiness, and computerized technology opinions to a large (N = 212) and diverse sample of individuals under custodial correctional supervision using either a tablet computer or traditional paper-and-pencil materials. Results revealed that participants in the paper-and-pencil condition completed the packet of questionnaires faster but omitted items more frequently. Older participants and those with lower levels of education tended to take longer to complete the tablet-administrated measures. The tablet format was rated as more usable irrespective of demographic and personal characteristics, and most participants across the 2 conditions indicated that they would prefer to use computerized technology to complete psychological testing. Administration format did not have a clear effect on attitudes toward correctional rehabilitation services. Noteworthy for researchers is the substantial time saved and absence of practical problems with the tablet condition. Implications for practitioners include the general usability of the devices, their appeal to incarcerated persons, and the potential for tablets to facilitate clinical and administrative tasks with corrections clients. Considering the novel nature of this study, its promising results, and its limitations, future research in this area is warranted.
Article
Full-text available
Imprisonment is the most severe punishment in democratic societies except for capital punishment, which is used only in the United States. Crime prevention is its primary rationale. Imprisonment may affect reoffending in various ways. It may be reduced by some combination of rehabilitation and what criminologists call specific deterrence. Sound arguments can be made, however, for a criminogenic effect (e.g., due to antisocial prison experiences or to stigma endured upon release). Remarkably little is known about the effects of imprisonment on reoffending. The existing research is limited in size, in quality, in its insights into why a prison term might be criminogenic or preventative, and in its capacity to explain why imprisonment might have differential effects depending on offenders' personal and social characteristics. Compared with noncustodial sanctions, incarceration appears to have a null or mildly criminogenic effect on future criminal behavior. This conclusion is not sufficiently firm to guide policy generally, though it casts doubt on claims that imprisonment has strong specific deterrent effects. The evidence does provide a basis for outlining components of an agenda for substantive and policy relevant research.
Article
Reentry correctional facilities play a critical role in preparing inmates to successfully transition back to the community. Part of this role includes providing a structured program, which allows for gradual transition from prison life to the community through work, education, and counseling programs. Little research reveals how correctional officers (COs) maintain control and promote rule compliance within a reentry environment. Using administrative, survey, and ethnographic data, we examine how COs in a reentry-focused prison manage the inmate population. Correctional officers do not report using misconducts in surveys and observations, but administrative data reveal staff often use formal misconducts even for minor infractions. The number of accumulated misconducts an inmate received, seriousness of the current violation, and officer tenure significantly relate to the severity of present misconduct outcomes. Considering the mission and goals of reentry facilities, this study has significant implications for the reentry process and inmate experience.
Article
This article discusses the digital inequalities experienced by prisoners and the potential opportunities that providing ‘new’ media in prisons offer for offender rehabilitation and resettlement. Currently denied access to online and social media that most of us take for granted, and unable to communicate in ways that have become ‘ordinary’ in the wider community, it is argued that prisoners experience profound social isolation and constitute one of the most impoverished groups in the digital age. In prisons which provide selected prisoners some access to information and communication technologies, their high socio-cultural status and consequent construction as a ‘privilege’ frequently results in them being used in the exercise of ‘soft’ power by prison officer gatekeepers. Moreover, when prisoners come to the end of their sentences, they not only are faced with prejudice and poor job prospects due to their criminal record, but their digital exclusion during a period of incarceration may have compound effects and lead to long-term and deep social exclusion.
Article
California is currently implementing a prison downsizing experiment of historical significance, and as such provides a critical test case for prison downsizing in America. The California Public Safety Realignment Law (Assembly Bill [AB] 109) shifts responsibility from the state to the counties for tens of thousands of offenders. If it works, California — the nation’s largest state and home to nearly one out of every ten U.S. prisoners — will have shown the nation how to downsize prisons safely by transferring lower-level offenders from state prisons to county systems. If it does not work, counties will be overwhelmed with diverted inmates, unable to operate needed programs, which ultimately results in continued criminality and jail (instead of prison) crowding. In a very real sense, California’s Realignment experiment is giving the rest of the nation a close look at “the day after” significant decarceration.
Article
Longer prison sentences correlate with lower recidivism. However, it is difficult to disentangle individual effects related to aging and prior criminal histories from state practices and policies that impact the likelihood of reoffending. This study examines a large, administrative data set containing the criminal histories of prisoners released in 2005 to investigate this problem. We used Cox models with a shared frailty framework to account for the latent, state-level correlations that have a multiplicative effect on the individual likelihood of rearrest. These models reveal that the state effects account for considerable variation in rearrests, while also clarifying the roles of sentence length and disparate criminal histories on recidivism.
Article
In 2009 the Administrative Office of the US Courts developed and piloted a training program (STARR) for probation and pretrial officers. The purpose of this program was to train officers in the use of core correctional practices in their one-on-one interactions with offenders. Two areas of interest were subsequently investigated by researchers. First, did the training impact officer behaviors and second did trained officers supervise offenders that had lower failure rates. The evaluation of this effort was published in 2012 and used a 12-month follow-up for the measure of recidivism. The current study, a research note, extends the follow-up period for recidivism to 24 months. While there is some decline in the overall treatment effects it appears that STARR training is associated with a reduction in recidivism for moderate-risk offenders. Further, when coupled with training in MI, STARR seems to provide a promising reduction in recidivism with high-risk offenders. The limitations of the study and policy implications are discussed.