Rethinking What Works with Offenders

Stephen Farrall

• Publisher: Willan
• Year: 2013
• Print ISBN: 9781843921028
• eBook ISBN: 9781134028658
• Edition: 1
• Topic: Social Sciences
• Subtopic: Criminology
• Index: Social Sciences

This important and original new book reports on a major investigation of the outcomes of probation supervision, is concerned with the key question of what works in probation, and comes at an important moment of change and development for the probation service in the UK. Unlike previous studies which have relied mostly on official data, this book makes use of over 200 interviews with men and women on probation, and their supervising Probation Officers. Rethinking What Works with Offenders has the following objectives: to understand probation work from the perspectives of those who deliver it and those to whom it is delivered to study probation intervention as a whole (in particular the probation order) rather than specific aspects to locate probation work in the wider social contexts of those on probation to analyse how probation works, and to reconceptualise probation outcomes in terms of degrees of success rather than as 'successful' or 'unsuccessful' to assess the policy implications of these conclusions This book presents an important and challenging range of findings on 'what works' in probation and with offenders, and will be essential reading for anybody professionally concerned with the present and future of probation. raises central issues at a critical time for the reorganised National Probation Servicebased on extensive research, including 200+ interviewsessential reading for anybody interested in 'what works' in probation

Part 1
Introduction

Chapter 1
Probation, social context and desistance from crime: introducing the agenda

The purpose of this book is to explore and throw further light on the processes that occur during probation supervision which are either conducive to desistance or which contribute to further offending. Whilst it is true that the correlates of recidivism and desistance are well known, the mechanisms by which these correlates are produced and the role that criminal justice sanctions play in this remain less well understood. In order to understand better the role of probation supervision in encouraging the processes associated with desistance, the criminal careers and behaviour of 199 people made subject to probation and combination orders¹ were examined in detail. This book reports on that four-year examination.

Many of those who embark on offending careers would appear to stop at around the same time that they gain stable employment, embark upon stable life-partnerships and disengage from a delinquent peer group. On the other hand, for those convicted and sentenced, their age and gender are better predictors of their likelihood of a further conviction than the type of sentence they received. Indeed, no study has yet demonstrated conclusively that sentences have markedly different outcomes from one another in terms of rates of reconviction. There are, however, numerous disadvantages with the research methodologies employed to investigate desistance and recidivism. Studies of sentence outcomes have generally been based on official statistics, have not considered in depth the nature of the interventions undertaken, have tended to be retrospective and have not collected the views and experiences of those made subject to such disposals. Thus these studies suffer from what Shadd Maruna (2000a: 12, emphasis in original) has referred to as the ‘black box’ syndrome – the outcomes are known, but the precise sequence of events and processes involved in their production has been left largely unexamined: ‘by concentrating almost exclusively on the question of "what works", offender rehabilitation research has largely ignored questions of how rehabilitation works, why it works with some clients or why it fails with others.’

This chapter provides the background to a project which has attempted to open this ‘black box’ so as to investigate the role that probation supervision, individual motivation and wider social and personal circumstances play in helping some people to stop offending. Unlike previous investigations of the impact of probation supervision, it has aimed to bring to the evaluation some of the techniques and styles of research developed by those investigating the progression of criminal careers. Thus, the research has departed from the conventional sources of data, types of variables collected and styles of analyses. It has

• not therefore relied predominantly on data derived from official sources but has collected data directly from probationers and probation officers by way of face-to-face interviews;
• examined the impact of changes in the social and personal circumstances of the probationer on their offending career; and
• employed a prospective longitudinal methodology.

Its origins, therefore, lay in two key areas of criminological research: the evaluation of criminal justice interventions and the study of the course of criminal careers, their persistence and desistance. Both these paradigms offer insights and set the agenda for many of the issues investigated, but a review of each suggests that there are areas of both where current knowledge is deficient. By addressing these deficiencies, the project throws new light on the impact of probation supervision on probationers’ lives and on the factors conducive to their desistance from crime.

The background

Some of the most commonly observed regularities in criminology concern rates of offending over the course of a person’s life. For example, by the time that they had reached their thirtieth birthday, approximately a third of the males in England and Wales born in 1953 had received a conviction for a non-motoring offence (Home Office 1995). However, only around 7 per cent had received six or more convictions and most of those convicted had an officially recognised criminal career which spanned no more than four years (Home Office 1995: Table 8). The peak age of conviction² for males in England and Wales is currently around 18 years of age (Barclay 1990; Newburn 1997; Flood-Page et al. 2000). This broad picture, based upon official statistics, has been supported by numerous other studies (e.g. Farrington 1992a) and by researchers collecting data in different countries (e.g. Blumstein et al. 1988) and during different historical eras (Gottfredson and Hirschi 1990). Taken together, the evidence suggests that whilst many people are caught offending at least once, few go on to become persistent offenders. Such is the uniformity of the relationship between age and crime that it has become one of the ‘laws’ of criminology, and is commonly referred to as the ‘age–crime curve’. For males, the line which describes the ‘age–crime curve’ starts around 10 years of age (until that age, for UK males, those under 10 are below the age of criminal responsibility). The line then starts to climb rapidly until it reaches a peak around the ages of 17–19. The curve then starts to decline – steeply at first, but becoming less steep as the years progress. For females the ‘curve’ is different. First of all, far fewer females would appear to embark upon offending careers. However, following a climb almost as steep as the male curve, the female ‘curve’ quickly flattens to something approaching a plateau between 14 and 18, before declining gradually.

Given the salience of this issue for policy-makers and its role as a source of much debate in criminology, it is little wonder that the 1980s and 1990s have witnessed a continuation of the efforts of criminologists to chart the extent, contours and correlates of patterns of offending with greater sophistication and in greater detail. In particular, interest in the reasons why some people stop offending and others persist has increased rapidly since the early 1980s. This same period has also witnessed a renewed interest in the investigation of the outcomes of various court disposals – both community disposals and custodial sentences (see Lloyd et al. 1994; McGuire 1995). For example, Lloyd et al. (1994) reported that whilst 43 per cent of probationers were reconvicted within two years of the start of their orders, the proportion of those reconvicted after community service orders was even higher (49 per cent), and higher still for those given probation orders with conditions (63 per cent).

These two bodies of work represent closely related, but nevertheless subtly different paradigms. One – referred to here as the criminal career paradigm – has charted the incidence and patterns of offending across the life spans of numerous cohorts of members of the general population. Such cohorts are followed from a young age for several years, often decades, and the data collected are usually based on interviews with cohort members or others involved in their lives (such as parents, teachers, social workers or peers). Official data relating to offending are frequently used to supplement and validate such reports. Although this paradigm has relied on quantitative research methods, it has not precluded the use of qualitative data sets.

The second paradigm – measuring the effectiveness of attempts by the criminal justice system to reduce such behaviour – has followed up those persons found guilty and sentenced to some form of punishment. These follow-ups often last no longer than two years, although in some instances follow-up periods of five or seven years have been recorded. This paradigm has relied – to the virtual exclusion of all other methodologies – on official records which have been analysed quantitatively (e.g. Lloyd et al. 1994).

The criminal career paradigm

Research by criminologists such as Sheldon and Eleanor Glueck, Alfred Blumstein, Marvin Wolfgang and Thorsten Sellin in the USA and David Farrington in the UK has suggested that whether or not an individual participates in offending is closely associated with his or her age and a number of social and psychological variables. For example, Farrington (1992b: 129) listed some of the variables found to be most strongly related to offending in the Cambridge Study in Delinquent Development – a follow-up study of over four hundred boys originally living in London. Included were ‘problematic’ behaviours between 8 and 14-years-old (e.g. bullying, lying and aggressiveness), teenage anti-social behaviours (e.g. heavy alcohol, tobacco use, gambling and frequent sexual activity), impulsiveness, school problems (e.g. low school attainment, frequent truancy and failure to take examinations), family factors (e.g. poor child rearing and supervision, and poor relationship with parents), anti-social factors (e.g. convictions of other family members and friends) and socioeconomic factors (low family income, poor housing and poor employment record). Several of these influences have also been found by other researchers to be associated with the onset of offending behaviours (see, for example, Graham and Bowling 1995: 33–43; also Wolfgang et al. 1972; Sampson and Laub 1993; Elliott and Menard 1996). The data employed in studies of this nature have usually been derived from repeated interviews with members of cohorts, often followed from school age to adulthood. During these interviews, self-reported data on offending and various other topics have been collected from the cohort members. As noted, these data are often supplemented by official records and further interviews with partners, peers and teachers in order to gain these perspectives as well as to validate responses.³

A number of researchers have focused their attention specifically upon the later stages of offending careers, and in particular on the factors conducive to desistance from offending. Drawing on the insights of earlier work on criminal careers, they have pointed to a number of correlates of desistance. The literature on desistance is commonly based on criminal career data sets (e.g. Knight and West 1975; Loeber et al. 1991; Sampson and Laub 1993), or on one-off retrospective research (e.g. Shover 1983; Cusson and Pinsonneault 1986; Graham and Bowling 1995). In a few cases (e.g. Burnett 1992; Leibrich 1993) data have been collected by following the careers of persons after they have been made subject to criminal justice interventions.⁴ The research undertaken so far has shed light on the role of social and personal factors in desistance (see Adams 1997; Farrall 2000; Laub and Sampson 2001, for outlines of this body of work).

The factors associated with desistance

A number of researchers (e.g. Uggen and Kruttschnitt 1998: 356) have provided evidence that desistance is associated with gaining employment, although the precise causal links between engaging in legitimate employment and desistance from offending have yet to be satisfactorily established. Mischkowitz (1994: 313) reported that ‘erratic work patterns were substituted by more stable and reliable behaviour’ amongst his sample of desisters. Meisenhelder (1977) noted that the acquisition of a good job provided the men in his sample with important social and economic resources, whilst Shover (1983: 214) reported how a job generated ‘. . . a pattern of routine activities – a daily agenda – which conflicted with and left little time for the daily activities associated with crime’.

Similar sentiments were expressed by Sampson and Laub (1993: 220–22) when they wrote that desisters were characterised as having ‘. . . good work habits and were frequently described as "hard workers" ’. Farrington et al. (1986: 351), reporting on the Cambridge Study in Delinquent Development, wrote that ‘proportionally more crimes were committed by . . . youths during periods of unemployment than during periods of employment’,afinding supported by the later work of Horney et al. (1995). However, as researchers such as Ditton (1977) and Henry (1978) have shown, full-time employment does not preclude either the opportunities to offend nor actual offending. Graham and Bowling (1995: 56, Table 5.2) found that for young males employment was not related to desistance, as did Rand (1987) when she investigated the impact of vocational training on criminal careers.

Another of the most common findings in the literature on desistance is that individuals cease to offend at about the same time that they start to form significant life-partnerships (e.g. Chylicki 1992). One of the clearest statements in support of this line of reasoning came from Shover (1983: 213), who wrote that ‘The establishment of a mutually satisfying relationship with a woman was a common pattern [and] an important factor in the transformation of their career line’. Cusson and Pinsonneault (1986: 79–80) and Mischkowitz (1994: 319) followed West (1982: 101–104) in arguing that what was important (in terms of facilitating desistance) was not marriage per se, but rather the quality of the relationship and the offending career of the person whom the would-be desister married. The recent work of Laub et al. (1998) supports this contention – as marriages became stronger amongst the men in their sample, so these men’s offending began to be curtailed. A number of studies have suggested that the experience of becoming a parent is also associated with desistance from offending (see, for example, Irwin 1970: 203; Parker 1976: 41; Trasler 1979: 315; Caddle 1991: 37; Leibrich 1993: 59; Sampson and Laub 1993: 218; Hughes 1997, 1998: 146; Uggen and Kruttschnitt 1998: 355; Jamieson et al. 1999: 130).

Despite the evidence suggesting that forming a life-partnership may result in desistance, some researchers have questioned this rather simple cause-and-effect model.⁵ Rand (1987: 137) tested the hypothesis that ‘. . . young men who marry are less criminal than th...

Table of contents

• Cover Page
• Half Title page
• Title Page
• Copyright Page
• Contents
• List of tables and figures
• Acknowledgements
• Foreword
• Part 1 Introduction
• Part 2 Probation, Motivation and Social Contexts
• Part 3 Persistence and Desistance
• Part 4 Conclusions
• Appendix
• References
• Index
• Case index