F O S T E R , C E D R I C 1 6 9 2 T S
1 Corrections in the Community.
© 2011, Elsevier Inc. All rights reserved.
C h a p t e r 1
It is hard to identify the benefits inmates gain from prison, but the harm done there is readily seen. If you want to increase the crime problem, incite men to greater evil, and intensify criminal inclinations and proclivities, then lock violators up in prison for long periods, reduce their outside contacts, stigmatize them and block their lawful employment when released, all the while setting them at tutelage under the direction of more skilled and predatory criminals. I know of no better way to gain your ends than these.
—Harry E. Allen
Crime is everywhere, in all nations great and small and, in this nation. In the United States, crime is a violation of criminal statutes passed by elected repre- sentatives. These statutes are enforced by a variety of social control agencies, including law enforcement, prosecution, court, and postadjudication compo- nents (e.g., prisons, probation, and parole). These varied agencies and actions, along with their philosophical bases and objectives, are usually called the “criminal justice system.”
No one imposed this specific set of agencies on the nation. We invented them ourselves and, if there is something amiss with an agency or mission, it can be changed. One fact about the American criminal justice system is that it
The Criminal Justice System
corrections
community corrections
criminal justice system
incarceration
intermediate sanctions
parole
prison
probation
Key terms
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 1 : the Criminal Justice system2
is rapidly evolving and changing as a result of the volume of crime, emerg- ing national priorities, available funding, and changing political ideologies. Behaviors deemed particularly heinous in one epoch may become regulated, if not accepted, behavior in another. For example, the “Great Experiment” of prohibition attempted to protect our national character and youth, increase productivity, lessen collateral problems of idleness, and improve the moral fiber of those using alcohol. It was later abandoned as a national crusade; earlier twentieth century law enforcement efforts instead lapsed into strate- gies to regulate alcohol as a controlled substance, concerned only in large part with keeping alcohol out of the hands of youthful consumers and col- lecting taxes.
One component of the criminal justice system is corrections, defined ear- lier as “postadjudication processing of convicted criminal offenders.” This definition, if it were ever adequate, probably best fits the correctional scene of the early twentieth century, when the major sentencing options available to sentencing courts were committing the offender to prison or granting
probation. In fact, the study of postadjudication processing of criminal law offenders was, until about 1969, commonly referred to as “penol- ogy.” As shown in subsequent chapters, postadju- dication has become much more complex in the United States.
The field of corrections, like most of the justice sys- tem, has undergone rapid change in the past three decades. Programs have been developed to allow prosecutors to suspend prosecution of alleged crim- inals, provided they became and remained actively involved in seeking personal development and rehabilitation under the “deferred prosecution” program. Pretrial detention of accused law viola- tors is now rare due to the development of personal recognizance programs that reduced the impor- tance of bondsmen in the pretrial portion of the system. In addition, the tools of technology have grown greatly in the past two decades, expanding probation supervision into conventional proba- tion, intensive supervised probation, house arrest (with or without electronic monitoring or global position tracking), community service, day atten- dance centers, and restitution programs. There are even probation variations that combine serving a sentence in jail before probation begins, and several
Booking in local jail. Credit: Beth Sanders
F O S T E R , C E D R I C 1 6 9 2 T S
3Corrections in the Community
probation programs that require a period of imprisonment prior to return to the community under probation supervision. These latter programs, inciden- tally, are part of the “intermediate sanctions” that have emerged in the past 25 years: offender control programs that fall somewhere between probation and imprisonment.
What has corrections become? How can we best define it at the present time? For us, corrections is the social control agency that provides societal protection through incarceration and community supervision and rehabilitation services to persons accused or convicted of criminal law violating behavior. This defi- nition includes restorative justice and pretrial diversion programs, as well as the more traditional probation and parole services. It also embraces interme- diate sanctions and alternative early release programs for inmates in prisons across the nation. In sum, corrections involves social control of persons whose behavior has brought them to the attention of the justice system. The missions, objectives, procedures, and even principles that govern our definition of cor- rections are continually changing. In this book, we hope to describe the recent developments and emerging dimensions that define community corrections today.
CorreCtions in the Community This book describes and explains corrections in the community, or “com- munity corrections.” This term refers to numerous and diverse types of supervision, treatment, reintegration, control, restoration, and support- ive programs for criminal law violators. Community corrections programs, as shown later, are designed for offenders at many levels of both juvenile and criminal justice systems. First, community corrections programs are found in the preadjudication level of the justice systems and include diver- sion and pretrial release programs, as well as treatment programs provided by private sector agencies, particularly for juveniles (Allen, Latessa, Ponder, & Simonsen, 2007; Maloney, Bazemore, & Hudson, 2001; Shaffer, Listwan, & Latessa, 2001; Travis & Petersilia, 2001).
As correctional clients enter the justice system, community corrections pro- grams have been developed and designed to minimize their further processing and placement into more secure settings. These preimprisonment programs include restitution, community services, active probation, intensive supervised probation, house arrest, and residential community facilities, such as halfway houses (please note that all of these programs are described in detail in later chapters). One assumption underlying this effort to minimize offender pen- etration into the justice system is that community corrections is more effective at reducing future crime and more cost-efficient. Community corrections is cer- tainly no less effective in reducing recidivism than is prison, and there is strong
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 1 : the Criminal Justice system4
evidence that community correctional programs, if administered properly, can significantly reduce recidivism.1
Another assumption is that community corrections is more humane, although there is some contemporary debate over whether corrections should be humane rather than harsh.
Community corrections continues after incarceration (and in some cases is com- bined with incarceration)2 and among the many programs found at this level are split sentences (jail followed by probation), shock incarceration and shock probation, prison furlough programs, work and educational release, shock parole, and parole programs and services.3 The various points at which com- munity corrections programs have been developed are suggested in Figure 1.1, which identifies the flow of clients into and through the justice system.
The diagram of Figure 1.1 first appeared in President Lyndon Johnson’s Crime Commission report, The Challenge of Crime in a Free Society (1969). It outlined the basic sequence of events in the criminal justice process. Police, courts, and corrections were thus viewed as elements that were interrelated and interdepen- dent. The idea was to demonstrate the manner in which successful crime preven- tion was the goal of the entire system. Community corrections fits squarely into this goal: offenders whose criminal behavior is reduced or eliminated through programs in the community will commit fewer if any crimes in the future.4
Two major factors should be pointed out in Figure 1.1. First, the major ways out of the system are probation and parole, shown here as system outputs. The second conclusion is that the number of cases flowing through the system decreases as offenders are processed at the various decision points (prosecu- tor, court, sentencing, and release from prison). Figure 1.2 depicts the flow of offenders through the system for those arrested for Type 1 felony crimes in 2003.
1An extensive body of research has demonstrated that community correctional programs can have a substantial effect on recidivism provided certain empirically derived principles are met. For a summary of this research see Latessa and Lowenkamp (2007). 2For example, in Ohio, the state funds “community-based correctional facilities.” These facilities are operated by local community corrections boards, are designed to provide treatment, and often utilize local community services. They are, however, secure facilities. For descriptions of the Florida circumstances, see Lucken (1997). For a California example, see Wexler, Melnick, Lowe, and Hiller (1999). 3Some would argue that many so-called “community” correctional programs are essentially institutional correctional facilities because they are state run. However, we believe that state-operated programs can indeed be considered community correctional programs, provided they include some type of supervision in the community. For a different perspective on this issue, see Duffee (1990). See also Burke (1997) and Gendreau, Goggin, and P. Smith (2000). 4See Lowenkamp, Latessa, and Holsinger (2006), and Wilson, Gallagher, and MacKenzie (2000).
F O S T E R , C E D R I C 1 6 9 2 T S
5 C
o rre
c tio
n s in
th e
C o
m m
u n
ity
Unsolved or Not Arrested
Released without Prosecution
Released without Prosecution
Charges Dropped or Dismissed
Charges Dropped or Dismissed
Entry into the System Prosecution and Pretrial Services Adjudication Sentencing and Corrections
Crimes Observed by the Police
Crime
Crimes Reported to the Police
Investigation Arrest Booking Preliminary
Hearing
Initial Appearance Bail or
Detention
Note: This chart gives a simplified view of caseflow through the criminal justice system. Procedures vary among jurisdictions. The weights of the lines are not intended to show the actual size of caseloads.
Juvenile Offenses
Nonpolice Referrals
Police Juvenile Unit
Release or Station Adjustment
Petty Offenses
Misdemeanors
Released
Intake Hearing Petition to Court
Information Arraignment
Felonies
Felonies Grand Jury
Refusal to Indict
Information
Arraignment
Charge Dismissed Acquitted
Trial
Guilty Plea
Charge Dismissed
Reduction of Charge
Acquitted
Trial
Guilty Plea
Adjudicatory Hearing
Released
Disposition
Sentencing
Appeal
Sentencing
Probation
Fine
Nonpayment
Probation
Revocation
Jail
Revocation
Juvenile Institution
Parole
Revocation
Out of System
Out of System
Habeas Corpus
Revocation
Parole
Out of System
Pardon and Clemency
Capital Punishment
Probation
Revocation
Penitentiary
Nonadjudicatory Disposition
FIGURE 1.1 What is the sequence of events in the criminal justice system? Source: Adapted from President’s Commission on Law Enforcement and Administration of Justice (1969).
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 1 : the Criminal Justice system6
The percentage of offenders in each major correctional sanction can be found in Figure 1.3. Nonincarceration sentences were imposed for nearly 60% of offenders in 2009. Another 11.2% that were sentenced were released from
prison onto parole supervision. Together this rep- resents nearly five million offenders. Even a large part of those offenders sentenced to jail may be released onto probation as part of a split-sentence. It should be obvious that community corrections handles a large proportion of the offenders in the nation. For example, the Bureau of Justice Statistics (2006) reported that one in every 32 adult resi- dents of the nation was under correctional control at the start of 2006. On the basis of 100,000 adult residents in the nation, 1884 were on probation, 347 on parole, 738 in prison, and 252 in jail (see Table 1.1). More than two of three offenders were living in the community on a given day in 2005.
2,320,900 Arrested 1,624,630 Adults Arrested
1,174,607 Adults Prosecuted
518,067 Adult Felony Complaints
696,270 Juveniles
451,049 Dismissed
24,144 Not Guilty
Total “Out:” 475,193 Adults 696,270 Juveniles
205,490 Pled Guilty to Misdemeanors
461,128 Pled Guilty
32,795 Guilty at Trial
Total “In” or Punished: 699,413 Adults
FIGURE 1.2 Outcomes for arrest for felony crime: 2003. Sources: Federal Bureau of Investigation (2004). Adapted from Silberman (1978).
Probation is a sentence imposed by the court that does not usually involve confinement and imposes conditions to restrain the offender’s actions in the community. The court retains authority to modify the conditions of the sentence or to resentence the offender if he or she violates the conditions. Parole is the release of an offender from confinement prior to expiration of sentence on condi- tion of good behavior and supervision in the community.
BoX 1.1 proBation and parole
FIGURE 1.3 Correctional populations in the United States, 2008.
Correctional populations in the United States, 2008
Probation 57.7%
Jail 10.6%
Parole 11.2%
Prison 20.5%
F O S T E R , C E D R I C 1 6 9 2 T S
7Probation in America
A list of the incarceration rate for each state is shown in Table 1.2. Louisiana leads the nation with a rate of 853 per 100,000 and Maine has the lowest with 151.
proBation in ameriCa Nearly 60% of the adults under correctional care or custody are on proba- tion, the largest single segment of the community correctional system. As shown in Table 1.3, Texas had the largest number of its citizens on probation, but Massachusetts had the highest rate: 3620 per 100,000 adult residents. Six other states each had a rate of more than 2500. The lowest state rate was New Hampshire (443 per 100,000).
table 1.1 Number of Adults Under Correctional Supervision in 2008 and per 100,000 Residents
Total Rate per 100,000
Probation 4,270,917 1845 Parole 828,169 358 Prison 1,518,559 504 Jail 785,556 258 Total correctional population 7,403,201 3150
Source: Glase and Bonsczar (2009).
A state or federal confinement facility having custodial authority over criminal-law violating adults sentenced to confinement for usually more than 1 year.
BoX 1.3 prison
A jail is a confinement facility, usually administered by a local law enforcement agency, intended for adults but sometimes containing juveniles, that holds persons detained pending adjudication and/or persons committed after adjudication for sentences of 1 year or less. Jails are usually supported by local tax revenues and, as such, are particularly vulnerable to resource reductions. Additional categories of jail inmates include mentally ill persons for whom there are no other facilities or who are awaiting transfer to mental health authorities, parolees and probationers awaiting hearings, court-detained witnesses and persons charged with contempt of court, fed- eral prisoners awaiting pick up by marshals, and offenders sentenced to state departments of corrections for whom there is not yet space but who cannot be released (“holdbacks”).
BoX 1.2 Jail
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 1 : the Criminal Justice system8
In all, it is clear that a great number of convicted persons are now being placed on probation. In most cases, probation agencies monitor the offender’s compliance with the conditions of probation release (restitution, community service, payment of fines, house arrest, drug/alcohol rehabilitation, etc.). The crucial roles that pro- bation plays in community corrections and the justice system become even more apparent when institutional and parole population figures are examined.
the u.s. prison population Because the rate of parole in a given state is affected by the size of the prison population, it is necessary to examine the size of the U.S. prison population before considering parole figures. A census of state and federal corrections institutions is conducted at midyear and year-end by the Bureau of Justice Statistics (Harrison and Beck, 2006). At midyear 2009, the number of people
table 1.2 Ranking of States by Prison Incarceration Rates, 2008 (Inmates per 100,000 Residents)
1. Louisiana 853 2. Mississippi 735 3. Oklahoma 661 4. Texas 639 5. Alabama 634 6. Arizona 567 7. Florida 557 8. Georgia 540 9. South Carolina 519 10. Arkansas 511 11. Missouri 509 12. Kentucky 492 13. Virginia 489 14. Michigan 488 15. Nevada 486 16. Idaho 474 17. California 467 18. Colorado 467 19. Delaware 463 20. Ohio 449 21. Indiana 442 22. Tennessee 436 23. Alaska 430 24. South Dakota 412 25. Connecticut 407 26. Maryland 403
27. Pennsylvania 393 28. Wyoming 387 29. Wisconsin 374 30. Oregon 371 31. Montana 368 32. North Carolina 368 33. Illinois 351 34. Hawaii 332 35. West Virginia 331 36. New Mexico 316 37. New York 307 38. Kansas 303 39. New Jersey 298 40. Iowa 291 41. Washington 272 42. Vermont 260 43. Nebraska 247 44. Rhode Island 240 45. Utah 232 46. North Dakota 225 47. New Hampshire 220 48. Massachusetts 218 49. Minnesota 179 50. Maine 151 51. Federal System 60
U.S. Total: 504
Source: Sabol, West, and Cooper (2009).
F O S T E R , C E D R I C 1 6 9 2 T S
9The U.S. Prison Population
incarcerated in prison was 1,617,478, an all-time high (West, 2010). These fig- ures are even more dramatic when you consider that an estimated 8% of black males in their late twenties were in prison (West, 2010). See Figure 1.4 for the number of persons under correctional supervision in 2000 and 2008.
These figures are important to the parole rates in part because they represent the source of clients for the parole system. Prisoners enter the parole system by a parole board decision or by fulfilling the condition of mandatory release. Typically, at some time between their minimum and maximum sentences,
table 1.3 Community Corrections among the States, End of Year, 2008
10 states with the largest 2008 community corrections populations
Number supervised
10 states with the highest rates of supervision
Persons supervised per 100,000 adult U.S. residents*
10 states with the lowest rates of supervision
Persons supervised per 100,000 adult U.S. residents*
Probation
Texas 427,080 Massachusetts 3620 New Hampshire 443 California 325,069 Rhode Island 3251 West Virginia 579 Florida 279,760 Minnesota 3202 Utah 582 Ohio 260,962 Ohio 2973 Nevada 684 Pennsylvania 186,973 Indiana 2727 Maine 718 Massachusetts 184,308 Delaware 2563 Kansas 770 Michigan 175,591 Texas 2401 New York 789 Illinois 144,904 Colorado 2358 North Dakota 845 Indiana 131,291 Michigan 2304 Virginia 896 New Jersey 128,737 Washington 2240 Iowa 998
Parole
California 120,753 Arkansas 920 Maine 3 Texas 102,921 Oregon 754 Florida 31 Pennsylvania 72,951 Pennsylvania 751 North Carolina 48 New York 52,225 Louisiana 742 South Carolina 57 Illinois 33,683 Texas 579 Massachusetts 63 Louisiana 24,636 Missouri 459 Nebraska 63 Georgia 23,448 South Dakota 446 Rhode Island 63 Michigan 22,523 California 438 Virginia 75 Oregon 22,195 Wisconsin 418 North Dakota 77 Missouri 20,683 Kentucky 375 Delaware 82
Note: This table excludes the District of Columbia, a wholly urban jurisdiction; Georgia probation counts, which included probation case-based counts for private agencies; and Idaho probation counts in which estimates for misdemeanors were based on admissions. Source: Glaze and Bonczar (2009). *Rates are computed using the U.S. adult resident population on January 1, 2009.
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 1 : the Criminal Justice system10
inmates are released from prison and placed on parole. Mandatory releasees enter parole supervision automatically at the expiration of their maximum terms (minus sentence reductions for time credit accumulated for good time, jail time, and other “gain” procedures). Traditionally, this has been the manner in which a parole system operated under the indeterminate sentencing model presently in force in one-half of the states. The “abandon parole” movement began in 1976, and a number of states have changed their statutes to remove the authority of the parole board to release offenders before the expiration of their sentences. This issue is discussed in more detail in the chapters that follow.
parole in ameriCa Adults on parole at the beginning of 2008 are found in Table 1.1 and totaled 828,169, the highest number ever on parole. Table 1.3 shows that the parole rate ranged from a high of 920 in Arkansas to a low of three in Maine. Maine abolished parole in the late 1970s, which explains the low rate; only those sen- tenced to parole before 1976 continue to be supervised.
In sum, parole statistics reveal the relationship between the size of the prison pop- ulation and the number of parolees. These figures indicate that both prison and parole populations increased dramatically from 1995 through 2008 (see Figure 1.4). Changes in sentencing options and sentence length have also meant that prisoners were actually serving longer sentences in 2008 than they were in 1995.
summary This brief consideration of statistics from major components of the correc- tional system (probation, prisons, and parole) demonstrates their crucial linkage within the criminal justice system. Imagine what would happen if
FIGURE 1.4 Correctional populations in the United States: 1995, 2000, and 2008.
Probation
Jail
Prison
Parole
0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000
1995 2000 2008
F O S T E R , C E D R I C 1 6 9 2 T S
11Review Questions
probation and parole were abolished completely and all convicted persons were required to serve their full prison terms; if this had happened in 2004, the prison population could have been nearly seven million! Naturally, the prison system is not equipped to handle such a large number of inmates, nor would it be good social policy to attempt such a foolish venture.
We do not wish to suggest that all offenders could and should be released to community corrections. At least 15–25% of the prison population are too dan- gerous or pose too great a threat to community safety to allow their immediate release, even onto “intensive supervised parole” (Allen et al., 2007).
It is the function of probation and its many variants (the so-called “interme- diate punishments”), as well as parole, to determine how the population of convicted persons can be managed in a fashion consistent with not only the capacity of the prison population but also the goals of societal protections and offender rehabilitation and reintegration.
In short, the examination of corrections in the community is the theme of this text. We consider such key issues as what are the best methods for classifying and supervising offenders? What background, education, and training should various community corrections agents possess? How effective are community corrections programs in terms of public safety? And at what cost? What are the recent innovations in community corrections and intermediate punishments? How effective are these compared to incarceration? The consideration of these (and other) issues will provide readers with the opportunity to form their own opinions and ideas concerning the proper use of community correctional programs and how to coordinate these in the criminal justice system.
review Questions 1. What is corrections in the community? 2. What is meant by the funnel effect and how does it occur? 3. If probation and parole were abolished completely, what effect would this
have on the prison system? 4. Develop an argument for increased use of community corrections. 5. How are offenders generally released from prison? 6. Describe the current distribution of offenders across the main
components of the criminal justice system.
reCommended readings Allen, H. E., Latessa, E. L., & Ponder, B. (2010). Corrections in America: An introduction. (12th ed.).
Upper Saddle River, NJ: Pearson/Prentice Hall.
Latessa, E. J., & Holsinger, A. (2006). Correctional contexts: Contemporary and classical readings. Los Angeles: Roxbury.
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 1 : the Criminal Justice system12
referenCes Allen, H. E., Latessa, E. L., Ponder, B., & Simonsen, C. E. (2007). Corrections in America. Englewood
Cliffs, NJ: Prentice-Hall.
Allen, H. E., Latessa, E. L., & Ponder, B. (2010). Corrections in America: An introduction (12th ed.). Upper Saddle River, NJ: Pearson/Prentice Hall.
Burke, P. B. (1997). Policy driven responses to probation and parole violations. Washington, DC: U.S. Department of Justice.
Duffee, D. E. (1990). Community characteristics: The presumed characteristics and argument for a new approach. In D. E. Duffee & E. F. McGarrell (Eds.), Community corrections: A community field approach. Cincinnati: Anderson.
Federal Bureau of Investigation. (2004). Crime in the United States. 2003. Washington, DC: U.S. Department of Justice.
Gendreau, P., Goggin, C., & Smith, P. (2000). Generating rational correctional policies. Corrections Management Quarterly, 4(2), 52–60.
Glaze, L. E., & Bonsczar, T. (2009). Probation and parole in the United States, 2009. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
Glaze, L. E., & Palla, S. (2005). Probation and parole in the United States, 2004. Washington, DC: U.S. Bureau of Justice Statistics.
Harrison, P., & Beck, A. (2006). Prison and jail inmates in midyear 2005. Washington, DC: Bureau of Justice Statistics.
Latessa, E. J., & Lowenkamp, C. (2007). What works in reducing recidivism. St. Thomas Law Journal, 3, 521–535.
Lowenkamp, C. T., Latessa, E. J., & Holsinger, A. (2006). The risk principle in action: What we have learned from 13,676 offenders and 97 correctional programs. Crime & Delinquency, 51(1), 1–17.
Lucken, K. (1997). The dynamics of penal reform. Crime, Law and Social Change, 26(4), 367–384.
Maloney, D., Bazemore, G., & Hudson, J. (2001). The end of probation and the beginning of commu- nity justice. Perspectives (Gerontological Nursing Association (Canada)), 25(3), 22–31.
President’s Commission on Law Enforcement and Administration of Justice. (1969). The challenge of crime in a free society. Washington, DC: U.S. Government Printing Office.
Sabol, W. J., West, H. C., & Cooper, M. (2009). Prisoners in 2008. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
Shaffer, D. K., Listwan, S., & Latessa, E. (2001). A description of Ohio’s drug courts. Cincinnati: University of Cincinnati Center for Criminal Justice Research.
Silberman, C. (1978). Criminal violence, Criminal Justice. New York: Random House.
Travis, J., & Petersilia, J. (2001). Reentry reconsidered: A new look at an old question. Crime & Delinquency, 47(3), 291–313.
West, H. C. (2010). Prison inmates at midyear 2009: Statistical tables. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.
Wexler, H., Melnick, G., Lowe, L., & Hiller, M. (1999). Three-year reincarceration outcomes for amity in-prison therapeutic community and aftercare in California. The Prison Journal, 79(3), 337–351.
Wilson, D. B., Gallagher, C., & MacKenzie, D. (2000). A meta-analysis of corrections-based edu- cation, vocation, and work programs for adult offenders. Journal of Research in Crime and Delinquency, 37(4), 347–368.
C h a p t e r 1 : the Criminal Justice system
F O S T E R , C E D R I C 1 6 9 2 T S
Corrections in the Community.
© 2011, Elsevier Inc. All rights reserved. 13
C h a p t e r 2
How could “nothing works” prevail and punishment be promoted when, at a minimum, the research evidence suggested that at least some programs appeared to be working for some offenders under some circumstances? The evidence was not consistent with the myths of sociological criminology.
—D.A. Andrews and James Bonta
The importance of evaluating correctional programs has never been more pro- nounced, especially given the current economic crisis. With vast sums of money being spent on corrections, the public is demanding programs that work. The critical questions considered in this chapter are as follow. What works? What do we know about program effectiveness? What harm is done when we fail to develop effective programs? Moreover, this chapter provides an evidence- based framework for discussing research and practice on community correc- tions throughout this book.
One of the most important areas of contemporary concern for corrections officials is the design and operation of effective correctional intervention pro- grams. This is particularly relevant as there is consistent evidence that the
What Works in Correctional Intervention?
cost-effectiveness
length of follow-up
parole conditions
principles of effective intervention
shock probation
Key terms
F O S T E R , C E D R I C 1 6 9 2 T S
14 C h a p t e r 2 : What Works in Correctional Intervention?
public supports rehabilitation programs for offenders (Applegate, Cullen, and Fisher, 1997). Survey research also reveals strong support for public protec- tion as an important goal of corrections (Applegate et al., 1997). As a result, disagreements are not uncommon about what the best methods are to achieve these goals. On one side are advocates for more punitive policies, such as an increased use of incarceration, “punishing smarter” strategies (e.g., boot camps), or simply increasing control and monitoring of offenders. The limits of these approaches have been outlined and debated by others (Currie, 1985; Bennett, DiIulio, & Walters, 1996).
As Cullen and Applegate (1998) imply, the most disheartening aspect of these “get tough” policies is their dismissal of the importance of programming designed to rehabilitate offenders. Cullen and Applegate further question whether this rejection of rehabilitation is sound public policy. As many states have found, simply locking up offenders and “throwing away the key” has proven to be a very expensive approach to crime control. This approach is also very limited, as the vast majority of offenders will one day return to society. Many will return at best unchanged, and at worst with many more problems and intensified needs for service (Petersilia, 1992). For those advocating inca- pacitation, one must also ask what should be done with offenders while they are incarcerated? Some scholars, such as Cullen and Applegate, do not believe that incapacitation and rehabilitation are mutually exclusive. Furthermore, because the vast majority of offenders are supervised in the community at dif- fering degrees of intensity, it is even more important that we develop programs that work toward reducing recidivism.
Many of the “intermediate sanctions” that have been developed over the past few years are but a few examples of “programs” that often fail to live up to their expectations, particularly in terms of reductions in recidivism (Latessa, Travis, & Holsinger, 1997; Petersilia, 1997). While programs such as boot camps, Scared Straight, and other “punishing smarter” programs remain popular, there is little evidence that they will lead to reductions in recidivism. Figure 2.1 shows some of the results from various studies. Unfortunately, evidence seems to indicate that in some cases, punishing smarter programs actually lead to increases in recidivism rates. One of the main problems with such approaches is that they only send a message about what the offender should not do; these approaches do not teach them the skills that they need to address high-risk situations in the future.
In a study funded by the National Institute of Justice, Sherman and colleagues (1998) summarized what does not work in reducing recidivism.
n Correctional boot camps using traditional military basic training n Drug prevention classes focused on fear and other emotional appeals,
including self-esteem, such as DARE
F O S T E R , C E D R I C 1 6 9 2 T S
15What Works in Correctional Intervention?
n School-based leisure-time enrichment programs n “Scared Straight” programs in which juvenile offenders visit adult
prisons n Shock probation, shock parole, and split sentences adding time to
probation or parole n Home detention with electronic monitoring n Intensive supervision n Rehabilitation programs using vague, unstructured counseling n Residential programs for juvenile offenders using challenging experiences
in rural settings
Despite the punitive movement, increasing evidence shows that correctional treatment can be effective in reducing recidivism among offenders (Andrews et al. 1990; Cullen & Gendreau, 1989; Gendreau & Andrews, 1990; Redondo, Sanchez-Meca & Garrido, 1999; Van Voorhis, 1987). Nonetheless, some scholars remain unconvinced (Antonowicz & Ross, 1994; Lab & Whitehead, 1988; Logan & Gaes, 1993). The debate surrounding treatment effectiveness has been ongoing since Martinson’s proclamation that “nothing works,” with many still clinging to this mantra, despite evidence to the contrary. Primary among the reasons for disbelief in the potential effectiveness of cor- rectional programming is the failure to measure outcome properly and the lack of quality programs.
Gendreau (1996) examined hundreds of correctional and rehabilitation pro- grams that attempt to intervene with offenders. His results indicated that 64 percent of the offender rehabilitation studies (that had control groups)
% Recidivism increased
% Recidivism reduced
12
10
8
6
4
2
0
-2
-4 Fines Drug
testing ISP Juvenile
boot camps *Not shown: restitution/electronic monitoring = 3%, Scared Straight = 4%
FIGURE 2.1 Effects of punishing smarter programs on recidivism.* Sources: Gendreau, Goggin, and Cullen (2000); Aos, Phipps, Barnoski, Lieb (1999).
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?16
reported reductions in favor of the treatment group; in fact, the average reduc- tion in recidivism was 10 percent. Others have subsequently conducted simi- lar studies (Lipsey & Wilson, 1997) and have come to the same conclusion: rehabilitation can be effective in reducing recidivism. For example, Figure 2.2 is based on a landmark meta-analysis (or quantitative review of the litera- ture) conducted by Lipsey in 1999. Figure 2.2 shows the expected recidivism rates when various programming characteristics are factored into probation.
Gendreau and Paparozzi (1995) also found that when rehabilitation pro- grams incorporated at least some of the eight principles of effective interven- tion, those programs reduced recidivism in the range of 25–70 percent, with the average about 50 percent. The principles of effective intervention are as follows:
1. Programs should have intensive services that are cognitive-behavioral in nature, that occupy 40–70 percent of the offender’s time in a program, and that are from 3–9 months in duration. Cognitive-behavioral programs incorporate elements of cognitive theories, behavioral theories, and social learning theories (see Spiegler & Guevremont, 2009).
2. Programs should target the criminogenic needs of high-risk offenders, such as antisocial attitudes, peer associations, personal and emotional factors (e.g., aggression, deficits in self-control), substance abuse, family and marital problems, and education/employment deficits.
3. Programs should incorporate responsivity among offender, therapist, and program. Simply said, treatment program should be delivered in a manner that facilitates the offender’s learning new prosocial skills and addresses potential barriers.
400 10 20 30 50
Routine probation (P)
P+minimal program
P+best intervention type (B)
P+B+good implementation (I)
P+B+I+over 6 months duration
FIGURE 2.2 Expected recidivism with various intervention characteristics for noninstitutionalized juvenile offenders. Source: Lipsey (1999).
F O S T E R , C E D R I C 1 6 9 2 T S
Parole Effectiveness 17
4. Program contingencies and behavioral strategies are enforced in a firm but fair manner; positive reinforcers outnumber punishers by at least 4:1.
5. Therapists relate to offenders in interpersonally sensitive and constructive ways and are trained and supervised accordingly. Treatment is systematically delivered by competent therapists and case managers.
6. Program structure and activities disrupt the delinquency network by placing offenders in situations (with people and in places) where prosocial activities predominate.
7. Provide relapse prevention in the community by such tactics as planning and rehearsing alternative prosocial responses, anticipating problem situations, training significant others (family and friends) to provide reinforcement for prosocial behavior, and establishing a system for booster sessions.
8. A high level of advocacy and brokerage as long as the community agency offers appropriate services.
Similarly, Gendreau (1996) lists those interventions that have not been found to be effective in reducing recidivism:
n Talking cures n Nondirective, relationship-oriented therapy n Traditional medical model approaches n Intensive services directed to low-risk offenders n Intensive services oriented to noncriminogenic needs (or factors
unrelated to future criminal behavior)
One example of a program that was not effective in reducing recidivism is found in Table 2.1. In a review of substance abuse treatment, Lightfoot (1997) identified effective and ineffective types of treatment. Interestingly, the types of effective and ineffective treatment models for substance abusers mirror find- ings from studies of other offender types. Taxman (2000) made similar conclu- sions after reviewing the research on substance abuse treatment. Her findings are summarized in Table 2.2.
parole effeCtIveness What is actually known about the effectiveness of probation and parole, and other community correctional alternatives, and what should be future research priorities? The next section summarizes what is generally con- cluded about selected topic areas of interest in parole effectiveness. This discussion of topic areas is basically organized along the general flow of criminal justice decision points as they relate to parole; however, most of
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?18
table 2.1 Review of Drug Treatment Effectiveness by Lightfoot (1997)
What treatment types were effective in quasi-experimental and/or controlled studies: n Social-learning based treatments n Aversion therapy: Electrical/chemical counter-conditioning n Covert sensitization n Contingency management/contingency contracting n Broad spectrum therapies n Individualized behavior therapy n Community reinforcement n Behavior self-control thinking n Relapse prevention
What treatment types showed no clear evidence of effectiveness from controlled studies: n Acupuncture n Education n Lectures n Bibliotherapy n Self-help n Alcoholics anonymous n Narcotics anonymous n Al-Anon n Adult children of alcoholics n Psycho-therapy n Supportive n Confrontational n Pharmacotherapies
Source: Lightfoot, L. (1997).
table 2.2 Review of Drug Treatment Effectiveness by Taxman (2000)
What treatment types were successful at reducing recidivism? n Directive counseling n Behavior modification n Therapeutic community n Moral reasoning n Social competency cognitive behavior models n Emotional skill development n Cognitive skills n Behavioral skills
What treatment types showed no clear evidence of effectiveness of reduced recidivism? n Nondirective counseling n Reality therapy n Psychosocial education n 12-step or other self-help groups n Psychoanalytical
Source: Taxman (2000).
F O S T E R , C E D R I C 1 6 9 2 T S
Parole Effectiveness 19
the findings also pertain to probation, particularly those on supervision and innovative programs.
Institutional factors
Several aspects of the institutional experience are thought to be related to parole and its effectiveness, such as length of time incarcerated, prison behav- ior, institutional programs, and parole conditions imposed as conditions of release.
time served
Early research that examined the effects of the amount of time served in prison on parole has generally concluded that the shorter the amount of time served, the greater the likelihood of successful parole (Eichman, 1965; Gottfredson, Gottfredson, & Garofalo, 1977).
Similarly, Smith, Goggin, and Gendreau (2002) conducted a meta- analysis of the prison literature. Results included a total of 27 studies comparing community-based offenders (e.g., probationers) to inmates, as well as 23 stud- ies comparing prisoners who served longer sentences to prisoners who served shorter sentences. Results indicated that offenders who were imprisoned had recidivism rates approximately 7 percent higher than community-based offenders, and inmates who served longer sentences had a recidivism rate that was 3 percent higher than inmates with shorter sentences (Smith et al. 2002) (see Figure 2.3).
0%
1%
2%
3%
4%
5%
6%
7%
8%
Incarceration vs. community Incarceration: more vs. less
FIGURE 2.3 Percent increase in recidivism by type of sanction. Source: Smith et al. (2002)
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?20
Most researchers, however, have concluded that longer prison terms have an adverse effect on parolee chances of success, implying that the nega- tive aspects of prisonization seem to intensify with time. For example, in his study of shock probationers, Vito (1978) concluded that even a short period of incarceration has a negative impact. The question that remains unanswered by this research is: Are there any characteristics of inmates who have served more time that are also associated with an unfavorable parole outcome?
prison programs
Does participation in prison programs have an effect on recidivism? Existing research on the effectiveness of institutional programs and prison behavior has been limited in its scope. Most such programs are analyzed in relation to institutional adjustment, disciplinary problems, and impact of program participation on the parole-granting process. The few evalua- tions that included a parole period usually show little if any positive effects with regard to recidivism. A study by Smith and Gendreau (2007), how- ever, examined the relationship between program participation and recidi- vism in a Canadian sample of 5469 federal offenders. Results indicated that programs targeting criminogenic needs reduced postrelease recidivism by 9 percent for moderate risk offenders and 11 percent for high-risk offenders. German correctional researchers evaluated the effectiveness of social ther- apy programs across eight prisons, and the results were remarkably similar (Egg, Pearson, Cleland, & Lipton, 2000). The overall average reduction in recidivism for what is generally described as moderate- to high-risk adult incarcerates was 12 percent.
Most research that has examined prison behavior has not found a relationship between prison behavior and success on parole (Morris, 1978; von Hirsch & Hanrahan, 1979). However, a study by Gottfredson, Gottfredson, and Adams (1982) found that there is some relation between institutional infractions and infractions while on parole, after controlling for prior record (Finchamp, 1988). French and Gendreau (2006) also examined the relationship between participation in prison-based programs and misconducts/postrelease recidi- vism using meta-analytic techniques. Prison-based programs targeting crimi- nogenic needs reduced misconducts by 26 percent and reduced postrelease recidivism by 14 percent (French & Gendreau, 2004). Overall, however, there has not been a great deal of attention given to the relationship among insti- tutional programs, prison behavior, and subsequent success or failure on parole.
F O S T E R , C E D R I C 1 6 9 2 T S
Parole Effectiveness 21
Work and education programs
Two areas that have received some attention are work and education programs for offenders. Although the literature on education programs is inconclusive, evidence does seem to suggest that educational programs can affect inmate behavior and recidivism positively (Ayers, Duguid, Montague, & Wolowidnyk, 1980; Eskridge & Newbold, 1994; Linden & Perry, 1982; Roberts & Cheek, 1994). A study by MacKenzie and Hickman (1998) examined 12 correctional education programs for adult offenders. Of the 12 studies, eight produced results, suggesting that correctional education may have a positive impact on the rate of recidivism. They also concluded that while there were some incon- sistencies in the findings, the preponderance of evidence suggested that voca- tional education programs were effective in reducing recidivism.
Likewise, the literature on work programs does not convincingly demonstrate reduced recidivism (Vito, 1985b; Zeisel, 1982). Gendreau and Ross (1987; MacKenzie and Hickman, 1998), however, provide some principles that should be followed with regard to work programs: (1) they must enhance practical skills, (2) develop interpersonal skills and minimize prisonization, and (3) ensure that work is not intended as punishment alone.
In a study that reviewed the available research on corrections-based education, vocational, and work programs, Wilson, Gallagher, Coggeshall, Mackenzie (1999) looked at 33 studies. The majority of the studies they reviewed reported that edu- cation or work programs had reduced recidivism, with the average reduction
Schumaker and associates (1990) reported on the effects of in-prison vocational and academic coursework for 760 inmates followed on parole for 12 months. Their employment information and criminal activity rates (technical violations and new arrests) were gathered. The vocational/ academic groups generally had the lowest criminal activity rates and the highest employment rates. Those who had earned a General Education Diploma had the lowest criminal activity rate, and the control group who did not participate in vocation/academic programming had the highest criminal activity rate. Stevens and Ward (1997) tracked North Carolina inmates who had earned their associate and/ or baccalaureate degrees while imprisoned and found that prison inmates who earned their degrees tended to become law-abiding citizens significantly more often that inmates who did not advance their education.
Sources: Schumaker, Anderson, and Anderson (1990). Stevens and Ward (1997).
BoX 2.1 voCatIonal and aCademIC IndICators of parole suCCess
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?22
between 4 and 14 percent. Despite these promising findings, the researchers did not believe the research provided sufficient evidence of the effectiveness of these programs. This was attributed to a lack of high-quality studies.
therapeutic Communities
In recent years, prison-based therapeutic communities (TCs) have made a resur- gence (see Box 2.2). This is due in part to increased federal funding. Although there is a great deal of variation in how therapeutic communities operate, the essential ingredient is the principle that all staff and offenders provide thera- peutic experiences. TCs are more common in prisons, but many operate in community-based facilities, such as halfway houses.
A number of studies have shown that TCs can have an appreciable effect on recidivism rates, especially when community follow-up aftercare is provided (see Knight, Simpson, & Hiller, 1999; Martin, Butzin, Saum, & Inciardi, 1999; Wexler, Melnick, Lowe, & Peters, 1999). Figure 2.4 shows results from one such program operating in Delaware.
faith-Based programs
One of President George W. Bush’s initiatives was the expansion of faith-based programs in human service. While faith-based programs have a long history in corrections, there has been surprisingly little empirical research conducted on their effectiveness, and results are mixed. Religious programs in prisons may help inmates cope; however, research indicates that offenders who had poor coping skills prior to prison have poor coping skills in prison (Porporino & Zamble, 1984). Since 1985, 23 studies have explored the relationship between
80
70
60
50
40
30
20
10
0 Comparison Dropout Treatment Treatment
& Aftercare
FIGURE 2.4 Therapeutic community treatment: arrest rates after a 3-year follow-up (percent arrested). Source: Martin et al. (1999).
F O S T E R , C E D R I C 1 6 9 2 T S
Parole Effectiveness 23
religion and deviance in the general population. Eighteen of those studies show evidence that religiosity reduces deviance (i.e., people of strong faith are generally less criminal than nonbelievers); however, this does not appear to translate well into correctional programming.
Only two studies have examined effects of religious participation on insti- tutional adjustment and infractions. In 1984, Johnson studied 782 inmates in Florida. Results indicated no differences in disciplinary problems or insti- tutional adjustment for religious and nonreligious inmates. In 1992, Clear, Stout, and Dammer studied nonrandom sample of 769 inmates in 20 prisons in 12 states. They concluded that a prisoner’s religious participation had a sig- nificant and positive relationship to prison adjustment. They also found that other factors, such as age and race of offender, played a role.
Similarly, only three studies have examined religion and postrelease behav- ior. In 1987, Johnson and colleagues studied inmates released from four adult male prisons in New York. One group participated in Prison Fellowship Program (PFP); one did not. Results from this research indicated that the level of participation influenced prison adjustment; however, the direction was not always as anticipated:
n Participants with a high level of involvement PFP were less likely to commit infractions than low or moderate participants.
n However, high-PFP participants received more serious infractions. n High-PFP participants were significantly less likely to be rearrested
during the follow-up, but this relationship was strongest for whites and nonsignificant for African Americans.
Young, Gartner, and O’Connor (1995) followed a group of 180 federal inmates trained as volunteer prison ministers who attended special seminars and a matched control group. Overall, the seminar group had a significantly
Therapeutic communities, or TCs as they are commonly known, are eclectic in nature and offer an intense self-help model that focuses on the whole person. Staff and offenders are intimately involved in the treatment process. Confrontation and accountability are key ingre- dients of a TC. Offenders who engage in appropriate behavior are given “pull-ups” (positive reinforcement) by other offenders and staff, while those who engage in behavior detrimen- tal to them or others are given “haircuts” (confronted about their behavior). One of the criti- cisms leveled at TCs is their use of shaming and other degrading sanctions. For example, some TCs have been known to have offenders wear diapers like a baby, sit in chairs for long periods of time, wear dunce hats, and other punishments designed to change the behavior of the participant.
BoX 2.2 therapeutIC CommunItIes
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?24
lower rate of recidivism and maintained a higher survival rate than the con- trol group. Seminars were most effective with lower-risk subjects, whites, and women.
A more recent study by Sumter (1999) followed inmates from the Clear, Stout, Kelly, Hardyman, Shapiro. (1992) study. There were no differences in the recidi- vism rates between “religious” and “nonreligious” inmates; furthermore, regard- less of how many times they attended chapel, inmates who had a greater religious orientation in terms of values were less likely to recidivate. Sumter also found that offenders who attended religious programs upon release were less likely to recidivate; however, no relationship existed between attending inside and out- side prison. Participation in religious programs was certainly no panacea: 66 per- cent of “religious” prisoners experienced one or more arrests in the follow-up period. Sumter concluded that religion as a correctional program is complicated and multifaceted—as personal as it is social—and becomes more complicated in a prison setting. Inmates embrace religion for a number of reasons, some hearten- ing (spirituality and coping mechanism) and some cynical (get snacks, time out of cell, more freedom, looks good for parole). What we do know is that prisons distort everything. What may seem like a quest for spiritual awakening on the sur- face can simply be a way to get around the strictures of confinement. Related to this is the fact that we have little understanding of precisely how religion works or what the best definition of “religious” might be (conversion, weekly service attendance, number of books read in Bible study, punitive vs. redemptive orien- tation, or frequency of participation in such religious rites as attending church, participation in services, tithing, frequency of prayer, or proselytizing.) Regardless of the findings on faith-based correctional programs, most of us would agree that pursuit of religious understanding is a basic human right—prisoners who wish to engage in spiritual expression should be encouraged to do so, but this is true regardless of what the research finds. It does not mean that faith-based programs will have a significant effect on recidivism rates.
Given all the contradictions from the research, it is often difficult to deter- mine what, if any, effects prison-based treatment has on offender behavior. In a large study conducted by researchers in Washington State (Aos et al.1999), the research examined all the available studies and conducted a meta- analysis to determine effect sizes on recidivism. As can be seen in Figure 2.5, they found that Life Skills programs produced no reductions in recidivism, but subsidizing jobs for offenders age 27 and over produced reductions of 24 percent.
parole Conditions
Offenders who are granted parole are required to follow rules and con- ditions. Failure to do so can lead to reincarceration. With regard to the
F O S T E R , C E D R I C 1 6 9 2 T S
Parole Effectiveness 25
imposition of parole conditions, a nationwide survey of 52 parolee field supervision agencies, Allen and Latessa (1980) found 49 had residency requirements as a condition of parole and 47 had an employment require- ment. The Travis and Latessa (1984) survey found similar results. Despite the widespread requirement of parole conditions, the literature produced only three studies that were directly related to the imposition of these conditions and parole effectiveness. Although two studies (Beasley, 1978; Morgan, 1993) showed a relationship between stability of residency and parole success, the lack of research in this area makes generalization difficult.
One of the most important conditions of parole is the requirement to report regularly to a parole officer and not to leave a prescribed area, such as the county, without permission. Offenders who fail to report or whose where- abouts are unknown are called absconders. A study by Williams, McShane, and Dolny (2000) found that 27 percent of parolees in California were listed as absconders, and another study conducted by Schwaner (1997) in Ohio found 11 percent. Absconders have problems with alcohol abuse, have been convicted of a property crime, and have a history of prior parole vio- lation and absence of suitable housing (Buckholtz & Foos, 1996). Despite these high numbers of absconders, there has been little research on this subject.
parole release
Primarily in response to the supporters of determinate sentencing, research- ers have increasingly turned their attention to evaluating the success of parole supervision.
25
20
15
10
5
0 Subsidized Paroles*
Voc. Ed. T.C. Fin. Support Paroles
*This effect was for inmates over 27 years of age. There was no effect for younger adults.
FIGURE 2.5 Washington state study of the average effect sizes of prison programs (average percent reduction in recidivism). Source: Aos, et al. (1999).
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?26
Critics of parole supervision rely on two basic arguments to support their views. The first is that parole supervision simply is not effective in reducing recidivism (Citizens’ Inquiry on Parole and Criminal Justice, 1975; Wilson, 1977). The second, more philosophical argument is that supervision is not “just” (von Hirsch & Hanrahan, 1979). A more plausible conclusion is that the evidence is mixed; that parole supervision is effective in reducing recidivism rates among parolees (Flanagan, 1985).
Several studies have compared parolees to mandatory releases, but they have failed to control for possible differences in the selection of the groups (Martinson & Wilks, 1977). Other studies controlled for differences have reported favorable results (Gottfredson, 1975; Lerner, 1977), whereas other studies have reported less positive results (Jackson, 1983; Nuttal et al., 1977; Waller, 1974). In one study, Gottfredson, Mitchell-Herzfeld, and Flanagan (1982:292) concluded “much of our data does indicate an effect for parole supervision, an effect that varies by offender attributes, and an effect that appears not to be very large.” The existing evidence seems to be in favor of parole supervision, although there is no clear consensus as to its effectiveness.
Even the most outspoken critics of parole agree that the agencies responsi- ble for the task of supervision are often understaffed and that their officers are undertrained, underpaid, and overworked. They are inundated with exces- sively large caseloads, workloads, and paperwork. Community services are either unavailable or unwilling to handle parolees; as a consequence, parole officers are expected to be all things to all people. As indicated in Chapter 7, they are also expected to perform the dual roles of surveillance–police officer and rehabilitator–treatment agent.
Some evidence suggests that by shortening the amount of time on parole, we could save a considerable amount of money and time while not seriously increasing the risk of failure. Most data seem to indicate that the majority of failures on parole occur during the first 2 years (Flanagan, 1982; Hoffman & Stone-Meierhoefer, 1980) and drop significantly thereafter. There is also some evidence that early release into the community and from parole incurs no higher risk to the community and, in fact, is justifiable on cost con- siderations (Holt, 1975), a conclusion echoed by MacKenzie and Piquero (1994). It is also important to note that easing the offender back into the community through community residential centers and furlough programs can facilitate the early release process. The definition and purpose of com- munity residential centers and furloughs are found in Boxes 2.3 and 2.4, respectively.
F O S T E R , C E D R I C 1 6 9 2 T S
Parole Effectiveness 27
A Washington State (1976) 10-year follow-up of parolees found that the first year of parole was critical, with more than one-half of those paroled return- ing to prison during this time period. In this study, there were more failures in the second 6 months after release than in first. It was also found that those convicted of murder and manslaughter were less likely to recidivate and that property offenders—especially those convicted of burglary, auto theft, and forgery—had the highest failure rate. As expected, younger parolees did signifi- cantly worse than those 40 years of age or older. Blacks did slightly worse than whites after the first 6 months, and Native Americans did significantly worse than all other groups.
It is important to note that many of the failures on parole supervision are a result of technical violations (TVs), that is, failure to abide by the conditions imposed by the parole board. TVs can range from a positive drug test to fail- ure to report as directed. Some states have reported that the reincarceration rates for parolees have dropped; Texas reported their high of 50 percent in 1992 declined to 31 percent in 1997 and Pennsylvania reported a decline from 50 percent in 1994 to 39 percent in 1996. Reasons offered for these
Community residential centers (also known as halfway houses) are residential facilities where probationers, furloughees, and parolees may be placed when in need of a more structured set- ting. The primary purpose of a halfway house is to limit an offender’s freedom while encouraging reintegration into society through employment, education, treatment, habilitation, restitution, training, compliance with financial sanctions, and other activities designed to rehabilitate the offender and deter future crime (Ohio Community Corrections Organization, 1993).
BoX 2.3 CommunIty resIdentIal Centers
Furlough is a phased reentry program designed to ease the offender’s transition from prison to the community. Furloughs include escorted or unescorted leaves from confinement, granted for designated purposes and time periods (funerals, dying relatives, etc.), before the formal sen- tence expires. Used primarily for employment, vocational training, or education, furlough in effect extends the limits of confinement to include temporary residence in the community dur- ing the last months of confinement. Furloughees are frequently required to reside in commu- nity residential centers. Furloughs allow parole boards to observe the offender’s behavior in the community and may lead to faster release from parole supervision for those adjusting favor- ably. Because furloughees are closely screened and supervised in the community, failure rates appear to be low. For example, Ohio reports a nine percent return to prison rate for calendar 1992 (Ohio Community Corrections Organization, 1993).
BoX 2.4 furloughs
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?28
declines are not known, but possible explanations include an older parole population and lower-risk offenders being sent to prison. Figures 2.6 and 2.7 illustrate the reason for failure for probationers and parolees.
Studies of parole success by type of offense indicate repeatedly that those who commit murder are among the best parole risks (Neithercutt, 1972). Reasons for this conclusion vary; the explanation offered most frequently is that most murderers tend to be first offenders who have committed crimes of passion. Another reason cited is age; because most convicted murderers spend a great amount of time incarcerated, they tend to be older (and more mature) when released, usually after the high-crime-risk years of 16–29 (see Table 2.3).
In a study of murderers who had been given a death sentence and then had that sentence commuted when Furman v. Georgia was overturned, Vito, Wilson, and Latessa (1991) found that 43.5 percent of death row inmates in Ohio were paroled and that 25 percent were returned to prison (recidivated). These results
73% 10%
17% New sentence
Same sentence
Other/unknown*
*Absconder: 7%; unknown: 12%; death: 2%; detainer or warrant: 2% all other: 49%.
Note: Detail may not total due to rounding.
FIGURE 2.6 Probationers returned to custody during 2005. Source: Bureau of Justice (2006).
47%
7%
22%
20%
New sentence
Absconder
PV/other
Miscellaneous*
*Transferred: 2%; death: 2%; other: 3%.
FIGURE 2.7 Parolees returned to prison during year 2005. Source: Bureau of Justice (2006).
F O S T E R , C E D R I C 1 6 9 2 T S
Parole Effectiveness 29
were very similar to those found in Texas, where 19 percent of the paroled Furman cases recidivated (Marquart & Sorensen, 1988), and in Kentucky, which had a 29 percent failure rate (Vito & Wilson, 1988). Overall, studies examin- ing murderers were found to generate consistent findings and conclusions over time.
A study by Austin (2001:331) examined the important issue of prisoner reen- try. He concluded “it is not clear that parolees, in the aggregate, pose as large a public safety problem as some believe.”
In order to summarize, we have selected data from a national study of parole recidivism (Beck, 1987). These data confirm two important points with regard to parole effectiveness: (1) recidivism rates vary depending on the definition of recidivism and (2) the type of offense and age are important factors in deter- mining parole success (Table 2.4). Other findings included the following:
n Approximately 10 percent of the persons paroled accounted for 40 percent of the subsequent arrest offenses.
n About one-fifth of the subsequent arrests occurred in states other than the original paroling state.
n An estimated 37 percent of the parolees were rearrested while still on parole.
n Recidivism rates were highest in the first 2 years after an offender’s release from prison. Within 1 year, 32 percent of those paroled had been arrested; within 2 years, 47 percent had been rearrested.
n Recidivism was higher among men, blacks, and persons who had not completed high school than among women, whites, and high school graduates.
table 2.3 Age of Parolees and Likelihood of Failure
Age at time of prison release
Rate of return to prison by years after release from prison
Years
1 2 3 4 5 6 7
18–24 years old 21% 34% 41% 45% 48% 49% 50% 25–34 12 21 28 33 37 41 43 35–44 7 14 18 22 26 30 34 45+ 2 4 6 8 10 11 12 All ages 14 23 29 34 37 40 42
Source: Adopted from Beck (1987).
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?30
n Almost three-quarters of those paroled for property offenses were rearrested for a serious crime compared to about two-thirds of those paroled for violent offenses.
n Approximately one-third of both property offenders and violent offenders were rearrested for a violent crime upon release from prison.
n The longer the parolee’s prior arrest record, the higher the rate of recidivism—more than 90 percent of parolees with six or more previous adult arrests were rearrested compared to 59 percent of the first-time offender.
table 2.4 Failure Rates of Parolees
Percent of Young Parolees Who within 6 Years of Release from Prison Were
Rearrested Reconvicted Reincarcerated
All Parolees: 69% 53% 49%
Sex: Male 70% 54% 50% Female 52 40 36
Race: White 64% 49% 45% Black 76 60 56 Hispanic 71 50 44 Other 75 65 63
Education: Less than 12 years 71% 55% 51% High school graduate 61 46 43 Some college 48 44 31
Paroling offense: Violent offense 64% 43% 39% Murder 70 25 22 Robbery 64 45 40 Assault 72 51 47
Property offense 73 60 56 Burglary 73 60 56 Forgery/fraud 74 59 56 Larceny 71 61 55
Drug offense 49% 30% 25%
Source: Adopted from Beck (1987).
F O S T E R , C E D R I C 1 6 9 2 T S
31Probation Effectiveness
n The earlier the parolee’s first adult arrest, the more likely the chances for rearrest—79 percent of those arrested and charged as an adult before the age of 17 were rearrested compared to 51 percent of those first arrested at the age of 20 or older.
n Time served in prison had no consistent impact on recidivism rates— those who had served 6 months or less in prison were about as likely to be arrested as those who had served more than 2 years.
proBatIon effeCtIveness As with parole, the quality of probation research is dubious. Unlike parole, which is found on state and federal levels, probation still remains primarily a local governmental function. Facts are that (1) probation can be found at local, state, and federal levels; (2) there are municipal and county probation depart- ments; and (3) probation serves both misdemeanants and felons. These, com- bined with the problems discussed previously, make research in probation very difficult to conduct. Indeed, much of the research has been limited to only the several probation departments to which researchers have been welcome. This event gives us a limited sense of the true picture of probation.
As with parole, the research on probation effectiveness is divided into sec- tions. However, unlike our presentation of parole, the research on proba- tion is divided into five groups: studies that (1) compare the performance of offenders receiving alternative dispositions; (2) simply measure probation outcome without comparison with any other form of sanction; (3) measure
If the current average sentence remained the same for violent offenders and the required policy were to be to serve 85 percent of the current sentence, the predicted times served would be increased by: n for new admissions, 26 months n for prisoners now present, 84 months n for releases, an average of 33 months longer in prison Based on current sentences, the estimated time predicted to be served (months) would be:
Percent of sentence served Prison admissions Prisoners present Prison releases
Current 62 100 43
75% 78 162 67
85% 88 188 76
100% 104 216 89
BoX 2.5 estImated tIme to serve
Source: Greenfield (1995).
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?32
probation outcome and then attempt to isolate the characteristics which tend to differentiate between successful and nonsuccessful outcomes; (4) examine the cost-effectiveness of probation; and (5) examine probation combined with therapeutic drug courts.
probation versus alternative dispositions
To examine the effectiveness of probation compared to other dispositions, we looked at six studies. Three of the studies compared recidivism rates of indi- viduals placed on probation with individuals sentenced to incarceration. Babst and Mannering’s (1965) study compared similar types of offenders who were imprisoned or placed on probation. The sample consisted of 7614 Wisconsin offenders who were statistically comparable in original disposition, county of commitment, type of offense committed, number of prior felonies, and mari- tal status. Parolees were followed for 2 years, and probationers were followed for 2 years or until discharge from probation, whichever came first. Violations were defined as the commission of a new offense or the violation of proba- tion/parole rules. Findings of this study showed that, for offenders with no prior felony convictions, the violation rate was 25 percent for probationers and 32.9 percent for parolees. For offenders with one prior felony conviction, viola- tion rates were 41.8 percent for probationers and 43.9 percent for parolees; for offenders with two or more felonies, rates were 51.8 percent for probationers and 48.7 percent for parolees. With respect to the difference in violation rates for first offenders (which was statistically significant at the .05 level), Babst and Mannering note that this finding could be a result of the fact that parolees are a more difficult group to supervise or could actually show that, at least for first offenders, incarceration does more harm than good.
Another study done in Wisconsin (Wisconsin Division of Corrections, 1965) compared the performance of burglars who had no previous felony convictions, sentenced to prison, or placed on probation. While this study also attempted to investigate the characteristics associated with successful and nonsuccessful probationers and parolees, we simply report at this point that the violation rate (based on a 2-year follow-up, using the same definition of violation rate as Babst and Mannering described earlier) for burglars placed on probation was 23 percent, whereas that for burglars who were incarcerated and then placed on parole was 34 percent. Thus, it appears that, as with the Wisconsin study, probation was more successful than parole.
The Pennsylvania Program for Women and Girl Offenders, Inc. (1976) com- pared recidivism rates among all women placed on state probation or released on state parole during a 2-year period. Recidivism was defined as any technical violation of probation or parole or any new criminal charge. Findings showed that, overall, women placed on probation had a 35.6 percent recidivism rate.
F O S T E R , C E D R I C 1 6 9 2 T S
33Probation Effectiveness
When only women with no prior convictions were considered, probation- ers had a 24 percent recidivism rate and parolees had a 23.1 percent rate. Differences between these rates were not statistically significant.
Vito (1978) compared regular probationers with shock probationers (who served at least 30 days in prison). He found that shock probationers had a 40 percent higher probability of failure than those released to regular supervision. Vito and Allen (1981:16) concluded:
… the fact of incarceration is having some unknown and unmeasurable effect upon (the more unfavorable) performance of shock probationers. … It could be that the negative effects of incarceration are affecting the performance of shock probationers.
Whereas these four studies compared probation with some form of incarcer- ation, a California study (California Department of Justice, 1969) compared violation rates among offenders placed on probation, offenders sentenced to probation following a jail term, and offenders given straight jail sentences. The study examined the performance of a cohort of offenders, all of whom had an equal exposure of one full year in the community. For the probation group, cohort status was gained on the date of the beginning of the proba- tion period; for the group receiving jail sentences, cohort status began on the date of release from jail. To evaluate the relative effectiveness of these dispositions, three violation levels were used: “none” or no known arrest for a technical violation or a new offense, “minor” or at least an arrest and per- haps a conviction resulting in a jail sentence of less than 90 days or proba- tion of 1 year or less, and “major,” signifying at least a conviction resulting in a jail sentence of not less than 90 days or a term of probation exceeding 1 year. Because each case was followed for only 1 year, the final outcome of a violation occasionally did not occur until after the year was over. If it could be inferred that the disposition or sentence was the result of an arrest that did occur within the follow-up year, the action was included in the vio- lation rate.
The findings of this study are illustrated in Table 2.5. Those offenders receiving jail sentences without the benefit of probation services have the worst record of recidivism.
These studies illustrate that, as a disposition, probation appears to be more effective than incarceration, even for a short period of time. This may be due, in part, to the fact that probationers immediately return to the community, their jobs, and their families.
Finally, an Alaska study (Alaska Department of Health and Social Services, 1976) utilized an experimental design to compare the performance of misdemeanant
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?34
offenders receiving probation supervision with offenders officially on proba- tion but not required to report to the probation unit. The groups were cre- ated by random assignment to the experimental group (under supervision) or the control group (no supervision) and were followed for periods ranging from 2 months to slightly more than 2 years. Performance was assessed by means of recidivism, defined as the conviction for a new offense. Findings of the study showed that 22 percent of experimental group members and 24 of control group members had been convicted of new offenses during the follow- up period.
Given the paucity of research and the caution with which recidivism data must be approached, it is nearly impossible, not to mention inappropriate, to attempt to draw any definitive conclusions from these studies about the effectiveness of probation compared to other alternative dispositions. Nonetheless, it appears from the limited research that has been conducted that the following tentative conclusions can be reached. Of studies that compared probation to incarcera- tion, it tentatively appears that probation may have a significant impact on first offenders. It may also be suggested that the severity of violations appears to increase in proportion to the severity of the disposition. It does not appear that the provision of probation supervision for misdemeanants is more effec- tive than an unsupervised probation period.
probation outcome
A number of studies reported recidivism rates only for probationers. Thirteen of these were reviewed, but one should remember that definitions of failure, follow-up periods, and types of offenders differ significantly from one study to another. Table 2.6 includes the author, types of instant offenses committed by the probationers in the study, and the definition of failure used in the study, the length of follow-up, and failure rates.
These summary descriptions illustrate many of the problems associated with attempting to assess probation effectiveness. The type of offenders constituting
table 2.5 Violation Levels of Sentenced Offenders in California
Violations
Sentence None Minor Major
Probation only 64.7% 23.7% 11.6%
Jail, then probation 50.3% 31.7% 18.0%
Jail only 46.6% 29.5% 23.9%
Source: California Department of Justice (1969). F O S T E R , C E D R I C 1 6 9 2 T S
35Probation Effectiveness
the samples (as represented by instant offenses) varies, as do the definitions used in each study to characterize failure. Four studies computed failure rates while offenders were on probation, and the length of follow-up periods ranged from several months to many years.
table 2.6 Studies Reporting Recidivism Rates for Probationers
Study Instant offense Failure Follow-up Failure rate (%)
Caldwell (1951) Internal revenue laws (72%)
Convictions Postprobation 5½–11½ years
16.4
England (1955) Bootlegging (48%) and forgery
Convictions Postprobation 6-12 years
17.7
Davis (1955) Burglary, forgery, and checks
Two or more violations (technical and new offense)
To termination 4–7 years
30.2
Frease (1964) Unknown Inactive letter, bench warrant, and revocation
On probation 18–30 months
20.2
Landis et al. (1969) Auto theft, forgery, and checks
Revocation (technical and new offense)
To termination 52.5
Irish (1972) Larceny and burglary Arrests or convictions Postprobation Minimum 4 years
41.5
Missouri Div. Probation & Parole (1976)
Burglary, larceny, and vehicle theft
Arrests and convictions
Postprobation 6 months–7 years
30.0
Kusuda (1976) Property Revocation To termination 1–2 years
18.3
Comptroller General (1976)
Unknown Revocation and postrelease conviction
Postprobation 20 months average
55.0
Irish (1977) Property Arrests Postprobation 3–4 years
29.6
Petersilia (1985) Felony probationers Arrests Tracked over 40 months
65.0
McGaha, Fichter, and Hirschburg (1987)
Felony probationers Arrests Tracked over 40 months
22.3
Vito (1986) Felony probationers (excluding drug offenses)
Arrests Tracked over 40 months
22.0
Maxwell, Bynum, Gray, and Combs (2000)
Felony probationers Revoked Tracked over 30 months
47.0
Source: Adapted and updated from Allen, Carlson, and Parks (1979).
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?36
Most of the studies reviewed here stated that their purpose was to assess “pro- bation effectiveness”; however, unlike the five studies examined earlier, none of these studies defined a base (such as a failure rate for comparable parolees or offenders on summary probation) against which to compare findings in order to support a claim that probation is an effective alternative for rehabili- tating offenders.
In a study of 1700 probationers in Michigan, Maxwell and colleagues (2000) found that only 24 percent of probationers had no technical violations dur- ing a 30-month follow-up period. Results are summarized in Figure 2.8. Treatment-related violations, such as dirty urine or failing to attend treat- ment programs, accounted for the largest proportion of violations, fol- lowed by failure to appear. Only 13 percent of the violations were for a new crime.
MacKenzie, Browning, Skroban, and Smith (1999) studied the impact of pro- bation on the criminal activities of offenders. They concluded that probation alone had an effect on property and dealing crimes. Probation was not sig- nificantly associated with reductions in personal crimes of forgery and fraud offenses. The conclusion reached by the authors was that “probation may be more effective than previously thought.”
In one of the more critical studies of probation effectiveness, Petersilia (1985) examined 1672 felony probationers from two counties in California over a 40-month period. She found that more than 67 percent were rearrested and 51 percent were convicted for a new offense. Petersilia concluded that felony probationers posed a significant risk to the community. Critics of the Petersilia study quickly pointed out that two urban counties in California are not representative of the rest of the states or the country. Two replication studies,
40
35
30
25
20
15
10
5
0 Treatment Related
Failure to Report
New Crime
FIGURE 2.8 Michigan study of probationer recidivism (percentage). Source: Maxwell et al. (2000).
F O S T E R , C E D R I C 1 6 9 2 T S
37Probation Effectiveness
one in Kentucky (Vito, 1986) and one in Missouri (McGaha et al. 1987), found quite different results. In both Kentucky and Missouri, felony probationers were rearrested at about one-third of the rate as those in California.
One suggested reason for these differences is the effects of budget cuts on pro- bation staffs in California. Caseloads of more than 300 are commonplace, and there is virtually no enforcement of probation conditions (Snider, 1986). Despite these reasons, it is important to note that the effectiveness of felony probation is still very much in debate.
Morgan (1993) studied 266 adult felony probationers in Tennessee to deter- mine factors associated with favorable probation outcome and those that would predict success. She found that only 27 percent of the probationers failed and that females, married probationers, and those with higher levels of education were most likely to succeed. Factors significantly related to probation failure were prior felonies, prior probation, prior institutional commitment, and pro- bation sentence length (the longer the sentence, the more likely the failure).
The review of these studies demonstrates that little progress has apparently been made in recent years toward an adequate assessment of probation. Conclusions drawn by the authors of these studies, however, appear to sug- gest that an unwritten agreement or rule of thumb exists that probation can be considered to be effective and that a failure rate above 30 percent indicates it is not effective. This tendency is suggested by the comments in Table 2.7.
State courts in 32 counties across 17 states sentenced 79,000 felons to probation in 1986. Within 3 years of sentencing, while still on probation, 43 percent were rearrested for a felony. An esti- mated 18 percent of the arrests were for a violent crime (murder, rape, robbery, or aggravated assault); 33 percent were for a drug offense (drug trafficking or drug possession). Of each 100 felony probationers tracked for 3 years: n 26 went to prison n 10 went to jail n 10 absconded These findings are based on a follow-up survey of felons on probation using a sample that represented a fourth of the total 306,000 felons sentenced to probation in 1986. The survey used state criminal history files and probation files to obtain information. It was not based on a nationally representative sample; 39 percent of the follow-up cases were from a single state (California). Nevertheless, based on 12,370 sample cases representing 79,043 felons placed on probation in the counties and states studied, the follow-up represents the largest survey of its kind ever done.
BoX 2.6 proBatIoners rearrested for a felony WIthIn three years
Source: Bureau of Justice Statistics National Update (1994).
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?38
probation outcome and statistics
In addition to measuring the effectiveness of probation, a number of studies have also attempted to isolate characteristics that could be related to offender rehabilitation. Table 2.8 presents a summary of the major factors that were found in each study to be statistically correlated with failure. Keeping in mind the methodological differences among the studies in terms of definition of failure and specification of follow-up period, it appears that the one character- istic found to be associated most commonly with failure is the probationer’s previous criminal histories. Other factors frequently cited are the youthfulness of the probationer, marital status other than married, unemployment, and educational level below the eleventh grade.
Factors such as employment and education are dynamic factors that are cor- related with outcome. Because these areas can be addressed during super-
table 2.7 Evaluations of Effectiveness of Probation
Year Author Failure rate Comments
1951 Caldwell 16% [P]robation is an effective method of dealing with federal offenders …
1955 England 18% A reconviction rate of less than one-fifth or one-quarter … is an acceptable performance for a probation service.
1976 Missouri 30% Probation is an effective and efficient way of handling the majority of offenders in the State of Missouri.
1976 Comptroller 55% [P]robation systems we reviewed were achieving limited success in protecting society and rehabilitating offenders.
1977 Irish 30% [S]upervision program is effectively accomplishing its objective.
1985 Petersilia 65% Felony probation does present a serious threat to public safety.
1986 Vito 22% Felony probation supervision appears to be relatively effective in controlling recidivism …
1987 McGaha et al. 22% In Missouri, it does not appear that the current use of felony probation poses a high risk to the security of the community.
1991 Whitehead 40% [C]alls for drastically reduced use of probation for felony offenders are only partially in order.
1994 Morgan 27% [I]nadequate employment and unemployment are major impediments to achieving successful probation adjustment and … outcome.
1997 Mortimer and May
18% Electronic monitoring and probation orders yield comparable success rates.
Source: Adapted and updated from Allen et al. (1979), Petersilla (1985), Vito (1986), McGaha et al. (1987), Whitehead (1991), Morgan (1993), and Mortimer and May (1997).
F O S T E R , C E D R I C 1 6 9 2 T S
table 2.8 Studies Reporting Factors Related to Probationer Recidivism
Study
Previous criminal history Youth
Status other than married
Not employed
Low income (below $400)
Education below 11th grade
Abuse of alcohol or drugs
Property offender
On-Probation maladjustment
Imposition of conditions
Caldwell (1951)
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
—a
England (1955)
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
—a Significant correlation
Davis (1955)
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Frease (1964)
Significant correlation
Significant correlation
—b Significant correlation
Significant correlation
Significant correlation
Landis (1969)
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Irish (1972)
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
—a
MO Div. of Prob. & Parole
Significant correlation
Significant correlation
Significant correlation
Significant correlation
—c Significant correlation
Significant correlation
Significant correlation
Kusuda (1976)
Significant correlation
Significant correlation
Significant correlation
—b —a Significant correlation
—a
Comptroller General (1976))
—a
Irish (1977)
Significant correlation
—a Significant correlation
Petersilia (1985)
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Benedict (1998)
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
Significant correlation
aIn these studies, instant and postprobation offenses committed by probationers were predominantly “property”; however, a correlation between property offenses and recidivism was not investigated. bCorrelation only with income between $100 and $400; those who make less than $100 and those who made above $400 both had an equal probability of success. cCorrelation only with income between $100 and $700; those who made less than $100 or above $700 both had an equal probability of success. Source: Adapted from Allen et al. (1979).
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?40
vision, one can reasonably view these factors positively; we have a clear indication of offender needs, and they can be improved. However, a ques- tion remains as to whether probation and parole officers are addressing these needs adequately. When probation and parole agencies fail to meet offender needs that are correlated with outcome, the result is often higher failure rates.
Cost-effeCtIveness While the public has demanded tougher sentences, it has become increas- ingly apparent that the cost associated with more incarceration and prison construction is astronomical. Estimates place the cost of constructing a max- imum-security prison at approximately $80,000 per bed, and the annual cost of maintenance and housing inmates at more than $21,000 (Camp & Camp, 2000). Despite the opening of 371 new prisons since 1991 and constructing an additional 558,000 beds since 1992 (Camp & Camp, 2000), the acute shortage of prison space has made incarceration a scarce resource. Many states are faced with severe budget deficits, and legislators and the public are reluctant to vote for new prison construction. However, ample evidence also shows that once prisons are built they are filled. In addition, 19 states are under court order to increase services or reduce or limit their prison populations or the population in a specific prison (American Correctional Association, 2000). Because of the increasingly high cost associated with incarceration, researchers have begun to focus on the cost-effectiveness of alternatives.
Shock probation (also known as “reconsideration of sentence” or “shock therapy”) is a program allowing sentencing judges to reconsider the offender’s original sentence to imprisonment and then recall the inmate for a sentence to probation within the community under conditions deemed appropriate. It is presumed that a short term of incarceration would “shock” the offender into abandoning criminal activity and into pursuit of law-abiding behavior. It can be seen as an alternative disposition for sentencing judges who wish to control probationer behavior through deterrence and tourniquet sentencing. It is a last-ditch program used by some judges in the difficult decision of how best to protect the public while maximizing offender reintegration. In more recent years, it has become a “front end” solution to prison overcrowding. Vito (1985a) found reincarceration rates to range from 10 to 26 percent across many studies; Boudouris and Turnbull (1985) found a rearrest/revocation rate of 39 percent in Iowa over a lon- ger follow-up period. The latter also found that sex and substance abuse offenders were most responsive to shock incarceration and that the cost savings of sentencing offenders to shock probation would be substantial.
BoX 2.7 shoCK proBatIonF O S T E R , C E D R I C 1 6 9 2 T S
41Cost-Effectiveness
In light of these factors, and in addition to research aimed at measuring effec- tiveness in terms of recidivism, there have been attempts to demonstrate cost- effectiveness of probation. Typically, with criminal justice agencies, costs are usually divided into three types: processing, program, and client-centered. Processing costs include monies spent in identifying and selecting individ- uals for a given program. Program costs are expenditures associated with incarceration and include direct costs, such as loss of earning, and indirect costs, such as psychological effects upon of alienation/prisonization, social stigma, and other detrimental effects upon the prisoner’s marriage and fam- ily (Nelson, 1975).
Similarly, benefits generated by probation could include savings to society through the use of diversion, wages, and taxes generated by the participants, and reduced crime or recidivism rates (Vito & Latessa, 1979). In addition, there are the costs associated with failure, such as the monetary loss and grief expe- rienced by the victims.
Studies that provided the most thorough financial comparisons were those that treated the cost–benefit analysis as their primary focus and considered direct and indirect costs and benefits.
In one of the most comprehensive studies of probation costs, Frazier (1972) attempted to develop realistic cost information on probation and incarcera- tion for the purpose of comparison. A number of estimates were used to com- pare the indirect costs associated with incarceration. These factors included the average wage and average months employed per year, average taxes paid on gross wages, and the cost of welfare support for children whose wage-earn- ing parent had been incarcerated. These figures were based on data collection from a representative sample of 115 inmates and were then extended to the
Morgan reviewed the probation outcome literature through 1991 and concluded that probation is effective as a correctional alternative. Failure rates ranged from 14 to 60 percent for a group that had already committed crime; success rates vary from 40 to 86 percent. Factors associated more frequently with failure on probation included age, sex, marital status, low income, prior criminal record, and employment status. Those most likely to fail were unem- ployed or underemployed young males with a low income and prior criminal record. The recon- viction offenses of those who failed were more likely to be minor misdemeanors rather than felonies. Probationers who were adequately employed, married with children, and had lived in their area for at least 2 years were most often successful when placed on probation.
Source: Morgan (1993).
BoX 2.8 proBatIon as a CorreCtIonal alternatIve
F O S T E R , C E D R I C 1 6 9 2 T S
C h a p t e r 2 : What Works in Correctional Intervention?42
entire inmate population in Texas for 1970. The authors concluded that if 3000 inmates were diverted, a 1-year savings of $5,715,000 could be generated, and this, in 1970 dollars.
There have been several other cost–benefit analyses, however, that tend to include cost comparisons as part of a large research effort, and thus are plagued by errors of omission and incomplete costs or benefit identifica- tions (Erwin, 1984; Latessa, 1985, 1986; Pearson, 1985; System Sciences, 1982; Wetter, 1985). Despite their limitations, these studies also support the contention that probation is a cost-effective alternative to incarceration. In one study that included cost-effectiveness as part of a larger research effort, Latessa et al. (1997) compared the cost of four correctional options [regular probation, intensive supervised probation (ISP), community-based correc- tional facilities (CBCFs), and prison]. Figure 2.9 shows the average cost per offender by for supervision through discharge. This shows that probation and ISP are clearly less expensive alternatives to residential and institutional options. They went on to conclude:
n Compared to incarceration, placement in ISP or CBCF program produces savings and revenues between $4500 and $5000 per offender when compared to imprisonment.
n Intensive supervised probation and CBCF programs were more expensive than regular probation, but substantially less expensive than imprisonment.
n If only half of the offenders served in the community had been incarcerated, the state realized a savings of $49 million.
$0
$10,000
$20,000
$30,000
$40,000
$50,000
$60,000
$70,000
Probation Community based correctional facility
Prison Prison and parole
FIGURE 2.9 Average cost of supervision through discharge in Ohio. Source: Latessa (2006).
F O S T E R , C E D R I C 1 6 9 2 T S
43Summary
n Offenders in the ISP and CBCFs paid more than $3.4 million in court costs, fines, and restitution and $1.7 million in the value of community service and labor over a 3-year period.
In a review of seven cost–benefit analysis studies conducted on correctional alternatives, Welsh and Farrington (2000) found that for each dollar spent on programs, the public received a return of $1.13 to $7.14 in various savings. Likewise, Cohen (1998) determined the monetary value of saving a high-risk youth at between $1.7 and $2.3 million. Aos and associates (1999) calculated the cost savings from selected correctional programs based on expected reduc- tions in recidivism rates (see Table 2.9).
summary There are now several meta-analytic reviews of the correctional treatment liter- ature conducted by different authors, and the results have been replicated with remarkable consistency. These findings are collectively referred to as the “what works” literature, and have been summarized into the “principles of effective intervention” with offender populations. In general, these meta-analyses have been very critical of the “nothing works” doctrine proposed by Martinson and others (Whitehead & Lab, 1989). Furthermore, the results have not supported the use of intermediate sanctions (e.g., electronic monitoring, house arrest, restitution, etc.) and other more punitive approaches. At the same time, there are certain strategies and programs that are well supported by the literature.
table 2.9 Cost Savings from Selected Correctional Programs
Juvenile boot camps Taxpayers receive $0.42 for every dollar spent In-prison vocational education Taxpayers receive $2.30 for every dollar spent Adult basic education Taxpayers receive $1.71 for every dollar spent In-prison therapeutic communities Taxpayers receive $0.76 for every dollar spent Sex offender treatment programs Taxpayers receive $0.25 for every dollar spent Life Skills programs Taxpayers receive $0.00 for every dollar spent Job counseling and job search for inmates leaving prison
Taxpayers receive $2.84 for every dollar spent
Drug courts Taxpayers receive $1.69 for every dollar spent Subsidized jobs for inmates leaving prison
Taxpayers receive $0.67 for every dollar spent
Adult intensive supervision programs Taxpayers receive $0.39 for every dollar spent Case management substance abuse programs
Taxpayers lost $0.15 for every dollar spent
Adopted from Aos et al.(1999)
