California Department
of Corrections and
Rehabilitation
Several Poor Administrative Practices
Have Hindered Reductions in Recidivism
and Denied Inmates Access to In‑Prison
Rehabilitation Programs
January 2019

REPORT 2018‑113

CALIFORNIA STATE AUDITOR
621 Capitol Mall, Suite 1200 | Sacramento | CA | 95814

916.445.0255 | TTY 916.445.0033

For complaints of state employee misconduct,
contact us through the Whistleblower Hotline:

1.800.952.5665

Don’t want to miss any of our reports? Subscribe to our email list at

auditor.ca.gov

For questions regarding the contents of this report, please contact Margarita Fernández, Chief of Public Affairs, at 916.445.0255
This report is also available online at www.auditor.ca.gov | Alternative format reports available upon request | Permission is granted to reproduce reports

Elaine M. Howle State Auditor

January 31, 2019
2018‑113
The Governor of California
President pro Tempore of the Senate
Speaker of the Assembly
State Capitol
Sacramento, California 95814
Dear Governor and Legislative Leaders:
As requested by the Joint Legislative Audit Committee, the California State Auditor presents this audit
report concerning the effectiveness of in-prison rehabilitation programs at the California Department of
Corrections and Rehabilitation (Corrections).
This report concludes that inmates who completed in-prison cognitive behavioral therapy (CBT) programs
recidivated at about the same rate as inmates who did not complete the programs. These results are serious
enough to highlight an urgent need for Corrections to take a more active and meaningful role in ensuring
that these programs are effective. In particular, Corrections has not revalidated the accuracy of the tools it
uses to assess inmates’ rehabilitative needs since recent statutory changes caused a major shift in the State’s
prison population. Inaccurate assessment tools could result in placing inmates in the wrong programs or
in no programs at all. Furthermore, Corrections has not ensured that all of its CBT class curricula are
evidence based, resulting in a significant portion of inmates that do not receive treatment that has been
proven effective in reducing recidivism. Addressing these two problems would help Corrections ensure that
rehabilitation programs are meeting their primary purpose of reducing recidivism.
Moreover, Corrections has neither consistently placed inmates on waiting lists for needed rehabilitation
programs nor prioritized those with the highest need correctly. This contributed to Corrections' failure to
meet any of the rehabilitative needs for 62 percent of the inmates released in fiscal year 2017–18 who had been
assessed as at risk to recidivate. One reason inmates may not be receiving needed rehabilitation programs
is that Corrections is having difficulty fully staffing its rehabilitation programs at all of its prisons. These
various issues have resulted in low inmate enrollment rates when compared to the programs’ budgeted
capacity at the three prisons we reviewed.
Finally, Corrections has neither developed any performance measures for its rehabilitation programs, such
as a target reduction in recidivism, nor assessed program cost-effectiveness. Moreover, Corrections has not
analyzed whether its rehabilitation programs reduce recidivism. To perform such an analysis, Corrections
needs to collect additional data and take steps to ensure it delivers CBT programs as intended across all of
its facilities. Although Corrections plans to coordinate with external researchers to conduct a performance
evaluation of the rehabilitation programs over the course of the next two years, Corrections has taken no
formal steps to initiate this process. Because the Legislature provided Corrections with a significant budget
increase so that it could expand rehabilitation programs to all prisons in the State, it is vital that Corrections
demonstrate that the additional investment was worthwhile. To this end, the Legislature should implement
new accountability mechanisms related to Corrections’ rehabilitation programs, including additional
oversight, performance targets, and recidivism evaluations conducted by an external researcher.
Respectfully submitted,

ELAINE M. HOWLE, CPA
California State Auditor

621 Capitol Mall, Suite 1200

|

Sacramento, CA 95814

|

916.445.0255

|

916.327.0019 fax

|

w w w. a u d i t o r. c a . g o v

iv

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Selected Abbreviations Used in This Report
board

Board of State and Community Corrections

Corrections

California Department of Corrections and Rehabilitation

DOJ

California Department of Justice

CalPIA

California Prison Industry Authority

C‑ROB

California Rehabilitation Oversight Board

EDD

Employment Development Department

Folsom

Folsom State Prison

Inspector General

Office of the Inspector General

PPIC

Public Policy Institute of California

R. J. Donovan

Richard J. Donovan Correctional Facility

San Quentin

San Quentin State Prison

UC Irvine

University of California, Irvine

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Contents
Summary	1
Introduction	5
Corrections’ Implementation of Certain Rehabilitation Programs
Has Not Resulted in Demonstrable Reductions in Recidivism	

13

Corrections Is Failing to Place Inmates Into Appropriate
Rehabilitation Programs, Leading to Inmates Being Released
From Prison Without Having Any of Their Rehabilitation Needs Met	

23

Additional Oversight Is Needed to Ensure the Effectiveness of
Corrections’ Rehabilitation Programs	

35

Appendix A
Technical Appendix: Data and Methodology 	

49

Appendix B
Scope and Methodology 	

53

Appendix C
The Number of Inmates on Waiting Lists and the Average
Time Spent on the List	57
Responses to the Audit
California Prison Industry Authority	

59

California Department of Corrections and Rehabilitation	

61

California State Auditor’s Comments on the Response From
the California Department of Corrections and Rehabilitation	
California Rehabilitation Oversight Board	

63
65

v

vi

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Blank page inserted for reproduction purposes only.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

SUMMARY
Although the number of inmates housed in state prisons has decreased in recent years,
recidivism rates for inmates in California have remained stubbornly high, averaging
around 50 percent over the past decade. The State defines recidivism as when a person
is convicted of a subsequent crime within three years of being released from custody.
Research shows that rehabilitation programs can reduce recidivism by changing inmates’
behavior based on their individual needs and risks. For example, inmates are more likely
to recidivate if they have drug abuse problems, have trouble keeping steady employment,
or are illiterate. Rehabilitation programs aim to address and mitigate those challenges.
In 2012 the California Department of Corrections and Rehabilitation (Corrections)
released a report, commonly known as the blueprint, that set a number of goals, including
increasing access to rehabilitation programs.1 Since 2012 Corrections has expanded
cognitive behavioral therapy (CBT), vocational education, and academic education to all of
its 36 prisons, including a corresponding increase in its budget for in‑prison rehabilitation
programs, from $234 million in fiscal year 2013–14 to $298 million in fiscal year 2018–19.
Corrections has also begun administering the tools it uses to assess rehabilitative needs
for a greater number of inmates, and has created ways to better ensure its CBT vendors are
providing services consistently and efficiently. However, since this expansion, Corrections
has not undertaken sufficient effort to determine whether these programs are effective at
reducing recidivism.

Corrections’ Implementation of Certain Rehabilitation Programs
Has Not Resulted in Demonstrable Reductions in Recidivism
Our analysis of inmates released from prison in fiscal year 2015–16
did not find an overall relationship between inmates completing
CBT rehabilitation programs and their recidivism rates. In fact,
inmates who completed their recommended CBT rehabilitation
programs recidivated at about the same rate as inmates who were not
assigned to those rehabilitation programs. One potential reason why
our overall analysis did not find that CBT rehabilitation programs
are related to reductions in recidivism is that Corrections has
not revalidated the accuracy of the tools it uses to assess inmates’
rehabilitative needs since recent statutory changes caused a major
shift in the State’s prison population. Another potential reason is that
Corrections has not ensured that vendors provide consistent and
effective CBT programs that have been proven through research to
reduce recidivism—otherwise known as evidence based. Specifically,
we reviewed contracts for vendors that provided CBT classes at 10 of
Corrections’ 36 prisons and found that nearly 20 percent of their
respective curricula were not evidence based.

1	

Corrections first released The Future of California: A Blueprint to Save Billions of Dollars, End Federal Oversight, and Improve the
Prison System in 2012, and updated it in 2016.

Page 13

1

2

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Page 23

Corrections Is Failing to Place Inmates Into Appropriate
Rehabilitation Programs, Leading to Inmates Being Released From
Prison Without Having Any of Their Rehabilitation Needs Met
Corrections has neither consistently placed inmates on waiting lists
for needed rehabilitation programs nor prioritized those with the
highest need correctly. This has contributed to Corrections' failure
to meet any of the rehabilitative needs for 62 percent of the inmates
released in fiscal year 2017–18 who had been assessed as at risk
of recidivating. One reason inmates may not be receiving needed
rehabilitation programs is that Corrections is having difficulty fully
staffing its vocational and academic rehabilitation programs at
all of its prisons. These various issues have resulted in low inmate
enrollment rates when compared to the programs’ budgeted capacity
at the three prisons we reviewed.

Page 35

Additional Oversight Is Needed to Ensure the Effectiveness of
Corrections’ Rehabilitation Programs
Corrections has neither developed any performance measures for
its rehabilitation programs, such as a target reduction in recidivism,
nor has it assessed program cost‑effectiveness. Further, Corrections
has not analyzed whether its rehabilitation programs reduce
recidivism. To perform such an analysis, Corrections needs more
time to collect data and take steps to ensure that it delivers CBT
programs as intended across all of its facilities. Although the Office
of the Inspector General and the California Rehabilitation Oversight
Board (C-ROB) perform some limited oversight of Corrections’
rehabilitation programs, neither is well suited to conduct the analysis
needed to determine whether those programs are effective at
reducing recidivism or are cost‑effective. Although Corrections plans
to coordinate with external researchers to conduct a performance
evaluation of the rehabilitation programs over the course of the
next two years, Corrections has taken no formal steps to initiate
this process. Because the Legislature provided Corrections with a
significant budget increase so that it could expand rehabilitation
programs to all prisons in the State, it is vital that Corrections
demonstrate that the additional investment was worthwhile.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Summary of Recommendations
Legislature
To ensure that Corrections’ rehabilitation programs reduce
recidivism, the Legislature should require Corrections to establish
performance targets, including ones for reducing recidivism and
determining the programs’ cost‑effectiveness, and to partner
with external researchers to evaluate the effectiveness of its
rehabilitation programs.
Corrections
To ensure that Corrections has reliable tools for assessing the needs
of its inmate population, it should validate its assessment tools.
To ensure that its CBT classes are effective at reducing recidivism,
Corrections should provide adequate oversight to ensure that its
vendors teach only evidence‑based curricula.
To ensure that it can meet the rehabilitation needs of its inmates,
Corrections should develop and begin implementing plans to meet
its staffing‑level goals for rehabilitative programming.
To ensure that Corrections effectively and efficiently allocates
resources to reduce recidivism, it should partner with a research
organization to conduct a systematic evaluation to determine
whether its rehabilitation programs are reducing recidivism and if
they are cost‑effective.
Agency Comments
Corrections agrees with our findings and will address the specific
recommendations in a corrective action plan within the timelines
outlined in the audit report. C-ROB and the California Prison
Industry Authority also agreed with our recommendations.

3

4

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Blank page inserted for reproduction purposes only.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

INTRODUCTION
Background
The California Department of Corrections and Rehabilitation (Corrections) is responsible
for protecting the public by safely and securely supervising adult and juvenile inmates,
providing effective rehabilitation and treatment, and integrating inmates successfully back
into their communities. Corrections operates three adult women’s prisons and 33 adult men’s
prisons across the State,2 and it housed 116,400 male inmates and 5,200 female inmates as of
September 2018.
In the early 2000s, California’s prison system faced a crisis due to overcrowding and a budget that
was increasing at an unsustainable rate, from $5 billion in fiscal year 2000–01 to $9 billion in fiscal
year 2010–11. Additionally, in 2011 the U.S. Supreme Court upheld an order requiring Corrections
to reduce its inmate population to 137.5 percent of the prisons’ design capacity in an effort to
ensure that it provided mental health and medical treatment that met constitutional standards. The
Legislature later passed Assembly Bill 109—the 2011 Realignment Legislation (realignment), which
generally shifted the responsibility for incarcerating lower‑level felons convicted of nonviolent,
nonserious, and non‑sex‑related crimes from the State to the counties.
Although the number of inmates housed in state prisons has decreased since realignment,
the recidivism rate for inmates in California has remained stubbornly high, averaging
about 50 percent from fiscal years 2002–03 through 2012–13, as shown in Figure 1.
Figure 1
Corrections’ Recidivism Rate Averaged Around 50 Percent From Fiscal Years 2002–03 Through 2012–13
100%

Recidivism Rate

90

(Percentage of inmates who were released from state prison and then
subsequently convicted of a new crime within three years of release)

Recidivism Percentage Rate

80
70
60
50
40
30
20
10
0
2002–03

2003–04

2004–05

2005–06

2006–07

2007–08

2008–09

2009–10

2010–11

2011–12

2012–13

Fiscal Year
Source:  Analysis of Corrections’ 2017 Outcome Report: An Examination of Offenders Released in Fiscal Year 2012–13.

2	

Folsom State Prison (Folsom) houses both adult men and adult women in different prison facilities. Because their rehabilitation programs
are separate, we treated the two facilities as separate prisons in our analysis.

5

6

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Recent Policy Changes Have Reduced the
State Prison Population
In recent years, the Legislature and voters enacted various
constitutional and statutory changes that significantly
changed the composition of the State’s inmate population.
Some of the major changes include the following:
•	 Realignment (2011): Realignment limited who could be
sent to state prison. Specifically, it required that certain
lower‑level offenders serve their incarceration terms in
county jail. Additionally, it required that counties, rather
than the State, supervise certain lower‑level offenders
released from state prison.
•	 Proposition 36 (2012): Proposition 36 reduced prison
sentences for certain offenders subject to the State’s
existing three‑strikes law whose most recent offenses
were nonserious, nonviolent felonies. It also allowed
certain offenders serving life sentences to apply for
reduced sentences.
•	 Proposition 47 (2014): Proposition 47 reduced penalties for
certain offenders convicted of nonserious and nonviolent
property and drug crimes from felonies to misdemeanors.
It also allowed certain offenders who had been previously
convicted of such crimes to apply for reduced sentences.
•	 Proposition 57 (2016): Proposition 57 expanded inmate
eligibility for parole consideration, increased the State’s
authority to reduce inmates’ sentences due to good
behavior and/or the completion of rehabilitation programs,
and mandated that judges determine whether youth
should be subject to adult sentences in criminal court.
Source:  Analysis of state law, regulations, and documents
related to propositions.

The National Institute of Justice describes recidivism as a
person’s relapse into criminal behavior, often after the person
receives sanctions or undergoes intervention for a previous
crime. In California, state law required the Board of State and
Community Corrections (board) to develop the State’s
definition of recidivism. In 2014 the board defined recidivism as
a conviction for a new felony or misdemeanor committed
within three years of release from custody or committed within
three years of placement on supervision for a previous criminal
conviction. The reduction in recidivism in fiscal year 2012–13
shown in Figure 1 is likely due to the passage of Proposition 47
in 2014, which reduced the classification of certain nonserious
and nonviolent property and drug crimes from felonies to
misdemeanors, thus reducing the penalties for those crimes.
Although the proposition passed in 2014, it likely kept some
individuals who were released in fiscal year 2012–13 from
returning to prison within three years. Despite the small
reduction, recidivism remains a serious issue. According to a
study referenced in the National Institute of Justice’s website,
California has the 13th highest recidivism rate in the country.
As described in the text box, major policy changes, including
Proposition 47, have reduced the State’s prison population
and the amount of time inmates serve, thus increasing the
importance of effective rehabilitation programs. As a result of
these key policy changes, a greater majority of inmates currently
housed in state prisons are now classified as serious and violent
offenders. Specifically, according to Corrections’ demographic
reports, the proportion of inmates in custody for crimes against
persons, which tend to be more serious than other crimes, grew
from 59 percent in 2010 to 76 percent in 2017.
Corrections Has Increased Its Focus on Rehabilitation Programs

Following realignment, Corrections began increasing inmates’ access to in‑prison
rehabilitation programs to meet their rehabilitative needs. Research shows
that rehabilitation programs can reduce recidivism by changing inmates’ behavior
based on their individual needs and risks. For example, inmates are more likely to
recidivate if they have drug abuse problems or have trouble keeping steady employment.
Rehabilitation programs aim to address and mitigate those challenges.
In 2012 Corrections released a report, commonly known as the blueprint,3 that set a
number of goals, including increasing access to rehabilitation programs in order to
meet the needs of inmates before their release. The Legislature subsequently provided

3	

Corrections first released The Future of California: A Blueprint to Save Billions of Dollars, End Federal Oversight, and Improve the
Prison System in 2012, and updated it in 2016.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Corrections with a significant increase in funding to provide increased
inmate access to rehabilitative services. As shown in Figure 2,
Corrections’ budget for in‑prison rehabilitation programs increased
by $64 million, or nearly 30 percent, between fiscal years 2015–16
and 2016–17. Corrections used this increase in funding to expand its
cognitive behavioral therapy (CBT) reentry programs—designed to
correct an inmate’s patterns of thinking and behavior—to all of its
36 prisons, as shown in Table 1 on the following page.
Figure 2
The Budget for Rehabilitation Programs Increased by $64 Million
From Fiscal Years 2013–14 Through 2018–19
(In Millions)
$350

$302

$293 $298

2016–17

2017–18

300

Rehabilitation Programs Budget

250

$234 $235 $238

200

150

100

50

0
2013–14

2014–15

2015–16

2018–19

Fiscal Year
Source:  Analysis of fiscal year 2015–16 Governor’s Budget, fiscal years 2016–17 through 2018–19
budget acts, and Corrections’ budget documents.
Note:  We excluded costs associated with Corrections’ community‑based programming and its
administration of rehabilitation programs from this analysis.

7

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Table 1
Corrections Has Increased Its Rehabilitation Programs to All of Its 36 Prisons
FISCAL YEAR
NUMBER OF PRISONS THAT PROVIDE
REHABILITATION PROGRAMS FOR:*

COGNITIVE
BEHAVIORAL
THERAPY

8

2010–11

2011–12

2012–13

2013–14

2014–15

2015–16

2016–17

2017–18

Academic education

33

33

34

35

36

36

36

36

Vocational education

29

29

33

35

36

36

36

36

Substance abuse disorder treatment

12

12

12

14

24

24

36

36

Anger management, family relationships,
criminological thinking

0

0

0

12

14

14

36

36

Source:  Analysis of Corrections’ budget reports for fiscal years 2010–11 through 2017–18.
*	 We excluded programs designated for inmates serving lengthy periods of incarceration, such as life with the possibility of parole.

For fiscal year 2018–19, the Legislature appropriated $298 million
for rehabilitation programs, which equates to 2 percent of
Corrections’ overall budget, as shown in Figure 3. In comparison,
rehabilitation funding made up 1 percent, or $127 million, of
Corrections’ budget in fiscal year 2011–12. A majority of the funding
increase related to the expansion of CBT programs. Specifically, in
fiscal year 2016–17, the budget for in‑prison rehabilitation programs
increased from $238 million to $302 million, with 70 percent of
this increase going toward in‑prison programs, including CBT
programs and support. According to Corrections’ reports, it
had budgeted capacity capable of providing
rehabilitation opportunities for up to 130,000
CalPIA Provides Vocational Training Programs
inmates in its academic education, vocational
in Nine Different Areas
education, CBT, and other programs in fiscal
year 2017–18.
•	Carpentry*
•	 Iron working*
•	 Construction labor*
•	 Commercial diving
•	 Facilities maintenance
•	 Computer‑aided design
•	 Computer coding
•	Culinary
•	Roofing*
Source:  Analysis of the September 2017 and April 2018
agreements between Corrections and CalPIA, CalPIA’s union
agreements, and CalPIA’s PIA Programs by Institution report
dated June 30, 2018.
*  Union affiliated.

In addition to the training provided by
Corrections, the California Prison Industry
Authority (CalPIA) provides vocational education
to another 580 inmates. A majority of CalPIA’s
resources are devoted to its prison industry
programs, such as manufacturing license plates
or furniture, but CalPIA also provides vocational
programs in a total of nine fields, as shown in
the text box. Inmates are generally eligible for a
CalPIA vocational program interview if they are
designated as a minimum or medium security
level, exhibit good behavior, do not currently
have a need for substance abuse treatment,
and have not been sentenced to life without
parole. Currently, Corrections has contracted
with CalPIA to establish and manage vocational

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

programming through fiscal year 2019–20, at a cost of $12.4 million.
CalPIA partners with unions to provide curriculum and instructors
for four of these programs—carpentry, iron working, construction
labor, and roofing. After they complete these programs, inmates
become eligible for each union’s apprenticeship program
upon release.
Figure 3
Rehabilitation Funds Make Up a Small Portion of Corrections’
Fiscal Year 2018–19 Budget
(Dollars in Millions)

$12,149

CORRECTIONS’ TOTAL BUDGET

98%

$11,851

Non-rehabilitation budget

2%
26%
52%

$78

CBT and
Transitions
programs*

$298

Rehabilitation programs’ budget

18%
4%

$55

Vocational education

$154

Academic education

$11

Volunteer programs†

Source:  Analysis of fiscal year 2018–19 Budget Act, and Corrections’ budget documents.
Note:  We excluded costs associated with Corrections’ community‑based programming and its
administration of rehabilitation programs from this analysis.
*	 The Transitions Program equips inmates with job search skills and financial literacy to help them
reintegrate into society once released.
†	 Volunteer programs include educational, social, cultural, and recreational activities provided by
volunteers or nonprofits. Programs can include Alcoholics Anonymous, yard time literacy, or yoga.

9

10

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Corrections Operates a Variety of Rehabilitation Programs
Within Corrections, the Division of Rehabilitative Programming
(division) is generally responsible for administering the
rehabilitation programs that Corrections provides to inmates.4
The division oversees programs at all 36 state prisons, including
programs run by Corrections staff, such as adult education and
vocational training, and programs such as CBT run by contract
staff. Additionally, inmates can participate in programs that
Corrections oversees but that are run by volunteers, such as
Alcoholics Anonymous and yoga.
Corrections determines what in‑prison rehabilitation programs
inmates need through assessments that it requires inmates to take
upon entering an institution. As shown in Figure 4, an inmate’s score
on these assessments indicates the type of rehabilitation program
that will address his or her needs. The first assessment, known
as Correctional Offender Management Profiling for Alternative
Sanctions (COMPAS), measures an inmate’s need for CBT and
vocational education. A second assessment, known as the Tests
of Adult Basic Education (TABE), measures an inmate’s need for
academic education programs. As shown in Table 2 on page 12,
Corrections gives priority for enrollment in rehabilitation programs
to inmates who have both a moderate to high risk of recidivating and
a moderate to high need for a rehabilitation program. Corrections
determines inmates’ risk of recidivating according to their California
Static Risk Assessment (CSRA) score, which is derived from
their prior criminal history. If inmates receive a moderate to high
COMPAS score in any of the four CBT categories—substance abuse,
criminal thinking, anger management, or family relationships—
Corrections places them on a waiting list for the class or classes that
address their rehabilitative needs. In addition, state law requires
Corrections to offer academic programming throughout an inmate’s
period of incarceration, focused on increasing the inmate’s reading
ability to at least a ninth‑grade level. Corrections also requires
inmates to participate in substance abuse disorder treatment if it
determines that they have a history of drug abuse.
Inmates enter the prison system through one of six institutions
known as reception centers. Correctional counselors at the
reception centers evaluate the inmates through assessments
such as COMPAS and TABE, and retrieve the inmate’s CSRA
score, which is automatically developed based on the inmate’s
demographics and criminal history using California Department
of Justice (DOJ) data. Medical staff also administer medical and

4	

Other departments within Corrections are responsible for administering smaller
rehabilitation programs.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

psychological evaluations. Corrections’ classification staff use the
assessments and input from the correctional counselors to assign
the inmates to a home institution. Within two weeks of arrival at
their assigned prison, inmates meet with a correctional counselor
who uses the assessments to recommend a course of treatment
to a classification committee. The classification committee
evaluates the inmate’s case factors, such as the assessments and
the counselor’s recommendations, and places the inmate on
appropriate waiting lists for in‑prison work or rehabilitation
programs. Inmate assignment officers place inmates who are on
waiting lists into in‑prison jobs or rehabilitation programs as those
opportunities become available. Inmates have a strong incentive
to participate because they can earn credits for early release by
participating in approved rehabilitation programs or by attaining
educational achievements.
Figure 4
Corrections Uses Inmate Assessment Scores to Assess an Inmate’s Need for Rehabilitation Programs

Inmate Assessments
Tests of Adult
Basic Education
(TABE)

Correctional Offender Management
Profiling for Alternative Sanctions
(COMPAS)

Substance
abuse
score

anger/
hostility
score

criminal
personality
score

Substance
abuse
disorder
treatment

anger
management

criminal
Thinking

Support from
family of origin
score

Employment
problems
score

Family
Relationships

Career
Technical
Education
Programs

General
population
academic
education

Vocational
education

Academic
education

Cognitive behavioral therapy (CBT)

In-Prison Rehabilitation Programs
Source:  Analysis of various Corrections policies and staff interviews.
*	 We excluded programs designated for inmates serving lengthy periods of incarceration, such as life with the possibility of parole.

Tabe
reading
score

11

12

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Table 2
Corrections Uses Assessment Scores to Identify Inmates With the Highest Risk to Recidivate and the Highest
Rehabilitative Needs and Places Those Inmates at a Higher Priority
INMATE ASSESSMENT SCORES
ASSESSMENT TOOL /
RISK OR NEED ASSESSED

CSRA
Inmate risk to recidivate

CORRECTIONS PLACES INMATES WITH THE FOLLOWING
ASSESSMENT SCORES AT A LOWER PRIORITY FOR
REHABILITATION PROGRAMS

CORRECTIONS PLACES INMATES WITH THE FOLLOWING
ASSESSMENT SCORES AT A HIGHER PRIORITY FOR
REHABILITATION PROGRAMS

Low risk

Moderate to high risk

+ one of the following:
COMPAS
Inmate behavioral needs

Low need

Moderate to high need

or*

TABE
Inmate education needs

Reading score ninth‑grade level or above

Reading score zero to eighth‑grade level

Source:  Analysis of state law and Corrections policies and staff interviews.
*	 Corrections prioritizes inmates for CBT and vocational education programming if they have a moderate to high risk to recidivate based
on their CSRA score and a moderate to high need for the program based on their COMPAS score. Corrections prioritizes inmates for
academic education programming if they have a moderate to high risk to recidivate based on their CSRA score and a reading score
below a ninth‑grade level based on their TABE score.

Two Entities Provide Oversight of Corrections’
Rehabilitation Programs
Oversight of Corrections’ in‑prison rehabilitation programs is
provided by two entities. State law requires the Office of the
Inspector General (Inspector General) to periodically review
Corrections’ implementation of the reforms outlined in the
blueprint in addition to monitoring Corrections’ delivery of medical
care for inmates and overseeing its internal affairs investigations
into allegations of wrongdoing by Corrections staff. Further, the
California Rehabilitation Oversight Board (C‑ROB) issues an
annual report examining some specific aspects of Corrections’
rehabilitation programs. C‑ROB’s members consist of various
agency executives and private professionals with varying expertise
and responsibilities related to inmate treatment. C‑ROB meets
at least twice a year to discuss the effectiveness of Corrections’
rehabilitation programs. However, for reasons described later in
this report, oversight of Corrections’ rehabilitation programs by
these entities is limited.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Corrections’ Implementation of Certain
Rehabilitation Programs Has Not Resulted in
Demonstrable Reductions in Recidivism
Key Points
•	 Our analysis suggests that inmates who completed their recommended CBT
rehabilitation programs recidivated at about the same rate as inmates who did
not complete their recommended rehabilitation programs, and we were unable
to identify a relationship between completing rehabilitation programs and
recidivism rates.
•	 Corrections needs to revalidate the tools it uses to assess inmates’ rehabilitative needs
to ensure that they are still accurate, given the change in the State’s prison population
following realignment.
•	 Corrections has not conducted sufficient oversight to ensure that its vendors provide
evidence‑based and researched CBT programs.
Corrections’ CBT Rehabilitation Programs Do Not Appear to Have Reduced Recidivism
Because of the financial cost and societal impact of a recidivating inmate, the State has a
vested interest in ensuring that rehabilitation programs are meeting their primary goal
of reducing recidivism, particularly in light of the fact that the State has significantly
increased its investment in these programs over the past five years, as noted in the
Introduction. However, our analysis of data for inmates who received rehabilitation
programs suggests that there was no overall significant connection between an inmate
completing these programs and the inmate’s likelihood to recidivate. These results,
although somewhat constrained by data limitations that we will discuss later, are serious
enough to highlight an urgent need for Corrections to take a more active and meaningful
role in ensuring that these programs are effective.
To determine whether in‑prison rehabilitation programs reduced recidivism rates
for the inmates who completed them, we began by making three key decisions. First,
although the board defines recidivism as a conviction for a new felony or misdemeanor
committed within three years of release from custody, we limited our analysis to one‑
and two‑year recidivism rates. We did this because Corrections has accurate data on
program participation only beginning in October 2014, when it implemented a new
data system. Because CBT programs expanded to 11 male prisons in fiscal year 2014–15,
we also gave inmates time to take CBT classes before they were released from prison
during fiscal year 2015–16. Second, we focused our analysis on two groups of inmates:
those who received four types of CBT programs—substance abuse disorder treatment,
anger management, criminal thinking, and family relationships—and those who did not

13

14

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

complete any CBT at the 11 prisons that offered CBT during the
period we examined.5 We focused our analysis on CBT because
the majority—70 percent—of the State’s recent expansion of its
rehabilitation programs budget was solely for expanding CBT to all
prisons. Third, we considered inmates as having their needs met if
they completed at least half of the CBT programs that Corrections
determined they needed. Conversely, we considered inmates as
having no needs met if they were not assigned to any CBT programs
that Corrections determined they needed. Appendix A beginning
on page 49 includes the detailed methodology used in our analysis
and its results and limitations.
As shown in Table 3, the recidivism rate for inmates in the group
who had most of their needs met was similar to that of the
group with inmates who had no needs met. In fact, the two‑year
recidivism rates for both groups of inmates were between 24 and
25 percent. Our review analyzed inmates who were released in
fiscal year 2015–16 and, using data from DOJ, determined whether
those individuals were subsequently convicted of a misdemeanor or
felony within two years of release.
Table 3
Recidivism Rates Generally Did Not Vary Between the Two Inmate Groups
We Examined
RECIDIVISM RATE
TIME ELAPSED
SINCE RELEASE

MOST CBT NEEDS MET

NO CBT NEEDS MET

1 year

12%

13%

2 year

25%

24%

Source:  Analysis of Strategic Offender Management System (SOMS) and DOJ data.
*	 We excluded programs designated for inmates serving lengthy periods of incarceration, such as
life with the possibility of parole.

Many factors influence an inmate’s likelihood to recidivate, such
as education, race, age, and crime risk.6 Failing to consider factors
such as these when conducting an analysis of recidivism has the
potential to undermine the results. For example, Corrections’
2017 Outcome Evaluation Report indicates that inmates over
age 55 are significantly less likely to recidivate than inmates
who are 25 or younger. Thus, a detailed analysis should consider
5	

We excluded programs designed for inmates serving lengthy periods of incarceration, such as life
with the possibility of parole.

6	

Corrections assesses crime risk based on an inmate’s CSRA score, which produces a risk number
that predicts the likelihood that an offender will recidivate for a certain type of crime.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

inmates’ ages, because each additional inmate over age 55 decreases
the recidivism rate irrespective of the programs those inmates
complete. To ensure that our results were not skewed in this
manner—either by education, self‑identified race, age, crime risk,
or prison—we performed additional analyses and found that, even
when taking the impact of these factors into account, we identified
no significant overall relationship between completing CBT
programs and reduced recidivism.
We conducted a more focused analysis of our inmate groups
by examining whether completing CBT classes was related to
reductions in the two‑year recidivism rate within each of the
four categories—education, self‑identified race, age, and crime
risk. For example, in examining the relationship between inmates
with different levels of crime risk, we found that the group of
inmates that showed the strongest relationship between completing
CBT classes and a decrease in recidivism rate were those with
a high risk of committing violent crimes. Specifically, for this
group, completing assigned CBT classes was strongly related to a
16 percent decrease in recidivism. For other inmate groups in which
there appears to be a relationship between completing CBT classes
and reduced recidivism, see Table A beginning on page 51.
We also analyzed our inmate groups to determine whether there was
a relationship between inmates completing CBT classes in a particular
prison and recidivism. Although we did not find a significant
relationship in overall recidivism rates across all 11 prisons, our results
indicated that inmates incarcerated at one prison had a significantly
stronger relationship between completing CBT classes and their risk
of recidivating. Specifically, it appears that at the Substance Abuse
Treatment Facility, located in Kings County, inmates who completed
CBT classes had a recidivism rate that was 18 percent lower than for
inmates who did not complete these classes. A prison administrator
stated that during the time period of our analysis they were doing
several things that may have made their CBT classes more effective.
The administrator cited an immersive CBT environment, in which
inmates taking CBT classes were housed together, and incentives
for the inmates, such as individual cells, as potential reasons for the
lower recidivism rates. In addition, the prison offers peer mentoring
programs in which inmates with high social standing in the prison act
as peer mentors in the CBT classes.
The deputy director of Corrections’ Department of Rehabilitation
Programs (deputy director) stated that, while disappointing, he
accepts the results of our analysis. He believes our results may be
a reflection of the fact that in‑prison treatment alone does not
adequately impact recidivism. He further stated that a statistical
model that included other in‑prison rehabilitation programs
beyond CBT, as well as follow‑up post‑release treatment in

15

16

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

the community, may yield positive results. Although it is unclear
whether a more comprehensive study would definitively show that
rehabilitation programs were reducing recidivism, based on our
analysis it is evident that Corrections should take substantive actions
to help ensure that its programs are effective. As we discuss later in
this report, rehabilitation programs need to reduce recidivism by
only a small amount to become cost‑neutral.
Corrections Should Revalidate Its Rehabilitative Assessment Tools and
Develop Procedures for Future Periodic Validations
One potential reason why our overall analysis did not find that
rehabilitation programs are related to reductions in recidivism is
that Corrections has not recently evaluated how well it is assessing
inmates’ needs and risks before placing them in rehabilitation
programs. Ineffective assessments could result in Corrections
failing to place inmates in the programs that would most effectively
reduce their risk of recidivating. Inaccurate assessments might place
inmates in the wrong programs or no programs at all, increasing
their chances of recidivating. State law requires Corrections to
examine and study all pertinent circumstances of an inmate’s life
that caused him or her to violate the law and be committed to
prison. State law, regulations, and Corrections’ practices further
require Corrections to administer assessments to all inmates during
the reception process or during their annual review to determine
their rehabilitation needs and risk of recidivating.

Ineffective assessments could result in
Corrections failing to place inmates in the
programs that would most effectively reduce
their risk of recidivating.

Specifically, Corrections uses COMPAS to assist in determining an
inmate’s placement in a rehabilitation program and gives a CSRA
score to all inmates to determine their risk of recidivating. CSRA
is an assessment tool that uses a set of risk factors, including age,
gender, and criminal convictions, to determine inmates’ recidivism
risk. Inmates’ COMPAS scores are used in conjunction with their
CSRA risk scores to determine what rehabilitation programs are
needed. Finally, Corrections uses TABE to assess inmates’ general
academic ability. Although an outside entity validated TABE to

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

ensure its accuracy in 2017, COMPAS and CSRA have not been
recently reviewed and may no longer accurately assess inmates’
rehabilitative needs and risk of recidivating.
Specifically, Corrections has not validated COMPAS since 2010, when
it used data for inmates that were released over 10 years ago, from fiscal
years 2005–06 through 2008–09, when conducting the validation.
Further, CSRA was last validated in 2013 using data for inmates that
were released 16 years ago, in fiscal year 2002–03. The validation of
COMPAS and CSRA confirmed that they appropriately identified the
needs and recidivism risks of inmates who were representative of
the inmate population in prison at that time. The validations included
Corrections testing whether COMPAS yielded the same rehabilitative
needs scale results when given to the same inmate multiple times and
whether applying CSRA to released inmates adequately predicted
their risk of recidivating. The validations also evaluated whether the
needs scale that COMPAS produced—low need, moderate need, and
high need—corresponded to inmates’ future recidivism rates.
However, because the studies validated the needs and risks of inmates
who were in prison at least 10 years ago, COMPAS and CSRA have
not accounted for the drastic changes in the prison population since
realignment. As discussed in the Introduction, the number of inmates
in custody for crimes against persons, which tend to be more serious
than other crimes, increased from 59 percent in 2010 to 76 percent
in 2017. Statistical researchers have indicated that assessment tools—
such as COMPAS and CSRA—may become unreliable and inaccurate
if not validated periodically, especially when there is significant change
in the target population.

COMPAS and CSRA have not accounted for
the drastic changes in the prison population
since realignment.

These assessment tools have not been validated on Corrections’ more
violent inmate population and thus may be failing to place inmates in
appropriate programs that reduce their recidivism risk.
Moreover, although the validation studies that Corrections last
conducted for COMPAS and CSRA concluded that the assessment
tools were generally sufficient, they both highlighted minor issues
related to their assessment of violent offenders. Specifically, the results
of the COMPAS validation showed that although the assessment
was generally adequate to predict recidivism, it failed to meet the

17

18

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

minimum risk prediction threshold for determining the specific
likelihood of inmate recidivism by violent crime. Similarly, the
results of the CSRA validation showed that although the assessment
was generally adequate to predict recidivism, the study identified a
relevant percentage of inmates assessed as low risk who ultimately
recidivated by committing violent crimes. The minor issues
identified in the COMPAS and CSRA validation studies further
support the need for Corrections to revalidate them, particularly in
light of Corrections’ inmate population changes.
According to the deputy director, revalidating COMPAS and CSRA
is a priority for Corrections, but Corrections has a number of other
initiatives that it wants to complete first, including implementing a
tool to ensure that its vendors deliver CBT programs consistently
and effectively across all prisons. Furthermore, Corrections
convened a workgroup in November 2018 to engage in further
discussions on assessment tools other than COMPAS or CSRA
to determine whether they would better measure inmates’
rehabilitation needs and risks. According to the deputy director,
Corrections does not have any formal procedures or guidelines that
delineate standards for when assessment tools need revalidation.
On a positive note, Corrections has made strides in administering
assessments to a greater number of inmates. According to
the Blueprint Monitoring Report published periodically by the
Inspector General, Corrections has increased the percentage of
inmates who have received these assessments from 44 percent
of the total inmate population in fiscal year 2012–13 to 93 percent of
the inmate population eligible to receive a COMPAS assessment
in fiscal year 2017–18. The Inspector General also reported a
slight increase in inmates assessed with a CSRA score, from
96 percent in fiscal year 2011–12 to 98 percent in fiscal year 2017–18.
Additionally, Corrections recorded a TABE score for 94 percent of
the prison population in fiscal year 2017–18. Corrections’ increase
in administering the COMPAS assessment across the inmate
population helps it meet its regulatory requirement to administer
COMPAS to all eligible inmates and creates greater opportunity to
provide inmates with targeted rehabilitation. Although Corrections
still needs to make progress with respect to determining the
validity of its assessment tools and ensuring that all eligible
inmates receive those assessments, according to Corrections’
Outcome Evaluation Report, assessment tools alone cannot
reduce recidivism. Rather, their results need to be combined with
appropriate, evidence‑based treatment. However, as we discuss in
the next section, Corrections is also failing to ensure that inmates
receive evidence‑based treatment, thereby increasing their risk
of recidivating.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Corrections’ Lack of Oversight Has Resulted in Vendors Using CBT
Class Curricula That Have Not Been Evaluated to Ensure That They
Reduce Recidivism
Another potential reason why we generally did not find a
relationship between rehabilitation programs and reduced
recidivism is Corrections’ failure to ensure that all of its CBT class
curricula are evidence based, resulting in a significant portion of
inmates receiving a curriculum that has not been proven effective
in reducing recidivism. Researchers designate a curriculum as
evidence based when it has been evaluated using strong research
designs and has been shown to have a positive impact on the
program’s participants. Corrections recently relied on the Pew
Results First Clearinghouse Database (Pew)—a database that
contains an extensive list of CBT programs that researchers have
designated as evidence based. Corrections contracts all of the
teaching of the four elements of its CBT programs we reviewed—
anger management, family relationships, criminal thinking, and
substance abuse disorder treatment—to vendors. Each prison
contracts with a single vendor to provide the CBT programs,
and every contract we reviewed requires vendors to identify and
use evidence‑based curricula for their CBT programs.7 However,
according to the deputy director, the contracts do not require that
vendors use CBT programs listed as evidence based in Pew.
Due to a lack of oversight by Corrections, a significant portion
of the CBT curricula that have been taught at some of the
prisons we reviewed were not evidence based. Specifically, we
reviewed contracts for vendors that provided CBT classes at 10 of
Corrections’ 36 prisons and found that 17 percent of their CBT
curricula were not designated as evidence based, either by Pew
or by other sources provided by Corrections, as shown in Table 4
on the following page. Moreover, three prisons contracted with
vendors that used non‑evidence‑based CBT curricula in at least
one‑quarter of their classes, and one prison’s vendor did not use
evidence‑based CBT curricula in over half of its classes.
Furthermore, the number of CBT curricula that were evidence based
varied widely among prisons, and Corrections is unable to track
whether an inmate received evidence‑based curricula. For example,
in Pelican Bay State Prison, only one of 32 curricula (3 percent)
were not evidence based, while at Richard J. Donovan Correctional
Facility (R. J. Donovan) nine of the 16 curricula (56 percent) were
not evidence based. We analyzed these data to ascertain whether
the portion of evidence‑based curricula offered by a vendor had a
7	

Corrections' contracts allow it to require vendors to use curricula that is not evidence based.
However, Corrections did not exercise this option for any of the CBT curricula we reviewed or
discussed in this report.

19

20

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

relationship to the recidivism rates of the inmates those vendors
served, but data limitations precluded us from making conclusions
in this analysis. According to the deputy director, Corrections has
not historically monitored whether inmates receive evidence‑based
programs, though Corrections has recently changed that.
Table 4
Number of Evidence‑Based CBT Curricula Varies Across Corrections’ Prisons

PRISON

VENDOR

NUMBER OF
NON‑EVIDENCE‑BASED
CBT CURRICULA

PERCENT OF
NON‑EVIDENCE‑BASED
CBT CURRICULA

R. J. Donovan

Epidaurus dba Amity Foundation

9 of 16

56%

Centinela State Prison

Community Education Centers, Inc.

3 of 11

27

Central California Women's Facility

Epidaurus dba Amity Foundation

2 of 8

25

High Desert State Prison

Phoenix House of California, Inc.

6 of 27

22

California Men's Colony

WestCare California, Inc.

1 of 6

17

San Quentin State Prison (San Quentin)

Center Point Inc.

1 of 7

14

California Institute for Men

Center Point Inc.

1 of 7

14

Kern Valley State Prison

Geo Reentry Services, LLC

6 of 55

11

Folsom

HealthRIGHT 360, Inc.

1 of 11

9

Pelican Bay State Prison

WestCare California, Inc.

1 of 32

3

31 of 180

17%

Total of all 10 prisons
Source:  Analysis of Corrections’ CBT curriculums and Pew’s Results First Clearinghouse Database.

Although Corrections collects information on the proportion of
curricula vendors offer that are evidence based, it has not imposed
sanctions on vendors that did not comply with their contracts.
Because these vendors’ identified and used CBT curricula that
was not all evidence based, the vendors are likely in violation of
their contracts. Moreover, Corrections’ contract language gives
it significant flexibility to sanction vendors for failing to comply
with contracts, including termination of the contract. However,
the deputy director stated that Corrections wants to ensure
that the system is not closed off from new curricula that may be
effective. Corrections has provided insufficient oversight to ensure
that its vendors at each prison used evidence‑based curricula,
thereby eliminating its ability to hold them accountable for failing
to provide evidence‑based curricula.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

According to the deputy director, Corrections plans to conduct
additional oversight and enforcement in the future. He stated that
the next round of CBT contracts will require each vendor to use
only curricula that are designated as evidence based by Pew and
indicated that Corrections will enforce that standard. He further
stated that Corrections will use the data it collects on vendors’
CBT curricula, as well as on‑site checks, to determine whether
vendors are abiding by the terms of their contracts by teaching only
evidence‑based curricula. If a vendor does not abide by the terms
of its contract, Corrections will initiate a disciplinary process,
beginning with a 30‑day corrective action plan and escalating to a
rebid of that contract if the vendor does not fix the issue.
Corrections has already taken some first steps to help ensure that
CBT vendors are complying with their contracts by commissioning
the University of California, Irvine (UC Irvine) to produce a
report that examined various existing industry best practices and
identified ways to strengthen Corrections’ contract compliance
process. The report examined Corrections’ oversight of vendors
responsible for conducting CBT programs and recommended
changes to the Program Accounting Review (program review)
used by Corrections to ensure that vendors are complying with
their CBT program contracts and that the programs are being
delivered as intended. Specifically, UC Irvine recommended that
the program review increase the number of group observations and
the number of participant interviews, and add questions to measure
factors such as the level of trust that class participants demonstrate.
These changes also included increasing the amount of time the
program analyst spends at the prison and conducting more staff
interviews to assess the program. Two prisons piloted the resulting
new program review in 2018. Corrections has stated that it agrees
with UC Irvine’s recommendations for improving contract
compliance and has adopted the new program review, beginning
in January 2019.
Recommendations
To ensure that Corrections has reliable tools for assessing the needs
of its inmate population, it should validate COMPAS and CSRA by
January 2020 and revalidate all of its assessment tools at least every
five years.
To ensure that Corrections is able to discover and prioritize the
most effective CBT rehabilitation curricula, it should begin using
its ability to record the individual CBT curricula inmates receive,
and then use this information in an analysis of its rehabilitation
programs in 2020.

21

22

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

To ensure that its CBT classes are effective at reducing recidivism,
Corrections should amend its CBT contracts to require vendors
to teach only evidence‑based curricula as designated by Pew
and should provide adequate oversight, including implementing
UC Irvine’s contract compliance recommendations, to ensure that
its vendors adhere to this standard by January 2020.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Corrections Is Failing to Place Inmates Into
Appropriate Rehabilitation Programs, Leading to
Inmates Being Released From Prison Without Having
Any of Their Rehabilitation Needs Met
Key Points
•	 Corrections has neither consistently placed inmates on waiting lists for needed
rehabilitation programs nor prioritized those with the highest need correctly. This
has contributed to Corrections' failure to meet any rehabilitative needs for 62 percent
of the inmates released in fiscal year 2017–18 who had been assessed as at risk
of recidivating.
•	 Corrections is having difficulty fully staffing its rehabilitation programs at all of
its prisons.
•	 Corrections’ rehabilitation programs at the three prisons we reviewed have lower
enrollment rates than their budgeted capacity.

Corrections Has Neither Placed Inmates on Program Waiting Lists Appropriately Nor Assigned
Inmates to the Programs Necessary to Address Their Rehabilitative Needs
Using Corrections’ inmate data, we determined that during fiscal year 2017–18 Corrections
did not meet its goal of providing rehabilitative programming to 70 percent of its target
population—inmates who have a moderate to high risk of recidivating and a moderate to
high need for rehabilitative programming prior to their release. Specifically, we found that
24,000 of the inmates released during fiscal year 2017–18 had a moderate to high risk of
recidivating and a moderate to high need for at least one rehabilitation program. Of those
inmates, only 9,000—or 38 percent—completed or were assigned to a program at some
point between October 2014, when SOMS went live, and the date the inmate was released
from prison. Thus, 62 percent of the inmates released in fiscal year 2017–18 who had been
assessed as having rehabilitative needs and having a risk of recidivating were released
without any of these needs having been met.
Although Corrections assesses the rehabilitative needs of inmates, it has not consistently
placed inmates onto waiting lists for rehabilitation programs or assigned inmates who are on
waiting lists to needed classes. Department policies require Corrections to give priority for
assignment to programs to inmates who are within two years of their release. Corrections’
policies and regulations further require its staff to administer COMPAS rehabilitative
assessments to inmates during the reception center process when first incarcerated, or
during the annual review process. Figure 5 on the following page depicts the assessment and
waiting list process. Corrections’ policies require that it place inmates onto rehabilitation
program waiting lists generally within five years of their scheduled release, depending on the
types of programs an inmate needs.

23

24

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Figure 5
State Law and Corrections' Policies Require Prisons to Assess Inmates’ Rehabilitation Needs and Attempt to
Address Those Needs Before Their Release Dates
Corrections
receives inmate
at prison
reception center

Corrections
completes
COMPAS and
other necessary
inmate
assessments*

Corrections
transfers inmate
from reception
center to
assigned prison

Classification
committee convenes
to discuss inmate
rehabilitation options;
may add inmate to
waiting list or assign
inmate to programs†

PRISON RECEPTION CENTER

While at
reception center
(0–90 days)

Corrections
suggested
deadline for
priority wait listing
and assigning
inmate to CBT
and vocational
programs

Corrections goal
is to provide
programming to
at least 70%
of the target
population‡

70%

Assigned PRISON

COMMITTEE
MEETING
IN PROGRESS

RECEPTION
Day 0

Corrections
prioritizes inmate
for placement onto
waiting list for CBT
rehabilitation
programs

Once approved by
classification staff
(90+ days)

By Day 14
following
arrival

CBT

priority

Within
5 years
of release

Within
4-12 months
of release

By release
date

Source:  Analysis of state law and Corrections’ policies and procedures.
Note:  Individual prisons may alter this timeline, if necessary, based on the needs of the prison.
*	 Corrections also performs COMPAS assessment during inmates' annual classification committee review process if they have not completed a
COMPAS assessment at the reception center.
†	 Corrections prioritizes inmates for placement into academic programs throughout their incarceration. Additionally, inmates can be prioritized for
vocational programs until they are within 12 months of release.
‡	 Corrections identifies inmates within its target population as inmates assessed with a moderate to high risk of recidivating and a moderate to
high rehabilitation program need.

Our detailed review of records for 60 inmates who were in prison
as of September 2018 showed that Corrections had failed to place
20, or 33 percent, of them onto waiting lists within five years of
their scheduled release.8 Of further concern is that 14 of these
20 inmates did not have enough time left to serve before their
scheduled release dates to complete their needed classes, making it
impossible for these inmates to receive the rehabilitation programs
they need.
Although Corrections’ policy requires prisons to place inmates onto
waiting lists within five years of release, it gives the correctional
counselors latitude as to when that placement actually occurs.
According to the deputy director, Corrections allows each prison,

8	

We excluded programs designed for inmates serving lengthy periods of incarceration, such as life
with the possibility of parole.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

through the classification committee and correctional counselors,
to independently determine when to place inmates on waiting lists.
As a result, some prisons immediately place inmates onto waiting
lists during the initial classification committee meeting, and other
prisons wait until inmates get closer to their expected release date.
However, if inmates are not on waiting lists, assignment officers—
local Corrections staff responsible for manually assigning inmates
from waiting lists into rehabilitation classes—may not be aware
that they are in need of a particular program and may fail to assign
them to one when space becomes available. In one example from
the 60 inmate records we reviewed, Corrections did not place an
inmate with a high rehabilitative need for substance abuse disorder
treatment on a waiting list for a substance abuse class, even though
the inmate was due to be released from prison in 2018 after being
incarcerated for more than three years. Because of Corrections’
inadequate oversight of correctional counselors at individual
prisons, inmates are not consistently being placed on waiting lists,
and as a result, some are not getting assigned to needed programs.
However, even when Corrections followed its own processes and
placed inmates on waiting lists for rehabilitation programs within
the appropriate time frame, it did not consistently assign all
inmates from the waiting lists to necessary rehabilitation classes
before their release from prison. Corrections’ policy requires
prisons to assign inmates to CBT classes within two years of
release, and to academic and vocational programs within four years
of release, but it gives assignment officers discretion as to when
that assignment actually occurs. Our detailed review of the
aforementioned 60 inmate records found that Corrections failed to
assign six inmates—10 percent—to the classes they needed to meet
their rehabilitative needs, even though they were on waiting lists for
those classes. Furthermore, at the time of our review none of the
six had enough time left in their respective sentences to complete
the courses they needed.
Of additional concern is that the CBT programs we reviewed could
accommodate additional inmates, as their vacancy rates average
24 percent. In one example, despite a 24 percent vacancy rate for
substance abuse programs during fiscal year 2017–18 at Folsom,
Corrections did not assign an inmate to the five‑month program,
even though the inmate was a high risk for recidivating, had a high
need for substance abuse programming, and was in prison for more
than a year. According to the Folsom correctional counselor III,
among the reasons that an inmate with a rehabilitative need
may not be assigned to a rehabilitation program are the inmate
being assigned to another program and concern that the inmate’s
enrollment in the class may be a security threat to the staff, other
inmates, or the prison.

25

26

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

In addition, the waiting lists that Corrections uses to prioritize
inmates’ placement into rehabilitation programs do not reflect the
priorities that Corrections identified in its blueprint. Specifically,
the blueprint prioritizes inmates with a moderate to high risk
of recidivating and a moderate to high need for rehabilitation
programs. However, in practice, Corrections is not prioritizing
inmates with a high risk to recidivate and a high need for the
rehabilitation programs. Moreover, when the correct inmates are
not assigned to rehabilitative programming, those inmates can be
denied the opportunity to earn rehabilitation program credits that
reduce their time left to serve.
We examined the records for a selection of 19 inmates placed into
rehabilitation programs during fiscal year 2017–18 at the three
prisons we reviewed—Folsom, San Quentin, and R. J. Donovan—
and found that 15 of the 19 should not have received priority for
enrollment. Specifically, of the 19 inmates, four had a low risk of
recidivating, four had a low need for the rehabilitation program
they were placed in, and seven had a combination of both a low
risk and a low need. In each of these instances, we found at least
22 other inmates at the prison who were not currently enrolled in
another rehabilitation program and who had a moderate to high
risk of recidivating, a moderate to high need for the program, and
a scheduled release date within the next five years. As shown in
Figure 6, Corrections assigned those 15 inmates ahead of a total
of more than 700 other inmates with higher risks and needs who
were not enrolled in another rehabilitation program. For example,
we reviewed the data for three inmates that Corrections placed
into vocational education programs at R. J. Donovan on the same
day in April 2018. None of them had a moderate to high need for
vocational education, and only one had a moderate to high risk of
recidivating. Further, when we reviewed the inmate population for
R. J. Donovan on that day, we found 89 inmates who should have
received higher priority because they were not currently enrolled
in another rehabilitation program and had a moderate to high risk
of recidivating, a moderate to high need for the program, and a
scheduled release date within the next five years.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Figure 6
Corrections Assigned 15 Inmates at the Three Prisons We Reviewed to Rehabilitation Programs Even Though
More Than 700 Other Inmates Had Higher Risks and Needs

Vocational
education

3

Academic
education

3

Vocational
Education

Folsom

89

3

R. J. Donovan

164
22

1

CBT

Academic
education

29

4

Vocational
education

San Quentin

31

1

Inmates that Corrections assigned to rehabilitation programs
Inmates who should have been a higher priority for assignment to rehabilitation programs
Source:  Analysis of SOMS data.

According to internal documents we reviewed, Corrections is aware
that its waiting lists have been ineffective at prioritizing inmates
based on their risks and needs. The deputy director noted that the
waiting list had been useful for certain purposes, but an internal
Corrections document from March 2018 stated that “the waiting
list has been useless for assigning inmates to rehabilitative program
assignments since implementation because it does not address any
rehabilitative case factors.” Specifically, the current waiting lists do
not consider rehabilitative case factors when prioritizing inmates
for assignment, such as the inmate’s CSRA and COMPAS scores.
As of December 2018, Corrections is in the process of developing a

369

27

28

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

solution to update the waiting list system to one that is more dynamic
and will allow Corrections to update the criteria to match its priorities,
with an expected completion date of April 2019. Because Corrections
has not held prisons accountable for putting inmates onto waiting
lists within five years of their scheduled release and has not enforced
a consistent process to assigning inmates into classes, it is failing to
maximize the number of inmates who receive rehabilitation programs.
Corrections Is Not Staffing Rehabilitation Programs at the Levels Needed
to Run Them Effectively
Corrections has yet to staff its rehabilitation programs at levels sufficient
to meet its staffing goals for its rehabilitation programs. As part of its
blueprint, Corrections committed to an increase in its rehabilitative
staff to support its new standardized staffing plan. The new staffing plan
redistributed resources to correspond to the reduction in the inmate
population due to realignment and the increased need for staff dedicated
to inmate rehabilitation programs. It set a goal of providing rehabilitation
programming to 70 percent of inmates who have a moderate to high risk
of recidivating and medium to high rehabilitation programming needs
before their release. Corrections has performed several staffing‑level
assessments over the past several years, and rehabilitation program
staffing levels have increased since the 2012 blueprint. Using inmate
data, we verified that Corrections’ budgeted staffing levels for fiscal
year 2017–18 for its vocational and academic education programs were
sufficient for it to achieve its goal of providing rehabilitative services
to 70 percent of its target population. Although we found that the
staffing assessments that Corrections has performed are reasonable,
the programs need to be fully staffed for the rehabilitation programs to
have their maximum impact.
Corrections has historically struggled with high staff vacancy rates
for its academic and vocational education programs, which are run
by its own staff, and this has limited the opportunities for inmates to
participate in these two types of rehabilitation programs. For example,
according to the Inspector General’s 2013 Blueprint Monitoring Report,
Corrections had vacancies in 20 percent of its rehabilitation staff
positions as of August 2013. Although Corrections has improved its
staffing levels, these two programs combined still had a 13 percent
vacancy rate according to the 2018 Blueprint Monitoring Report, as
shown in Table 5.9 The deputy director believes that an appropriate
level for rehabilitative programming would be to have vacancies in

9	

Corrections contracts with third‑party vendors to provide inmate programming for its CBT programs,
which do not use Corrections staff. According to the Blueprint Monitoring Report, these CBT programs
are largely operational and vendors were operating all but a small number of CBT classes. According to
the deputy director, it would be cost prohibitive for Corrections to create and fund state positions for
supervision and lower‑level facilitators for its CBT programs.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

less than 10 percent of budgeted positions. Additionally, the deputy
director stated that some reasons the vacancy levels were higher
than 10 percent included the large‑scale expansion of rehabilitation
programs over the last few years and the difficulty of hiring certain
educational classifications in remote locations. However, Corrections’
inability to meet its rehabilitative staffing requirements means it may
not be able to meet its goals of providing rehabilitation programs to
those inmates who need them prior to release and maximizing the
programs’ positive impacts.
Table 5
Corrections Had Vacancies in More Than 10 Percent of Its Budgeted Rehabilitation Program Staff Positions
DECEMBER 2017 TO JANUARY 2018
BUDGETED STAFF

ACTUAL PROGRAM STAFF

NUMBER OF
VACANT POSITIONS

PERCENTAGE OF
VACANT POSITIONS

Academic education

543

491

52

10%

Vocational education

304

250

54

18

847

741

106

CORRECTIONS’ STAFFED
REHABILITATION PROGRAMS*

Totals

13%

Source:  Analysis of the Inspector General’s Ninth Report: Blueprint Monitoring, dated July 2018.
Note:  Actual numbers were identified by the Inspector General’s staff during on‑site visit reviews of each prison from December 2017 through
January 2018.
*	 This table does not include staffing levels for CBT, substance abuse, or preemployment Transitions programs because those programs are staffed
through third‑party vendors. According to the July 2018 Blueprint Monitoring Report, these vendor‑operated programs had a combined vacancy rate
of 7 percent. The table also does not include CalPIA staffing vacancies.

The Low Enrollment Rates of Corrections’ Rehabilitation Programs
Reduce Their Benefit to Inmates and the State
High staff vacancy rates and a failure to place inmates on program
waiting lists has resulted in Corrections not utilizing all of its
programs’ budgeted capacity. Although Corrections has expanded
its rehabilitation programs to all 36 prisons, prison staff have not
enrolled the maximum number of inmates in each rehabilitation
class. As shown in Table 6 on the following page, the three prisons
we reviewed enrolled inmates to fill 76 percent of their budgeted
capacity for academic education programs, 76 percent for CBT
programs, and 68 percent for vocational education, on average, during
fiscal year 2017–18.10 We calculated enrollment rates by comparing
the number of inmates enrolled in each program category with each
category’s budgeted capacity. At individual prisons, we found that
R. J. Donovan, at 67 percent, had the lowest academic enrollment rate;

10	

We could not calculate the enrollment rates for volunteer programs because Corrections does not
adequately track this information.

29

30

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

San Quentin had the lowest CBT enrollment rate, at 67 percent;
and R. J. Donovan had the lowest vocational enrollment rate, at
40 percent.
Table 6
The Three Prisons We Reviewed Did Not Enroll the Maximum Number of Inmates in Their Rehabilitation Programs
During Fiscal Year 2017–18
MONTHLY AVERAGE BUDGETED
INMATE CAPACITY

MONTHLY AVERAGE NUMBER OF
ENROLLED INMATES

ENROLLMENT
RATE

Folsom

537

415

77%

R. J. Donovan

702

471

67

PRISON

Academic education

CBT programs*

Vocational education

CalPIA’s vocational programs†

San Quentin

319

292

92

Average

519

393

76%

Folsom

251

190

76%

R. J. Donovan

240

206

86

San Quentin

264

178

67

Average

252

191

76%

Folsom

351

300

85%

R. J. Donovan

270

107

40

San Quentin

162

125

77

Average

261

177

68%

100

26

26%

Folsom
San Quentin

56

44

79

Average

78

35

45%

Source:  Analysis of SOMS data for fiscal year 2017–18.
*	 We excluded programs designated for inmates serving lengthy periods of incarceration, such as life with the possibility of parole.
†	 CalPIA does not have any vocational programs at R. J. Donovan.

Prisons have been unable to address the causes of low enrollment,
which mainly involve problems in hiring staff and the lack of adequate
space. Specifically, the R. J. Donovan principal stated that staff
shortages and difficulty filling vacant positions affected its vocational
enrollment rate, as four of its nine positions were vacant during fiscal
year 2017–18 and had been vacant since 2016. Both R. J. Donovan and
Folsom indicated that they have continuously posted the vacancies
and have even interviewed people to fill the positions. However,
for various reasons, they have each been unable to fill them. The
Folsom and R. J. Donovan school principals stated that physical
space limitations also have restricted the number of students they

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

can enroll in their classes, preventing the prisons from filling all of
the budgeted slots in their academic and vocational courses. For
example, the roof in the building where Folsom held three of its
vocational classes needed repairs, causing Folsom to reduce the
class size in those programs. The R. J. Donovan school principal
stated that the fire marshal limited the capacity in at least three of
the prison’s academic education classes from 27 to 16, a reduction
of 33 slots. While Corrections does use space surveys to identify
available space for program expansion and select areas, such as
kitchens and closets, that it can retrofit for educational purposes, it
does not use these reports to find more permanent space solutions.
CalPIA provides additional vocational courses with unique benefits
to inmates, including potentially decreasing recidivism at a higher
rate than Corrections’ rehabilitation programs. As discussed in the
Introduction, CalPIA has partnered with various unions to provide
inmates with benefits beyond vocational training. Upon completion
of its union‑affiliated programs, inmates become eligible for each
union’s apprenticeship program upon release, and receive tools
and their first year of union dues from CalPIA. Additionally, in
July 2011 CalPIA reported that the recidivism rate for those inmates
who completed its vocational education programs was 7 percent,
significantly lower than Corrections’ overall recidivism rate of
51 percent for inmates released in fiscal year 2010–11. CalPIA’s ability
to select inmates with a minimum or medium security level for
placement into its programs could be affecting its recidivism rate.

In July 2011, CalPIA reported that the
recidivism rate for those inmates who
completed its vocational education
programs was 7 percent.

However, CalPIA has not updated this rate in seven years, and
in February 2017 it entered into an agreement with UC Irvine
to recalculate the recidivism rate. CalPIA estimated that this
recalculation will be complete in May 2019. In addition, as noted
previously, drastic changes in the prison population, including an
increase in the number of inmates in custody for violent crimes,
warrant a recalculation of the recidivism rate for inmates that have
completed CalPIA’s vocational programs.
Despite the potential benefits, CalPIA’s vocational programs have
low inmate enrollment rates. CalPIA’s enrollment rates at all of
the nine prisons where it has vocational programs averaged only

31

32

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

48 percent during fiscal years 2014–15 through 2017–18. Further,
we found that Folsom used only 26 percent of CalPIA’s vocational
programs’ capacity, while San Quentin used 79 percent, for an
average of 45 percent across the two prisons. CalPIA’s manager of
vocational education programs (CalPIA manager) stated that the
process by which Corrections assigns inmates to CalPIA programs
has limited the number entering its programs. Once an inmate
arrives at a prison, Corrections requires that he or she meet with
a classification committee that is responsible for placing inmates
into programs based in part on their needs. However, according
to the CalPIA manager, CalPIA staff cannot regularly attend
these meetings, and as a result the committee is giving priority
to Corrections’ rehabilitation and work programs over CalPIA’s
programs, thereby limiting the number of inmates CalPIA can
select. The deputy director indicated that classification is a local
responsibility and that Corrections does not have a statewide
policy that requires inmates to take specific rehabilitation courses
over others.
Although CalPIA has been working with Corrections to address
the low enrollment in its vocational programs, resulting in modest
improvements, its programs continue to be underutilized. To better
ensure that inmates receive vocational education, CalPIA and
Corrections established a pilot program in May 2018 that would
automatically enroll inmates into Corrections or CalPIA vocational
courses before assigning them to prison jobs. Specifically, at select
prisons CalPIA hired representatives to play a more active role
in the classification process by attending committee meetings
and increasing interaction with inmates and classification staff to
encourage enrollment in CalPIA vocational education programs.
Corrections began implementing this pilot program at five prisons
in August 2018, encompassing 12 CalPIA vocational programs,
and thus far the enrollment rate has increased from 44 percent of
capacity in July 2018 to 53 percent in October 2018. According to
CalPIA’s general manager, it is in the process of expanding the pilot
program to all prisons where CalPIA offers vocational education.
Although CalPIA’s programs remain significantly underutilized,
it is planning to expand from the nine prisons where it currently
provides vocational programs to offer opportunities to inmates at
18 prisons in fiscal year 2018–19.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Recommendations
To ensure that inmates with the highest risks and needs are
wait listed, prioritized, and assigned appropriately, Corrections
should do the following:
•	 Require correctional counselors to place inmates onto waiting
lists once they have five years or less on their sentences.
•	 Update its waiting list system to prioritize inmates with
rehabilitative needs and risks in its target population.
•	 Assign inmates to rehabilitation programs in accordance with
its policies.
To ensure that it can meet the rehabilitation needs of its inmates,
Corrections should develop and begin implementing plans to
meet its staffing‑level goals for rehabilitative programming by
January 2020 and should implement a process to continuously
update and monitor these goals.
To increase the space available for rehabilitation programs, by
January 2020 Corrections should analyze and report on its current
infrastructure capacity compared to its needs for the programs. The
report should include the current space available and the square
footage needed. If the report indicates that additional space is
necessary, Corrections should work with the Legislature to address
those needs.
To improve the inmate enrollment rates in CalPIA’s vocational
education programs, CalPIA and Corrections should require
a CalPIA representative to attend all classification committee
meetings at all nine prisons where CalPIA offers vocational
education. Corrections should also ensure that it enrolls eligible
inmates in CalPIA’s vocational programs before filling spots in its
own vocational programs. In addition, if the CalPIA recidivism
study indicates that CalPIA’s vocational programs are better at
reducing recidivism than Corrections’ vocational programs,
CalPIA should request funding from the Legislature to expand its
vocational training program.

33

34

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Blank page inserted for reproduction purposes only.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Additional Oversight Is Needed to
Ensure the Effectiveness of Corrections’
Rehabilitation Programs
Key Points
•	 Corrections has neither developed any performance measures for its rehabilitation
programs, such as a target reduction in recidivism, nor assessed program
cost‑effectiveness.
•	 Corrections has not analyzed whether its rehabilitation programs reduce recidivism,
and it needs to collect additional data and take steps to ensure that CBT programs are
being delivered as intended across all of its prisons.
•	 Although the Inspector General and C‑ROB perform some limited oversight of
Corrections’ rehabilitation programs, neither is well suited to conducting the
comprehensive analysis needed to determine whether those programs are effective at
reducing recidivism or are cost‑effective.
•	 C‑ROB is well positioned to oversee Corrections’ processes for creating and
monitoring rehabilitation programs’ performance for cost‑effectiveness, and for
contracting with an external researcher to analyze the programs’ effectiveness at
reducing recidivism.
Corrections Has Neither Established Performance Measures for Its Rehabilitation Programs
Nor Measured Their Cost‑Effectiveness
Because one of the primary goals of rehabilitation programs is to reduce recidivism, we
expected to find that Corrections had established performance measures for its programs—
such as a target decrease in recidivism—to ensure that they achieve their intended result
and are cost‑effective for the State; however, we found that it has not done so. Developing
meaningful and accurate targets is a critical step in ensuring that rehabilitation programs
are achieving their desired outcomes. The deputy director stated that several barriers have
historically inhibited Corrections from setting recidivism goals, including statutory changes
for inmate sentencing. However, he noted that if done appropriately, a specific goal for
reducing recidivism would be useful. We believe that establishing annual targets for reducing
recidivism would be both useful and feasible.
Many other states have set specific goals for reducing recidivism and are receiving assistance
in meeting these goals from a federal grant. Specifically, the Statewide Recidivism Reduction
Grant (recidivism reduction grant), operated by the U.S. Department of Justice’s Bureau
of Justice Assistance, requires grantees to focus on specific priorities related to reducing
recidivism, including implementing evidence‑based practices and focusing on the most
at‑risk inmates as a condition of receiving up to $3 million. As an example, Minnesota has
set a five‑year goal of reducing recidivism by 3 percent with the help of nearly $3 million
in recidivism reduction grant funds. Minnesota is pursuing that goal in a number of ways,

35

36

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

including focusing efforts on the highest‑risk inmates and using
evidence‑based programs. According to the deputy director, he does
not believe the division has ever applied for this grant, but it may be
interested in doing so in the future.
In addition to setting recidivism targets, Corrections could use other
metrics to help improve its performance and to illustrate how effectively
it uses state resources. These metrics include the percentages of inmates
who enroll in and complete the rehabilitation programs that match
their needs. Although Corrections has been increasing the number
of inmates it enrolls in its CBT classes, as shown in Table 7, over the
time period we reviewed its prisons enrolled only from 16 percent to
22 percent of inmates in their needed CBT courses before they were
released. Table 7 also shows that completion percentages for CBT
programs ranged from 67 percent to 71 percent over the time period we
reviewed. Corrections has a goal of providing rehabilitative services to
70 percent of its target population. However, Corrections does not have
specific goals for inmates to complete CBT classes. According to its
2018 annual report, C‑ROB emphasized the importance of measuring
program outcomes, such as the percentage of inmates completing CBT
classes. Until Corrections develops meaningful measures pertaining to
program participation and completion, it will not be able to adequately
track or evaluate its performance over time.
Table 7
The Number of Inmates Who Enrolled in and Completed at Least One CBT Program Remains Low
INMATES ENROLLED IN
AT LEAST ONE CBT PROGRAM

INMATES WHO COMPLETED
AT LEAST ONE CBT PROGRAM

NUMBER

NUMBER

FISCAL YEAR
OF RELEASE

TOTAL NUMBER OF INMATES RELEASED WITH
AT LEAST ONE MODERATE TO HIGH CBT NEED

2015–16

25,041

4,006

16%

2,699

67%

2016–17

26,379

4,681

18

3,325

71

2017–18

32,221

7,168

22

4,983

70

Totals

83,641

15,855

PERCENTAGE

19%

11,007

PERCENTAGE

69%

Source:  Analysis of SOMS data.

In addition to encouraging better performance outcomes, setting
performance targets will also help Corrections and the Legislature
ensure that rehabilitation programs are cost‑effective. Corrections
currently does little to ensure that its rehabilitation programs are
cost‑effective and has not conducted cost‑effectiveness studies.
However, Corrections has much of the information needed to do so.
For example, Corrections estimated that it cost approximately $79,000
annually to incarcerate an inmate in fiscal year 2017–18. Based on its
rehabilitation budget and the number of inmates who completed

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

rehabilitation programs during that same fiscal year,
we estimate that it cost the State about $8,500 for each
inmate that completed one or more rehabilitation
programs. Thus, as shown in the text box, Corrections
might be able to demonstrate that its programs are at
least cost neutral if they keep just one out of every
nine inmates who completed them from recidivating.
It is important to note that this example does not
include the avoided societal cost for the crime that
triggered the reincarceration.
Although Corrections plans to coordinate with
external researchers to conduct a performance
evaluation of the rehabilitation programs—which
may include a high‑level cost‑effectiveness analysis—
over the course of the next two years, Corrections
has taken no formal steps to initiate this process.
Because the Legislature provided Corrections with
a significant budget increase so that it could expand
rehabilitation programs to all prisons in the State, it is
vital that Corrections demonstrate that the additional
investment was worthwhile.

Rehabilitation Programs Need to Reduce
Recidivism by Only a Small Amount
to Be Cost Neutral
•	 Total annual cost of incarcerating an inmate = $79,000
•	 Cost to provide one or more rehabilitation courses to
an inmate =
$298 million
(rehabilitation budget)
35,000 inmates completing
programs during fiscal
year 2017–18

=

$8,500 per inmate
completing one or
more rehabilitation
programs*

•	 $8,500 = 11% of $79,000 (meaning that an 11% decrease in
recidivism will be cost neutral)
•	 Therefore, if rehabilitation programs keep 11 percent, or one
out of nine inmates who complete them, from recidivating,
the programs become cost neutral.
Source:  Analysis of the fiscal year 2018–19 Budget Act and
Corrections’ monthly Rehabilitation Program reports throughout
fiscal year 2017–18.
*	 We have excluded the cost and number of completions
associated with CalPIA vocational education programs from
this calculation.

Corrections Has Not Conducted a Comprehensive
Evaluation of Its Rehabilitation Programs to
Determine Whether They Are Effective
Corrections has not determined the overall effectiveness of
its rehabilitation programs, including whether its programs
reduce recidivism. State law encourages Corrections to develop
rehabilitation programs designed to promote behavioral change and
prepare inmates to successfully reintegrate into their communities.
While Corrections has examined the effectiveness of its substance
abuse classes, it has not conducted an analysis of the outcomes of its
other rehabilitation classes or an analysis to see what classes are most
effective at reducing recidivism. Thus, it is unclear how Corrections
can demonstrate that its programs promote behavioral change and
prepare inmates to successfully reintegrate into their communities.
A comprehensive review of the rehabilitation programs’ effectiveness
is especially important given that our preliminary analysis found
little evidence that Corrections’ programs as they are currently
administered reduce recidivism.
According to the deputy director, although systematically evaluating
Corrections’ rehabilitation programs would be valuable, this has not
yet been possible, as Corrections did not extend its CBT programs
to all of the prisons in the system until 2016. Therefore, according to
the deputy director, it will not be possible to obtain recidivism

37

38

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

information on inmates enrolled in CBT, vocational, and academic
education programs until fiscal year 2020–21 because California
calculates recidivism rates by examining whether inmates have
been convicted of committing another misdemeanor or felony
within three years of their release. He also stated that conducting
such a study of Corrections’ rehabilitation programs would be
costly and would require a good deal of time. Further, Corrections
never conducted this type of study in the past because, according
to Corrections’ SOMS user project manager and data integrity
group lead, before the implementation of SOMS in October 2014
Corrections did not have reliable rehabilitation program data. With
the implementation of SOMS, Corrections now centrally tracks
program information, such as enrollment and outcomes, which will
make a future recidivism study possible. A rehabilitation program
administrator at one of the prisons we examined stated that outcome
data for CBT programs would be very valuable in comparing
the effectiveness of programs from year to year. However, until
Corrections both analyzes its existing data and enough time passes to
ascertain inmates’ recidivism rates, details on the effectiveness of its
rehabilitation programs will continue to be lacking.

Corrections now centrally tracks program
information, such as enrollment and
outcomes, which will make a future
recidivism study possible.

Although Corrections has never conducted an analysis to determine
the effectiveness of its rehabilitation programs, it has taken some
initial steps to ensure that it gathers program‑related data consistently
at all of its prisons and that vendors conduct CBT programs in the
same manner statewide. Specifically, it has contracted with UC Irvine
to examine ways to improve the uniformity of some of its CBT classes
and the data gathered from those classes. UC Irvine produced a
report for Corrections that provides recommendations to ensure that
vendors implement CBT programs consistently across all prisons.
Although this evaluation by UC Irvine tested only a small number of
CBT classes, and its review did not attempt to determine their overall
effectiveness, it was an important first step to ensure that Corrections
is gathering data and implementing CBT programs appropriately and
consistently across its prisons.
UC Irvine’s report examined how Corrections could improve
measurements of program fidelity in CBT programs. Program fidelity
concerns the degree to which a program is delivered as intended.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Ensuring that these programs are delivered as intended is vitally
important, as the UC Irvine report states that program fidelity in
evidence‑based CBT programs is strongly correlated to reduced
recidivism. The report recommends that Corrections adopt a program
fidelity checklist (checklist) to help ensure that its CBT programs are
being taught in a similar manner across its prisons. The checklist—
which UC Irvine created for Corrections—would integrate measures
such as adherence to the type of treatment, the amount of time inmates
spend in class, and the quality of teaching. The deputy director stated
that historically a lack of program fidelity has made it difficult to
assess program performance and the performance of vendors across
its facilities. According to the deputy director, Corrections has not had
time to implement the checklist, as UC Irvine released the report at the
end of June 2018, but it plans to do so by the end of fiscal year 2018–19.
Until Corrections implements the checklist across all of its prisons,
the vendors may not be implementing CBT programs as designed.
Therefore, Corrections will be unable to ensure that the programs are
as effective as possible at all of its prisons.
Additionally, the deputy director stated that Corrections intends
to measure the effectiveness of its vocational education programs
in both reducing recidivism and increasing the ability of inmates
to find employment after release, but it has had difficulty in doing
so. Corrections uses an agreement with the California Workforce
Development Board and Employment Development Department (EDD)
to track employment and the type of employment for former inmates
who received vocational education programs. For example, Corrections
could track whether inmates who received plumbing classes eventually
became plumbers. However, according to EDD’s privacy and disclosure
program coordinator, its system uses Social Security numbers to
identify individuals. According to a work group report issued by the
Prison to Employment Initiative—a state‑funded group created to
improve inmates’ employability—the Social Security numbers in SOMS
come from DOJ “rap sheets.” State regulations restrict Corrections’
ability to provide inmates’ Social Security numbers, permitting it to
do so only on a need‑to‑know basis to persons or agencies specifically
authorized to receive the information, which, according to Corrections’
assistant general counsel, is preventing EDD from getting the data it
needs to track inmate employment.

Corrections has had difficulty measuring
the effectiveness of its vocational education
programs in both reducing recidivism
and increasing the ability of inmates to
find employment.

39

40

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

To attempt to correct this issue, representatives from DOJ,
Corrections, and EDD stated that they have made recent efforts to
change state law to explicitly allow them to share the Social Security
numbers, but thus far they have been unsuccessful. According to
the deputy director, Corrections would like to conduct this analysis
of vocational education outcomes, but it has no plans to do so until
the issue of using inmates’ Social Security numbers is resolved.
Additionally, a former prison warden now involved in efforts to
assist inmates with finding employment after release noted that
some inmates claim to have multiple Social Security numbers,
which would further complicate tracking inmate employment
post‑release. The inability of Corrections to track inmates’
employment after release means that it is very difficult to know how
effective its education and job training programs are at preparing
inmates for the workforce.
Although Corrections has expanded its volunteer programs to
underserved locations and increased the number of volunteer
program opportunities for inmates, it has not yet determined
whether any of its volunteer programs are effective at reducing
recidivism or produce other positive effects. Provided by nonprofits
and volunteers, the programs include substance abuse support
groups, animal therapy, parenting classes, and health and wellness
programs. Corrections staff do not teach or participate in the
programs, but rather oversee them and perform administrative
tasks related to these programs. In fiscal year 2017–18, volunteer
programs had a budget of over $11 million and enrolled an average
of 61,000 inmates per month.11 The significance of these programs
increased in 2016 with the passage of Proposition 57, which
provided the opportunity for inmates to receive time off their
sentence if they completed some of these volunteer programs.
However, according to a SOMS user project manager and data
integrity group lead, Corrections is not required to report on the
outcomes of these volunteer programs, and according to the deputy
director, it has only recently begun to contemplate how best to
define the effectiveness of these programs.

Corrections has only recently begun
to contemplate how best to define the
effectiveness of its volunteer programs.

11	

This amount includes $3 million for innovative grants and $8 million for programs associated
with the California Arts Council.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Although Corrections is planning to measure the effectiveness of
some of its volunteer programs, it has not made sufficient progress
in doing so. The deputy director noted that it is considering
partnering with the Public Policy Institute of California (PPIC),
with a focus on determining how to measure the effectiveness of
the 192 volunteer programs that receive grants from Corrections.
Corrections or PPIC could expand the research to include all
volunteer programs, which total approximately 3,800. However, as
of November 2018, they had yet to reach a formal agreement and
had not yet defined the research scope. Without measuring the
effectiveness of these programs, Corrections will not know which
volunteer programs are working well and should be expanded to
other prisons and which it should shut down because they provide
no positive outcome.
Entities Responsible for Providing Oversight of Corrections’
Rehabilitation Programs Have Limited Resources to Determine
Whether the Programs Reduce Recidivism
Although the Legislature has tasked the Inspector General and
C‑ROB with evaluating elements of Corrections’ rehabilitation
programs, these entities do not have sufficient resources to
conduct an analysis of whether the programs reduce recidivism
or are cost‑effective. The Legislature gave the Inspector
General responsibility for conducting oversight of Corrections’
implementation of the blueprint but, as described below, eliminated
its authority to self‑initiate reviews not specifically requested by
the Governor or state lawmakers. Within its current authority and
resources, the Inspector General reviews unaudited Corrections
rehabilitation program data and performs site visits to determine
whether Corrections is meeting the goals set forth in the blueprint.
For example, the Inspector General reviewed rehabilitation
program data and found Corrections deficient in meeting its
blueprint goal to provide programming to at least 70 percent of
the target inmate population.
A second entity, C‑ROB, has been tasked by the Legislature with
conducting reviews of designated rehabilitation programs operated
by Corrections, including reviewing “the effectiveness of treatment
efforts.” However, C‑ROB is not structured or staffed adequately
to determine whether these programs are effective at reducing
recidivism. C‑ROB is an 11‑member board that includes state
officials, such as the inspector general serving as chair, the secretary
of Corrections, and the state superintendent of public instruction,
as well as university faculty members and local law enforcement
representatives. State law requires C‑ROB to meet at least twice
a year and to examine the various Corrections rehabilitation

41

42

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

programs for inmates and report its findings and recommendations
to the Legislature, including the effectiveness of treatment efforts
and the assessed rehabilitation needs of the inmates, among
other topics. Although state lawmakers provided initial funding
of $517,000 to the Inspector General to support C‑ROB, which
allowed the Inspector General to hire two C‑ROB support staff,
they discontinued this funding in 2011.
Officials from the Inspector General’s office explained that,
despite the loss of designated resources, the Inspector General has
continued to support C‑ROB by absorbing certain administrative
expenses into its own budget and by having its staff work
on C‑ROB initiatives. According to these officials, Inspector
General staff work on behalf of C‑ROB to conduct site visits to
prisons and gather rehabilitation program data, which they use
to create an annual report given to C‑ROB for approval. For
example, Inspector General staff collected data for C‑ROB on
Corrections’ rehabilitation programming enrollment and capacity
from June 2016 through June 2018 to analyze how effectively
Corrections fills spots for its rehabilitation programs. Similar
to our own analysis, the Inspector General found low inmate
enrollment. However, with a recent 42 percent budget reduction
in the Inspector General's budget, we do not believe it would be
reasonable to assume that the Inspector General could support
an effort by C‑ROB to use its authority to examine whether
Corrections’ rehabilitation programs are effective. Rather, as
outlined in the next section, C‑ROB—with limited support from
Inspector General staff—could provide oversight of Corrections’
efforts to complete and report the results of a comprehensive
analysis of the effectiveness of its rehabilitation programs.
Oversight Is Necessary to Ensure That Corrections Evaluates the
Effectiveness of Its Rehabilitation Programs
The Legislature, Corrections, and C‑ROB must work in concert
to determine the extent to which rehabilitation programs are
reducing recidivism and are cost‑effective. We did not identify any
law or regulation requiring Corrections to establish performance
measures, track how well it meets those goals, or conduct any
analysis to determine whether its adult, in‑prison rehabilitation
programs are effective at reducing recidivism. Further, there
is no executive branch oversight entity specifically responsible
for ensuring that Corrections performs any of these activities.
However, the Legislature can require Corrections to undertake
these endeavors and report on its progress annually. In addition,
C‑ROB has the expertise to oversee Corrections’ rehabilitation

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

programs to ensure that Corrections contracts with an external
researcher, develops appropriate targets for its rehabilitation
programs, and issues reports about its progress on an annual basis.
As noted earlier, because Corrections began to offer programs in
all prisons in fiscal year 2016–17, the recidivism data necessary to
conduct an analysis will not exist until fiscal year 2020–21. Thus, we
believe a three‑year plan—beginning in fiscal year 2019–20—would
provide ample time for Corrections and its external researcher to
collect, analyze, and report on whether its programs are effective.
Figure 7 on the following page summarizes our recommendations
in this area.
Although C‑ROB does not currently conduct extensive oversight to
determine the effectiveness of Corrections’ rehabilitation programs,
with additional resources and statutory authority, it could be
well positioned to conduct this oversight. As noted earlier, the
Inspector General currently provides the staff support for C‑ROB.
Thus, under a similar staffing model, C‑ROB could work with the
Inspector General’s staff to monitor Corrections’ evaluation of the
effectiveness of its rehabilitation programs, including monitoring
Corrections’ contract with an external researcher tasked with
conducting a systematic evaluation of all rehabilitation programs.
In addition, C‑ROB could work with the Inspector General’s
staff to monitor Corrections’ efforts to develop and meet annual
recidivism targets. According to the C‑ROB chief counsel, C‑ROB
could monitor Corrections’ progress in meeting its performance
targets and Corrections’ contracting process with an external
researcher, but would require additional statutory authority. State
law also currently requires C‑ROB to issue an annual report to the
Legislature that includes findings on the effectiveness of treatment
efforts; however C‑ROB’s chief counsel noted that additional
resources for the Inspector General would be necessary for its staff
to report the results of the monitoring to C‑ROB for inclusion in
its annual legislative report. Regardless of the entity given funding
and authority, Corrections requires oversight to ensure that it takes
necessary steps to evaluate the effectiveness of its rehabilitation
programs at reducing recidivism.

43

44

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Figure 7
These Actions Will Improve the Oversight and Accountability of Corrections’ Rehabilitation Programs

THE LEGISLATURE
To hold Corrections accountable for the effectiveness of its
rehabilitation programs, the Legislature should do the following:
• Require Corrections to establish performance targets and partner with external
researchers to evaluate the effectiveness of its rehabilitation programs.
• Task C-ROB with monitoring Corrections’ evaluative process and with
providing annual updates to the Legislature on progress in implementing the
three-year plan outlined below.

CORRECTIONS

Annual
reports

C-ROB

• Create targets for measuring program
cost-effectiveness.
• Partner with external researchers to
conduct a systematic evaluation of its
rehabilitation programs’ effect on
recidivism, based on scope of work
with data elements the researcher
may require.
• Collaborate with the external researcher
to establish annual targets for reducing
recidivism based on findings from
the analysis.

Oversight and
approval

Progress
reports

• Approve Corrections’ performance
targets and monitor its progress in
meeting those targets.
• Provide independent oversight of
Corrections’ external researcher
selection, scope of work, and
report presentation.
• Annually report to the Legislature on
Corrections’ progress on the three-year
plan described below.

Monitor and approve

Collaborate

EXTERNAL RESEARCHERS

• Provide input on performance targets.
• Conduct analysis on rehabilitation
programs’ effect on recidivism.

THREE-YEAR PLAN
Fiscal year 2019–20

Fiscal year 2020–21

Fiscal year 2021–22

Corrections drafts scope of
work, selects external
researcher to conduct analysis,
defines what data elements
the researchers may require,
and creates targets.

External researcher conducts
recidivism analysis and
Corrections takes corrective
action as necessary.

Corrections reports to the
Legislature and creates new targets
and policies given the results of
the recidivism analysis. Depending
upon the results of the analysis,
Corrections eliminates or modifies
programs that prove ineffective.

Source:  State Auditor’s recommendations to the Legislature, Corrections, and C‑ROB.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

Recommendations
Legislature
To ensure that Corrections’ rehabilitation programs reduce
recidivism, the Legislature should require Corrections to do
the following:
•	 Establish performance targets, including ones for reducing
recidivism and determining the programs’ cost‑effectiveness.
•	 Partner with external researchers to evaluate the effectiveness of
its rehabilitation programs and implement the three‑year plan
described below.
•	 Issue an annual report beginning in fiscal year 2021–22 that
shows the percentage reduction in recidivism that can be
attributed to the rehabilitation programs.

Year One: Fiscal Year 2019–20
Corrections drafts scope of work, selects an external researcher to
conduct the analysis, defines what data elements the researchers
may require, and creates targets.
Year Two: Fiscal Year 2020–21
External researcher conducts recidivism analysis and Corrections
develops and begins implementing a corrective action plan.
Year Three: Fiscal Year 2021–22
Corrections modifies as necessary and continues implementing its
corrective action plan. It also reports to the Legislature and creates
new targets and policies given the results of the recidivism analysis.
Depending upon the results of the analysis, Corrections eliminates
or modifies programs that prove ineffective.

To ensure that Corrections and its external researcher conduct a
comprehensive analysis of the rehabilitation programs’ effect on
recidivism, the Legislature should provide authority and funding for
C‑ROB to monitor the contracting process and provide progress
updates to the Legislature in its annual report.

45

46

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

To ensure that Corrections remains on track to complete its
analysis and develop performance targets, the Legislature should
require C‑ROB to monitor Corrections’ progress in developing
appropriate recidivism targets and meeting those targets, and to
provide annual updates on Corrections’ progress in implementing
the three‑year plan.
To ensure that Corrections and EDD can collaborate effectively
to track whether inmates that received vocational training found
work in a related field after release, the Legislature should amend
state law to explicitly allow Corrections to provide inmates’ Social
Security numbers to EDD.
Corrections
To ensure that Corrections effectively and efficiently allocates
resources and reduces recidivism, it should do the following:
•	 Partner with a research organization to conduct a systematic
evaluation during fiscal year 2020–21 to determine whether its
rehabilitation programs are reducing recidivism and if they are
cost‑effective. In addition, the external researcher should provide
input on the development of performance targets, including
recidivism reduction. Depending upon the results of the analysis,
Corrections should then eliminate or modify programs that
prove ineffective.
•	 Partner with an external researcher to help it quantify the effect
volunteer programs have on inmate outcomes and consider
expanding those programs if they prove effective or ceasing them
if they are not effective.
•	 Collaborate with C‑ROB to establish annual targets for reducing
recidivism and determining the cost‑effectiveness of the
programs. Corrections should also request federal grants tied to
setting targets for recidivism reduction.
To ensure that it has reliable tools to measure program fidelity
in its CBT programs, Corrections should implement UC Irvine’s
recommendation by June 2019.
To ensure that its vocational training programs are effectively
preparing inmates for the workforce upon their release and
reducing recidivism, Corrections should collaborate with EDD to
track the employment and the industry of employment for former
inmates by January 2020.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

C‑ROB
To ensure that Corrections is taking steps to reduce recidivism,
C‑ROB should monitor whether Corrections is developing
appropriate recidivism targets and, in its annual report, should
evaluate Corrections’ progress toward meeting those targets.
We conducted this audit under the authority vested in the California State Auditor by Government
Code 8543 et seq. and according to generally accepted government auditing standards. Those
standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to
provide a reasonable basis for our findings and conclusions based on our audit objectives specified
in the Scope and Methodology section of the report. We believe that the evidence obtained
provides a reasonable basis for our findings and conclusions based on our audit objectives.
Respectfully submitted,

ELAINE M. HOWLE, CPA
California State Auditor
January 31, 2019

47

48

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Blank page inserted for reproduction purposes only.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

APPENDIX A
TECHNICAL APPENDIX: DATA AND METHODOLOGY
Our analysis of the effect of Corrections’ CBT rehabilitation
programs on recidivism rates for inmates released in fiscal
year 2015–16 drew on data from Corrections and DOJ. During
fiscal year 2014–15, Corrections offered all four CBT programs at
11 men’s prisons. We compared recidivism rates and other data for
inmates that received CBT classes to the data for those that did
not. Our analysis examined inmates who completed CBT programs
but excluded inmates who completed programs for long‑term
offenders, defined as inmates with an indeterminate sentence
with the possibility of parole. We used a regression analysis to
control for many observable characteristics between the test and
control groups in an effort to isolate the effect on the recidivism
rates of inmates completing at least half of their assigned CBT
classes. Due to limitations in Corrections’ data, we were able to
examine only the one‑ and two‑year recidivism rates. In developing
our methodology we consulted with the director of the Center
for Evidence‑Based Corrections at UC Irvine, and a statistician
reviewed our statistical model and its results.
Data
Our analysis used data from both SOMS and DOJ. Corrections
implemented its SOMS inmate database statewide in October 2014.
It was designed to consolidate existing databases and replaced
multiple manual paper processes to standardize adult inmate data
and inmate population management practices. SOMS includes
personal information on individual inmates, such as their date
of birth, education level, and current location within the prison
system. SOMS also includes information on the programming
inmates received, such as when an inmate is enrolled in or
completed a rehabilitation class. We also obtained conviction data
from DOJ to determine whether an inmate recidivated within
two years of his release from prison.
Methodology
We obtained SOMS data to calculate recidivism rates for inmates
released in fiscal year 2015–16. Corrections began using SOMS to
track whether inmates completed CBT classes in October 2014.
Due to this data limitation, we were able to select only inmates that
completed CBT classes from October 2014 through their release in
fiscal year 2015–16 to determine whether the inmates recidivated

49

50

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

within two years.12 Thus, inmates released in fiscal year 2015–16
are the earliest release group our analysis could examine and still
calculate the one‑ and two‑year recidivism rates.
Our analysis required inmates to have completed CBT classes
for at least half of their needs as the minimum before being
considered to have had their rehabilitative needs met. Corrections
assessed inmate needs in four CBT categories (anger management,
substance abuse, criminal thinking, and family relationships).
Inmates who take classes that address at least half of their assessed
needs should have reduced their likelihood to recidivate because
Corrections’ contracts require vendors to use only evidence‑based
programs, meaning that the classes have been shown to be effective.
We used half, rather than all needs met, due to the limited number
of prisons that offered CBT classes during the time period we
examined. If we had focused only on inmates who had all of their
needs met, it would have severely limited the sample size for
our analysis.
To create the control and test groups for our analysis, we split the
cohort from the 11 prisons that were released in fiscal year 2015–16
into inmates who had at least half of their CBT needs met and
inmates who had not been assigned to CBT classes. Corrections’
policies prioritize inmates’ assignment to any CBT classes based on
a combination of their COMPAS and CSRA scores. Our analysis
examined inmates who were rated as having a moderate to high
need in at least one of the CBT categories. Given that our test and
control groups represented subsets of the prison population, we took
three steps to adjust for the potential differences in characteristics
between the two groups. First, we grouped inmates within the test
group by the total number of their CBT needs and their level of need
for each of the four CBT categories. Second, we matched a selection
of inmates from the control group with each member of the test
group based on their CBT needs combination. We did this to ensure
that we compared inmates who had the same CBT needs and who
differed in whether or not they completed at least half of the classes
they were assigned. Finally, we controlled for important observable
differences between the control group and the test group, including
age, self‑identified race, education (as measured by the inmate’s
most recent TABE reading score and whether or not the inmate
had a verified high school diploma or equivalency), crime risk (as
measured by the inmate’s CSRA score), and prison assignment with
variables in our regression analyses, as discussed below.

12	

Individual CBT classes generally last three to four months and no longer than six months.
Inmates would have had the opportunity to take multiple classes before their release during
fiscal year 2015–16.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

We calculated the overall recidivism rates between the control and
test groups with a regression analysis of the correlation between
inmates completing at least half of their assigned CBT classes
(treatment) and recidivism rates using probit regressions. Our
regressions examined the entire sample of the control and test groups
to determine the relationship between completing CBT classes and
inmates’ one‑ and two‑year recidivism rates while controlling for
prison location, age, self‑identified race, education, and crime risk.
We examined the correlation between completing CBT classes and
the one‑ and two‑year recidivism rates for each subset of our control
variables. This means that we conducted individual regressions for
prison, self‑identified race, education level, age, and crime risk. Table
A shows the results of these regressions, including the percentage
change in recidivism for inmates completing CBT classes and the
probability (p-value) that the results could be caused by chance.
The likelihood of obtaining a result that appears to be significant,
but actually occurs by chance, increases with the total number of
regressions performed. Nonetheless, we are confident in our analysis
that did not find an overall relationship between inmates completing
CBT rehabilitation programs and their recidivism rates.
Table A
P‑Values for Our Analysis of Inmates Who Completed CBT Classes for at Least
Half of Their Needs

FACTOR

CATEGORY

Overall

Prison

P‑VALUE

0.285

0.228

Avenal State Prison

0.806

0.671

California City Correctional Facility

0.288

0.821

California Institution for Men

0.894

0.656

California Men’s Colony

0.636

0.136

Correctional Training Facility

0.411

0.789

Chuckawalla Valley State Prison

0.936

0.162

High Desert State Prison

0.471

0.677

Ironwood State Prison
California Substance Abuse Treatment Facility

0.750
0.086†
0.016‡

Valley State Prison

0.319

Black

0.049‡

California State Prison, Los Angeles County

Self‑Identified
Race

P‑VALUE*

PERCENTAGE DECREASE
OR INCREASE IN
ONE‑YEAR RECIDIVISM

‑15%
‑10%

0.162
0.063†
0.000§

PERCENTAGE DECREASE
OR INCREASE IN
TWO‑YEAR RECIDIVISM

‑18%
‑18%

0.732
‑5%

0.027‡

Hispanic

0.672

0.776

White

0.379

0.974

Other

0.554

0.583

‑7%

continued on next page . . .

51

52

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

FACTOR

CSRA

Education

Age

CATEGORY

P‑VALUE*

PERCENTAGE DECREASE
OR INCREASE IN
ONE‑YEAR RECIDIVISM

P‑VALUE

Low Risk

0.145

Moderate Risk

0.969

High Risk ‑ Drug

0.086†

11%

0.199

High Risk ‑ Property
High Risk ‑ Violence

0.106
0.002§

‑13%

0.794
0.001§

No education

0.129

Elementary School

0.530

0.992

Middle School

0.314

0.245

Freshman/Sophomore

0.051†

Junior/Senior

0.807

0.456

High school diploma/GED

0.278

0.083†

No High school diploma/GED

0.452

0.592

18 to 29 years old

0.255

0.269

30 to 39 years old

0.090†

40 to 49 years old

0.458

0.604

50+ years old

0.707

0.958

PERCENTAGE DECREASE
OR INCREASE IN
TWO‑YEAR RECIDIVISM

0.915
0.857

‑16%

0.357

‑6%

‑5%

0.006§

‑11%
‑6%

0.324

Source:  Regression output based on SOMS and DOJ data.
*	 The p‑values test the null hypothesis that inmates’ assignment to at least half of their needed CBT classes has no effect. The lower the p-value, the
more likely you are to reject the null hypothesis. A higher p‑value indicates that you cannot reject the hypothesis. In other words, a low p‑value is
likely to be meaningful. Therefore, we only present recidivism increases or decreases for p‑values of 0.10 or less.
†	 p < 0.1	
‡	 p < 0.05	
§	 p < 0.01	

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

APPENDIX B
SCOPE AND METHODOLOGY
The Joint Legislative Audit Committee (Audit Committee)
directed the California State Auditor to perform an audit to
examine the effectiveness of in‑prison rehabilitation programs
at Corrections. Table B on the following page outlines the Audit
Committee’s objectives and our methods for addressing them.
Assessment of Data Reliability
The U.S. Government Accountability Office, whose standards
we are statutorily required to follow, requires us to assess
the sufficiency and appropriateness of computer‑processed
information that we use to support our findings, conclusions,
and recommendations. In performing this audit, we obtained
Corrections’ inmate information to identify inmates’ demographics,
rehabilitative needs, locations, and program participation.
Additionally, we obtained DOJ’s statewide conviction information
to identify inmates who recidivated. To evaluate these data,
we performed electronic testing of the data, reviewed existing
information about the data and systems, and interviewed agency
officials knowledgeable about the data. However, during our review
we identified data limitations. Specifically, program participation
information was not centrally tracked until the current system was
implemented in October 2014. Additionally, we were unable to
identify the needs of some inmates because Corrections does not
perform assessments on all inmates. Therefore, we found these data
to be of undetermined reliability for the audit purposes. Although
this determination may affect the precision of the numbers we
present, there is sufficient evidence in total to support our findings,
conclusions, and recommendations.

53

54

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Table B
Audit Objectives and the Methods Used to Address Them
AUDIT OBJECTIVE

1

Review and evaluate the laws, rules, and
regulations significant to the audit objectives.

2

Identify the roles and responsibilities of
Corrections and any other parties related to
the oversight of state‑funded rehabilitation
programs and assess the adequacy of their
oversight, including the extent to which
they ensure appropriate and consistent
implementation across institutions.

3

Determine whether Corrections has conducted
an assessment to determine the level of
resources required to meet the rehabilitative
needs of inmates in all of Corrections’ facilities.

4

Review and evaluate the effectiveness of the
rehabilitation programs by determining
the following:
a.  Corrections’ justification for the program’s
use and whether the program is based on
evidence and research.

METHOD

Identified and reviewed the laws, rules, and regulations for the Inspector General,
C‑ROB, Corrections, and CalPIA that were applicable to Corrections’ in‑prison
rehabilitation programs.
•  Documented the Inspector General and C‑ROB’s efforts in reviewing Corrections’
rehabilitation programs. In addition, we identified the statutorily mandated oversight
responsibility of the Inspector General and C‑ROB and found they were generally
complying with those requirements.
•  Reviewed a selection of 10 academic education programs and determined they were
accredited with the Western Association of Schools and Colleges.
•  Evaluated the adequacy of oversight of prison rehabilitation programs by Corrections,
Inspector General, C‑ROB, CalPIA, and the Prison Industry Board.
Obtained and evaluated assessments conducted by Corrections from 2012 through 2018
to determine if Corrections is meeting its resource allocation goals and whether those
goals are adequate to meet the rehabilitative needs of its inmates.

•  Reviewed and documented the process used by Corrections when procuring CBT
services. Reviewed 10 CBT vendor contracts to determine if Corrections procured
those services appropriately.
•  Selected 10 contracts to determine whether the vendors were using evidence‑based
curricula. In addition, we tested three vendors' contracts to determine whether
the curricula being taught at the prisons were the same as the curricula maintained
in Corrections’ database.

b.  Corrections’ method for evaluating the
cost‑effectiveness of the program and
whether it has considered investing in an
independent oversight entity to perform
this function.

•  Interviewed Corrections staff and reviewed policies and procedures to evaluate the
methodology for assessments Corrections and CalPIA have conducted regarding
the cost‑effectiveness of their rehabilitation programs.
•  Reviewed and evaluated best practices from other state and federal entities to
determine an appropriate methodology for Corrections and CalPIA to follow
for determining the cost‑effectiveness of its rehabilitation programs.
•  Interviewed Corrections and CalPIA staff to determine if they have considered
using an independent oversight entity to evaluate the cost‑effectiveness of their
rehabilitation programs.
•  Evaluated the objectives and methodology of the proposed study that UC Irvine is
conducting on the impact of Corrections’ rehabilitation programs on recidivism.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

AUDIT OBJECTIVE

5

Determine whether Corrections maintains
adequate data on its rehabilitation programs—
such as enrollment, attendance, and outcomes,
by facility—to allow stakeholders to compare
the effectiveness of rehabilitation resources
across facilities.

METHOD

•  Obtained a complete list of the rehabilitation programs offered at all prisons, and
interviewed staff to determine how long the agency has offered the programs,
and how Corrections measures program completeness (Corrections operates
36 prisons at 35 locations; the Folsom location houses two prisons, a men’s prison—
Folsom State Prison—and a women’s prison—Folsom Women’s Facility).
•  Obtained SOMS data for all inmates incarcerated from October 2014 through
September 2018 and determined the number of inmates enrolled, attending, and
completing rehabilitation programs.
•  Obtained all rehabilitation program vendor contracts for fiscal year 2017–18 and
determined the budgeted capacity for all programs offered by Corrections.
•  Compared the number of enrolled inmates to the budgeted capacity for each
program during fiscal year 2017–18 at the three prisons we reviewed.
•  Determined that Corrections maintains adequate data on rehabilitation programs,
including enrollment, attendance, and outcomes, by facility to allow stakeholders to
compare effectiveness of rehabilitation resources across facilities.
•  Analyzed CalPIA data from July 2014 through October 2018 to determine its
vocational programs’ capacity, enrollment, and vacancy rate.

6

7

Determine whether Corrections effectively
assesses inmates’ risks and rehabilitation needs
and reviews and evaluates how it ensures that
the tools used to identify the needs and risks
are valid. Assess the adequacy of Corrections’
policies, procedures, and practices for selecting
and prioritizing inmates for participation in
rehabilitation programs.

•  Interviewed Corrections staff and reviewed policies, procedures, and practices
regarding the assessment and prioritization of inmates’ needs.

Identify and assess Corrections’ performance
measures—such as whether inmates’
rehabilitative needs were met prior to their
release and whether inmates are progressing in
their programs—to evaluate the effectiveness
of rehabilitative services.

•  Obtained all policies and reports regarding performance measures, including
historical documents.

•  Reviewed and evaluated assessments (COMPAS, CSRA, and TABE) used by Corrections
and determine whether the tests have been validated.
•  Selected 60 inmates incarcerated by Corrections from fiscal years 2015–16 through
2016–17 to determine whether Corrections followed established policies and
procedures to place inmates into appropriate rehabilitation programs.
•  Examined how Corrections prioritized inmates for rehabilitation programs for
19 inmates during fiscal year 2017–18 at the three prisons we selected to determine if
the inmate assignments into their respective rehabilitation programs were appropriate.

•  Determined the recidivism rate for inmates that Corrections released from prison in
fiscal year 2015–16 and compared the rate of recidivism for those inmates that had at
least half of their rehabilitation needs met to those that had none of their needs met.
•  Identified best practices for performance measures used by other states, federal
agencies, and countries applicable to California.
•  Determined what issues are preventing Corrections from working with EDD to track
inmates’ employment outcomes post‑release.
•  Reviewed whether Corrections’ policies are tracking academic educational
performance through academic testing. Although Corrections does track inmates’
academic performance, it does not set performance measures or determine whether
inmates’ rehabilitative needs were met prior to release.

8

Determine whether Corrections maintains a
waiting list for rehabilitative programs, the
number of inmates on the waiting list, how
long they have been on the waiting list, their
risks, and their needs.

•  Interviewed correctional counselors at the three prisons we selected and determined
how staff place inmates on waiting lists at each institution.
•  Determined if Corrections’ waiting list process effectively prioritizes inmates
with a moderate to high risk of recidivating and a moderate to high need into
rehabilitation programs.
•  Determined the number of inmates on waiting lists for academic, vocational, and CBT
programs, how long they have been on the waiting list, and their risks and needs.

9

Review and assess any other issues that are
related to the audit.

Interviewed Corrections staff and obtained a complete list of all volunteer programs and
determined whether Corrections evaluated these programs for their effectiveness.

Source:  Analysis of Audit Committee's audit request number 2018‑113, planning documents, and analysis of information and documentation
identified in the table column titled Method.

55

56

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Blank page inserted for reproduction purposes only.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

APPENDIX C
THE NUMBER OF INMATES ON WAITING LISTS AND THE
AVERAGE TIME SPENT ON THE LIST
We reviewed Corrections’ waiting list data, and we present the
number of inmates on waiting lists as of July 2018 in Table C. This
table shows that there were between 18,000 and 31,000 inmates on
waiting lists as of July 2018, and that those inmates were on waiting
lists for an average of between 211 and 351 days. However, these
data are presented for informational purposes only because, as we
discuss on pages 23 through 28, we identified significant flaws with
Corrections’ waiting lists.
Table C
The Number of Inmates on Waiting Lists and the Average Time Spent on the List
WAITING LIST THAT
ADDRESSES THE NEED FOR:*

NUMBER OF INMATES
ON WAITING LIST

AVERAGE NUMBER OF
DAYS ON WAITING LIST

Academic education
Target population†

7,178

230

Not target population

10,820

338

Totals

17,998

295

Vocational education
Target population†

6,702

285

Not target population

23,008

351

Totals

29,710

336

Anger management
Target population†

10,951

217

Not target population

19,983

268

Totals

30,934

250

9,554

215

Criminal thinking
Target population†
Not target population

21,524

261

Totals

31,078

247

12,796

211

Substance abuse disorder treatment
Target population†
Not target population

18,172

289

Totals

30,968

257

continued on next page . . .

57

58

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

WAITING LIST THAT
ADDRESSES THE NEED FOR:*

NUMBER OF INMATES
ON WAITING LIST

AVERAGE NUMBER OF
DAYS ON WAITING LIST

Family relationships
Target population†

15,733

212

Not target population

13,313

295

Totals

29,046

250

Source:  Analysis of SOMS data as of July 24, 2018.
*	 We excluded programs designated for inmates serving lengthy periods of incarceration, such as
life with the possibility of parole.
†	 Corrections defines its target population as inmates with a moderate to high risk of recidivating
and a moderate to high need for the program.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

59

60

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Blank page inserted for reproduction purposes only.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

*

1

2

*  California State Auditor’s comments appear on page 63.

61

62

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

COMMENTS
CALIFORNIA STATE AUDITOR’S COMMENTS ON THE
RESPONSE FROM THE CALIFORNIA DEPARTMENT OF
CORRECTIONS AND REHABILITATION
To provide clarity and perspective, we are commenting on the
response from Corrections. The numbers below correspond to
the numbers we have placed in Corrections’ response.
The main focus of the audit report was not Corrections’
CBT programs. We discussed its academic and vocational
education programs on pages 28 through 31, and its volunteer
rehabilitation programs on pages 40 and 41. Furthermore, we
analyzed Corrections’ wait-list process for all program types,
including academic and vocational education, beginning on
page 23. We also reviewed Corrections’ policies and procedures
as they pertained to all types of rehabilitation programs. CBT was
the main focus of only a portion of our analysis. Specifically, as
we state on page 14, we analyzed whether CBT programs reduced
recidivism rates because the majority—70 percent—of the State’s
recent expansion of its rehabilitation programs budget was solely
for expanding CBT to all prisons.

1

We acknowledged in the audit report many of the steps Corrections
has taken to improve its rehabilitation programs. Specifically,
we discuss the program accountability tools it developed in
conjunction with UC Irvine and the modifications it plans to make
to its CBT contracts on page 20. We also discuss the program's
checklist it plans to implement on page 39.

2

63

64

Report 2018-113 | C ALIFO R N IA S TAT E AUD I TO R
January 2019

Blank page inserted for reproduction purposes only.

C ALIFO R N IA S TAT E AUD I TO R | Report 2018-113

January 2019

65