Performance audit, Department of Corrections, institutional security and staffing;

              PERFORMANCE AUDIT
DEPARTMENT OF CORRECTIONS
INSTITUTIONAL SECURITY
AND STAFFING
Report to the Arizona Legislature
By the Auditor General
May 1991
91- 4
DOUGLAS R. NORTON, CPA
AUDITOR GENERAL
STATE OF ARIZONA
OFFICE OF THE
AUDITOR GENERAL
May 22, 1991
Members of the Arizona Legislature
The Honorable Fife Symington, Governor
Mr. Samual A. Lewis, Director
Arizona Department of Corrections
Transmitted herewith i s a report of the Auditor General, A Performance
Audit of the Department of Corrections, I n s t i t u t i o n a l Security and
Staffing. This report i s i n response to a June 14, 1989, resolution of
the Joint Legislative Oversight Committee.
The report i s a follow- up review that assesses actions taken by the
Department to implement recommendations made in two previous performance
audit reports. We found that DOC has made significant improvements in
i n s t i t u t i o n a l security, and in i t s classification and personnel systems.
My s t a f f and I w i l l be pleased to discuss or c l a r i f y items in this report.
The report w i l l be released to the public on May 24, 1991.
DRN : l mn
STAFF :
Sincerely,
v sR. N orton
Au tor General
William Thomson
Peter N. Francis
Lucinda A. Trimble
Jean Wood
Ned E. Parrish
Leonard P. Wojciechowicz
2700 NORTH CENTRAL. AVENUE SUITE 700 " PHOENIX, ARIZONA 85004 . ( 602) 255- 4385 " FAX ( 602) 255- 1251
SUMMARY
The Office of the Auditor General has conducted a performance audit of
institutional security and staffing at the Arizona Department of
Corrections ( DOC). The audit was conducted in response to a
June 14, 1989, resolution of the Joint Legislative Oversight Committee as
part of the Sunset Review set forth in Arizona Revised Statutes ( A. R. S.)
$ 941- 2351 through 41- 2379.
This report is a follow- up review that examines the steps taken by DOC to
implement recommendations made in two previous performance audit reports
issued by our Office in 1985 and 1986: Report No. 85- 12, an audit of
Adult Institutions Security; and Report No. 86- 1, Security Staffing
i ssues .
DOC Has Taken Siclnificant Stem To
lrn~ rove Securitv At Its
Adult Institutions ( see pages 3 through 10)
DOC has made significant improvements in both perimeter and internal
security since our last audit.
Perimeter Securi tv
- Fencing - DOC has increased the use of razor wire on i t s
perimeter fences.
- Electronic Detection Systems - DOC has addressed detection
system failures through improved maintenance.
- Escapes - DOC has reduced the number of escapes from 38 in 1986
to 12 in 1990.
Internal Security
- Lockina Svstems - DOC has replaced or refurbished broken
systems, including some that had deteriorated to the point that
in order to open one cel I door, as many as 26 c e l l doors had to
be opened at once.
- Commrnication Systems - DOC has increased the number of radios
and radio frequencies throughout the system.
Pol icies and Procedures
- Visitation - DOC has established security procedures and
eliminated or reduced the number of food v i s i t s .
- Nedication - DOC now requires that psychotropic medicines and
controlled substances be delivered in no more than a one- day
SUPP IY.
- Inmate ~ ersonal property - DOC now l i m i ts the amount and type of
personal property inmates may possess.
While marked improvements have been made, problems noted in some areas in
1985 s t i l l exist. These areas include tool control, inmate counts,
u t i l i z a t i o n of temporary structures, the need for additional isolation
cells, and f a c i l i t y designs.
New security issues were also identified during this audit. For
instance, security inspections were found to be inadequate at most
f a c i l i t i e s ; post orders were deficient at some institutions; key control
problems were identified at some f a c i l i t i e s ; and personal body alarms are
needed, particularly for security personnel in remote units.
DOC Has lm~ lemented A More
Obiective Inmate Classification
~ vstem ( see pages 11 through 13)
Since the last audit, DOC has implemented a more objective inmate
classification system. The Department's previous system underclassified
inmates leading to increased security risks by placing higher risk
inmates in lower security f a c i l i t i e s . The new system has been reviewed
and modified twice. Additionally, DOC plans to validate the system
during 1991.
DOC Has Made Improvements In Its
Personnel Svstem ( see pages 15 through 20)
With the implementation of a centralized hiring unit in 1985, DOC has
significantly improved the efficiency and effectiveness of Correctional
Service Officer ( CSO) recruitment. There appears to be l i t t l e more that
DOC can do to hasten the hiring process without the authority and funding
for advanced hi ring. Currently, i t takes approximately 11 weeks to f i l l
a position, much of that time ( seven weeks) is for pre- service training
at the Correctional Officers Training Academy. To reduce that time, DOC
would need the authority and funding to hire and train CSOs before
vacancies occur. However, this would require an additional expenditure
of $ 5.5 mi l lion annual ly.
In addition, DOC has also been able to reduce CSO turnover from
approximately 31 percent five years ago, to the current rate of 20
percent for CSO Is and 8 percent for CSO I l s . However, DOC o f f i c i a l s are
concerned that salary, transportation, and housing problems may
contribute to a future increase in turnover.
DOC Needs To Do More To Define
Securitv Staffina Ne& ( see pages 21 through 26)
As noted in the 1986 report, DOC needs to periodically review i t s
staffing formula. DOC has never recalculated the formula and has
con t i nua l l y requested the staff i ng l eve l s recommended i n 1986. However ,
we recalculated the formula and found that staffing needs have changed
since the last audit. For instance, in 1986 we recommended a staffing
level of 5.3 Full- Time Equivalent ( FTE) positions to cover a 24- hour
post; however, since 1986 this figure has declined to 5.1.
In addition to a periodic review of the staffing formula, DOC also needs
to do more to identify the number of posts needed at each institution.
While DOC has identified security posts, the Department has not completed
a comprehensive analysis of i t s security post needs. Such a review
should include inspecting post positions, physical plant and ongoing
program a c t i v i t i e s , and a review of staff assignment records, as we1 l as
the use and ef f i cacy of management systems.
While DOC Has Made Strides To Increase
In- Service Trainina. Manv CSOs Still Do
Not Meet Requirements ( see pages 27 through 32)
Since the last audit, DOC has established a training academy to provide
pre- service training and has increased the amount of in- service training
that i t provides to i t s correctional service s t a f f . A l l CSOs now receive
288 hours of pre- service training and are required to obtain 24 hours of
in- service training annually. However, a large number of staff s t i l l do
not meet the requirement for in- service training. We found that 44
percent of a l l CSOs in 1989- 90 did not obtain the required 24 hours.
Additionally, over 10 percent received no in- service training. Many
higher ranking security staff also failed to obtain the mandatory 24
hours .
Factors such as a lack of commitment at some institutions, staff
vacancies, and a lack of resources to implement an effective tracking
system may have hampered DOC'S efforts to provide training to a l l of i t s
correctional s t a f f .
Pane
1
FINDING I: DOC HAS TAKEN SIGNIFICANT STEPS
SINCE 1985 TO IMPROVE SECURITY
AT ITS ADULT INSTITUTIONS. . . . . . . . . . . . . . . . .
DOC Has Improved Security A t I t s Adult
Institutions Since 1985. . . . . . . . . . . . . . . . . .
Despite Marked Improvement In Security,
Some Problems Identified In
Our Previous Report Continue . . . . . . . . . . . . . . .
Several Additional Security
Deficiencies Were ldentified . . . . . . . . . . . . . . .
Recommendations. . . . . . . . . . . . . . . . . . . . . .
FINDING 11: DOC HAS IMPLEMENTED
A MORE OBJECTIVE IWTE
CLASSIFICATION SYSTEM. . . . . . . . . . . . . . . . . . .
DOC Has Implemented A New System To
Resolve Previous Problems With
InmateClassification. . . . . . . . . . . . . . . . . . .
DOC Has Reviewed And Modified
TheSystemTwice. . . . . . . . . . . . . . . . . . . . .
DOC Plans To Evaluate The System
On An Ongoing Basis. . . . . . . . . . . . . . . . . . . .
Recommendation . . . . . . . . . . . . . . . . . . . . . .
FINDING Ill: DOC HAS MADE IMPROVEMENTS IN
ITS PERSONNELSYSTEM. . . . . . . . . . . . . . . . . . .
DOC Has Done Much To Speed Up Hiring . . . . . . . . . . .
CSO Turnover Rate Has Decreased. . . . . . . . . . . . . .
Recomnendation . . . . . . . . . . . . . . . . . . . . . .
FINDING IV: DOC NEEDS TO DO UORE TO DEFINE
smuR1- w STAFFING NEEDS
DOC Should Review I t s Staffing Formula . . . . . . . . . .
Post Positions Should Be Reviewed. . . . . . . . . . . . .
Recommendations. . . . . . . . . . . . . . . . . . . . . .
TABLE OF CONTENTS con ' t
FINDING V: WILE DOC HAS UAM STRIDES TO INCREASE
l N- SERVl CE TRAIN1 NG, WNY CSOs ST l LL DO NOT
MEET REWIREYNTS. . . . . . . . . . . . . . . . . . . . .
While More Training Is Being Provided, Many
CSOs S t i l l Do Not Meet Requirements. . . . . . . . . . . .
Lack Of Commitment And Resources
May Have Hindered Training Efforts . . . . . . . . . . . .
Recomnendations. . . . . . . . . . . . . . . . . . . . . .
OTHER PERTINENT INFOWATION. . . . . . . . . . . . . . . . . .
Growing Impact of Compensatory Time. . . . . . . . . . . .
AGENCY RESPONSE
APPENDICES
LIST OF TABLES
TABLE 1 Department Of Corrections Adult
l n s t i tutions Escape Statistics
1986- 1990. . . . . . . . . . . . . . . . . . . .
TABLE 2 Comparison Of CSO Salaries With
Those Of Similar Positions In- State
And Out- Of- State . . . . . . . . . . . . . . . .
TABLE 3 Comparison Of 1985 And 1990
Office Of The Auditor General
Staffing Studies . . . . . . . . . . . . . . . .
TABLE 4 Department Of Corrections Percentage Of
In- Service Training Hours Received
By Correctional Service Staff
I n 1989- 90 By Emp l oyee
Classification . . . . . . . . . . . . . . . . .
TABLE 5 Department Of Corrections Percentage Of
In- Service Training Hours Received
By Correctional Service Staff
In 1989- 90 By Institution. . . . . . . . . . . .
Paae
The O f f i c e o f the Auditor General has conducted a performance audit of
i n s t i t u t i o n a l security and s t a f f i n g at the Arizona Department of
Corrections ( DOC). The audit was conducted in response to a June 14,
1989, resolution of the Joint Legislative Oversight Committee as part of
the Sunset Review set forth in Arizona Revised Statutes ( A. R. S.)
$ 541- 2351 through 41- 2379.
This report i s a follow- up review that examines the steps taken by DOC to
implement recommendations made in two previous performance audit reports
issued by our Office in 1985 and 1986: Report No. 85- 12, an audit of
Adult I n s t i t u t i o n s Security; and Report No. 86- 1, an audit of Security
Staffing Issues.
Backaround
DOC i s charged with serving and protecting the people of the State by
imprisoning those adult offenders legally committed to the Department.
In i t s mission statement, DOC states that i t w i l l accomplish i t s charge
by maintaining effective custody control over offenders; maintaining a
healthy, safe, and secure environment for s t a f f and offenders; and
providing quality programs to offenders.
As of February 1, 1991, there were approximately 13,900 inmates l i v i n g in
DOC'S nine i n s t i t u t i o n s . This i s a 66 percent increase over the number
of inmates incarcerated i n 1985. To accommodate the growing prison
population, DOC has added three new i n s t i t u t i o n s ( Douglas, Winslow, and
Yuma) since the l a s t audit. The prison system has also expanded through
the construction of new units in Florence, Tucson, Safford, and Winslow
and double- bunking in P e r r y v i l l e . Additional bed space has also been
creatsd through the modification of existing space ( such as day rooms).
Institutional Security
In 1985, we reported several deficiencies in the area of security at
adult i n s t i t u t i o n s : 1) perimeter and internal security was inadequate at
some DOC adu l t i ns t i tu t ions ; 2) con t raband was widespread and eas i ly
accessible within the prisons; 3) the Department's inmate c l a s s i f i c a t i o n
system did not properly categorize inmates, thereby increasing security
risks; and 4) the Department's construction program, at that time, did
not provide enough maximum custody beds to meet current or future needs.
Security Staff inq
In 1986, we reported several deficiencies in the area of security
s t a f f i n g : 1) DOC could not properly determine i t s security s t a f f i n g
needs; 2) DOC was not able to h i r e enough Correctional Service Officers
( CSOs) to meet i t s security s t a f f i n g needs; 3) DOC'S process for
background checks for Correctional Service Officer applicants needed
improvement; and 4) DOC had not provided Correctional Service Officers
with adequate training.
Sco~ e And Methodology
The primary focus o f t h i s audit was to follow up on recommendations made
in our previous reports in 1985 and 1986; and to ascertain whether
improvements in the areas of security and security s t a f f i n g have been
made. To assist in our review of i n s t i t u t i o n a l security, we contracted
with security consultants to conduct a comprehensive review of six of
DOC'S nine adult i n s t i t u t i o n s .
This audit was conducted in accordance with government auditing standards.
The Auditor General and s t a f f express appreciation to the Director and
s t a f f of the Department s f Corrections for their cooperation and
assistance during the audit.
FINDING I
DOC HAS TAKEN SIGNIFICANT STEPS
SINCE I985 TO IMPROVE SECURITY
AT ITS ADULT INSTITUTIONS
The Department of Corrections ( DOC) has made substantial progress in
upgrading i n s t i t u t i o n a l security in the past f i v e years. Consultants
retained by our Office found that a l l the recommendations in our 1985
report have been addressed to some extent.(') Although security e f f o r t s
have been enhanced, and significant improvements have been made in both
perimeter and internal security, some problems discussed in our previous
audit report continue to exist. In addition, during t h i s review, several
other security deficiencies were i d e n t i f i e d that need to be addressed to
ensure the safe operation of the Department's f a c i l i t i e s .
We f i r s t reviewed security at DOC i n s t i t u t i o n s in 1985. A t that time,
security e f f o r t s were found to be inadequate at some f a c i l i t i e s .
Perimeter security was i d e n t i f i e d as a major problem area. Perimeter
security deficiencies were found in fencing, electronic detection
systems, use of guard towers, and v i s i b i l i t y . A number of internal
security problems were also identified. For instance, locking systems at
some f a c i l i t i e s were found to be improperly designed and inadequately
maintained. Tool control was weak. Hacksaws, wire cutters, welding and
cutting torches, and other tools were not secured. Control rooms, the
" nerve centern of a prison, were poorly secured. Security personnel were
also found to lack s u f f i c i e n t radios and frequencies to provide for
proper inmate control.
( 1) We contracted with James D. Henderson and Associates, a criminal j u s t i c e consulting
firm, to assist with a follow- up review o f s e c u r i t y measures at the Department's adult
correctional f a c i l i t i e s . The consultant team v i s i t e d six DOC complexes: Florence,
Tucson, P e r r y v i l l e , Phoenix, Douglas, and Winslow. Together these i n s t i t u t i o n s
contain 26 inmate housing units, with a t o t a l inmate population of 12,522. The
consultant team evaluated perimeter security measures a t a l l units. They a1 so
examined a v a r i e t y of factors a f f e c t i n g i n t e r n a l security ( such as building design,
yard 1 i ghti ng, inmate count procedures, and t o o l control procedures). The Appendix
provides information regarding the consultants1 backgrounds and a sumnary of t h e i r
findings.
DOC Has Improved Securitv At Its
Adult Institutions Since 1985
During the past f i v e years the Department has s i g n i f i c a n t l y improved
security at i t s correctional f a c i l i t i e s . Our consultants found that
perimeter security now i s generally adequate at the i n s t i t u t i o n s
surveyed. In addition, some improvements in internal security were
noted. F i n a l l y , DOC has revised or added a number of security- related
policies and procedures.
Perimeter Security - DOC has improved perimeter security at i t s adult
i n s t i t u t i o n s since 1985. lmprovements were noted in the following areas:
Fencing - In 1985 we noted that perimeter fencing was inadequate at
most of the Department's adult i n s t i t u t i o n s . A t that time, fences
often were not secured in cement and lacked s u f f i c i e n t razor wire to
discourage escape attempts. In addition, the minimum custody prison
at Fort Grant, which had more escapes than any other prison complex
in the Arizona system, had no perimeter fence. Our security
consultants found the fencing now surrounding DOC i n s t i t u t i o n s i s
excellent. The Department now makes increased use of razor wire on
perimeter fences. In addition, in October 1986, to discourage escape
attempts, DOC instal led an eight- foot fence around the perimeter of
the Fort Grant prison complex.
Electronic Detection Svstems - Deficiencies in electronic perimeter
detection equipment were also noted in our 1985 report. in
p a r t i c u l a r , the systems at ASPC- Tucson and ASPC- Perryville had
persistent system failures. Auditor General security consultants
tested the electronic perimeter detection systems at the i n s t i t u t i o n s
they v i s i t e d in late 1990 and early 1991 and found that t h i s
equipment worked properly in most areas. In addition, most
f a c i l i t i e s that u t i l i z e electronic detection systems ( such as
microwaves and shaker wires) now have contracts with companies to
maintain t h i s equipment.
Esca~ es - The 1985 report indicated that perimeter security
deficiencies created opportunities for escapes. Between January 1984
and May 1985, 56 inmates escaped through the perimeters of DOC
i n s t i t u t i o n s . lmprovements in perimeter security since that time
appear to be responsible, at least in part, for a reduction in
- escapes from DOC's adult i n s t i t u t i o n s . As Table 1 ( see page 5),
shows, the number of escapes has declined s i g n i f i c a n t l y since 1986.
A t o t a l of 38 inmates assigned to DOC'S adult i n s t i t u t i o n s escaped in
calendar year 1986. In contrast, only 12 inmates assigned to DOC's
adult i n s t i t u t i o n s escaped in 1990. It i s worth noting that 83
percent of these escapes involved minimum custody inmates.
TABLE 1
DEPARTMENT OF CORRECTIONS
ADULT l NSTlTUTlONS
ESCAPE STATISTICS( a)
19861990
Adult Institutions( b)
Escapes 38 40 29 32 12
Apprehensions 36 36 25 26 10
Outstanding 2 4 4 6 2
( a) Tab1 e excludes escapes involving inmates assigned to re1 ease centers, work
furlough, or home arrest.
( b) Some of the escapes recorded i n the table do not involve a breach of perimeter
security, but occurred while inmates were temporarily outside of t h e i r i n s t i t u t i o n s
f o r a c t i v i t i e s such as work d e t a i l s or court hearings.
Source: Office of the Auditor General staff analysis of DOC Escape Log,
1986- 1990 and DOC Escapee Hotsheet.
Internal Securitv - A number of internal security problems identified in
our previous report have also been addressed by DOC. For example,
improvements in internal security were identified in the following areas:
Lockina S ~ s t m s - In 1985 we identified problems with the locking
systems at Florence's maximum custody Central Unit. The locking
system in cellblock 1 had shorted out and rusted due to leaky
showers; consequently, c e l l doors had to be opened manually. In
cellblock 2 and a portion of cellblock 4, staff were unable to open
individual c e l l doors without opening a l l c e l l doors on a t i e r . In
the case of cellblock 2 this involved opening 26 cells at one time.
These locking sys tem prob l ems have been addressed . The Department
replaced the locking mechanisms in cellblock 1 approximately three
years ago and recently refurbished the locking systems in cel lblock 2
and cellblock 4.
Communication Svstems - Significant communication system problems
were also found in 1985. Our previous report noted that security
personnel lacked sufficient radios and radio frequencies to provide
adequate inmate control. A t that time, housing officers in the
medium custody South Unit at Florence were not issued radios.
Further, complex security lacked a separate frequency for emergency
communication at some institutions. Now, the consultants report that
there are adequate numbers of radios and radio frequencies throughout
the system. South Unit housing officers are now issued radios to
reduce their vulnerability in case of an emergency. In addition,
most institutions now have a separate radio channel for complex
security personnel.
Policies and Procedures - A number of security- related policies and
procedures have been improved or added since our previous audit.
Improvements have been made in the following areas:
Visitation - Our 1985 report was c r i t i c a l of DOC'S monitoring of
v i s i t a t i o n and of DOC'S policies allowing v i s i t o r s to bring food into
v i s i t a t i o n areas. Both conditions provided an opportunity for the
introduction of contraband. In 1989, the Department revised i t s
v i s i t a t i o n policy and established security procedures for v i s i t a t i o n
to address these problems. In addition, the policy eliminated food
v i s i t s at medium and maximum custody f a c i l i t i e s and limited the
number of food v i s i t s at minimum custody units.
Control of Uedication - In 1985, inmates at some institutions were
issued medication in multiple doses, enabling them to possibly save
their medication for sale at a later date. Since that time, the
Department has adopted policies and procedures for delivering
medications to inmates. DOC now requires that psychotropic medicines
and controlled substances be delivered in no more than a one day
SUPP IY.
Inmate Personal Pro~ ertv - In our previous report, i t was noted that
inmates were allowed to possess an excessive amount of property,
making searches for contraband d i f f i c u l t . The Department's policy
was revised in 1988 to limit the amount and type of property inmates
were allowed.
Despite Marked lm~ rovernent In Securitv. Some Problems
Identified In Our Previous Re- port Continue
Although DOC has improved security at i t s adult institutions, a number of
problems identified in 1985 persist. Tool control remains a problem at
many DOC f a c i l i t i e s . The Department s t i l l does not require a sufficient
number of formal inmate counts. F a c i l i t i e s available for inmate
detention and isolation remain insufficient. Temporary structures
continue to be used for housing inmates at some i n s t i t u t i o n s . F i n a l l y ,
the design of some DOC housing structures s t i l l limits the s t a f f ' s
a b i l i t y to control inmates.
Tool Control - Our previous security audit found that tool control
policies and procedures were not consistently followed by institutional
staff and needed to be revised. At that time, potentially hazardous
tools were frequently observed to be unsecured. Our consultants found
that tool control continues to be a problem at many DOC f a c i l i t i e s . The
consultant team again observed unsecured hacksaws, blades, cutting tools,
and portable welding equipment. In addition, tool inventories were
sometimes found to be incomplete or did not exist. Finally, tools were
often stored incorrectly, making i t d i f f i c u l t to determine i f items were
missing.
Inmate Counts - Our 1985 report also noted that the number of formal
counts required by DOC policy was insufficient. The policy required only
three formal inmate counts each day. The Department's policy in 1991
continues to require only three formal counts daily. Our consultants
also indicated that the length of time between counts is too long,
reducing the s t a f f ' s a b i l i t y to account for inmates. They recommend that
DOC increase the number of formal counts from three to five a day.
Detention and Isolation Cells - The number of detention and isolation
cells continues to be insufficient at some institutions. In 1985, DOC
had only 134 isolation cells, accounting for less than 2 percent of the
Department's permanent bed capacity. A t that time, we reported that the
number of cells available for inmate lockup was not adequate to ensure
effective inmate discipline.
DOC con t i nues to have an i nadequate supp l y of lockup ce l l s at some
institutions. Both the consultant team and an o f f i c i a l from the American
Correctional Association estimate that at any given time, 10 to 13
percent of the inmate population w i l l require some form of locked- status
housing for purposes such as detention or isolation.(') However, the
Department currently has only 367 cells for detention and isolation,
enough for only 2.6 percent of the total inmate population. In addition,
because only limited space is available, inmates in locked status
frequently share a c e l l . The consultants believe this practice limits
the s t a f f ' s a b i l i t y to adequately control inmates.
Temporary Structures - Our previous audit report noted that temporary
structures were widely used for inmate housing at DOC f a c i l i t i e s . The
use of tents, quonset huts, and t r a i l e r s for housing inmates was
c r i t i c i z e d because these structures lacked basic security features.
Although the use of tents has been discontinued, temporary structures are
( 1 ) An inmate may be placed i n locked- status housing f o r disciplinary reasons if he or she
i s found gui 1 ty of v i o l a t i n g departmental rules. Additional 1 y, 1 ock- up c e l l s are used
to i s o l a t e inmates who are considered a risk to the general inmate population or an
escape r i s k .
7
s t i l l being used at the ASPC- Florence East and North Units, the
ASPC- Tucson Echo Un i t , the ASPC- Doug l as G i la Un i t , and ASPC- Sa f ford .
Quonset huts and modular buildings are not designed to provide adequate
observation of inmates. These structures are also not designed to serve
as security barriers. The security consultants indicated that inmates
could not be secured within their quonset huts during a disruption.
F a c i l i t v Desi~ n - Some permanent inmate housing structures continue to
pose risks to inmates and staff because of their design. In our 1985
audit, ASPC- Florence's South Unit was rated by security consultants as a
low, medium- custody f a c i l i t y because i t has dormitory- style housing and
no s i ng l e ce l l s . The South Un i t cont i nues to be used to house " heavy",
medium- custody inmates in a dormitory setting. Our consultants noted
that the design of these housing structures limits the s t a f f ' s a b i l i t y to
observe a l l inmate a c t i v i t y . The consultants were also c r i t i c a l of the
design of inmate housing structures at the ASPC- Douglas Mohave Unit,
because i t contains a series of small rooms in which inmates could hide
or isolate security personnel.
4
Deficiencies Were ldentif ied
Our current review identified several security problems that were not
discussed in our previous report. The consultant team found that
security inspections were inadequate at most DOC f a c i l i t i e s . In
addition, post orders were found to be deficient at some institutions,
key control problems were i d e n t i f i e d a t some f a c i l i t i e s , and body alarms
were not issued to staff at any of the institutions in the State prison
system.
Security Inspections - Although DOC policy requires the routine
inspection and maintenance of security devices, the policy is not readily
adhered to by institutional s t a f f . The Department's policy requires
line and supervisory staff to inspect security devices such as gates,
fences, locking mechanisms, and alarm systems during each s h i f t . The
consultants found that " security inspections are not generally being
carried out in a consistent, complete, well- defined manner throughout the
institutions visited." In addition, security consultants found that DOC
institutions have not developed standardized methods for recording and
reporting the security inspections conducted by s t a f f .
8
Post Orders - Post orders were found to be deficient at a number of
institutions. Our consultants found that the " nature and quality of post
orders varied from i n s t i t u t i o n to institution." The consultant team
indicated that post orders frequently lacked information regarding
hostage situations, the use of force, and emergency evacuations.(') In
addition, post orders sometimes did not contain a face sheet for
employees to record that they had reviewed the post orders.
Key Control - Auditor General security consultants found that few DOC
f a c i l i t i e s have established comprehensive key- control systems. Master
key inventories were found to be inaccurate or incomplete at several
prison complexes. Listings of the number and type of keys assigned to
each key ring sometimes were not maintained in the control rooms where
the keys are issued. Keys were sometimes stored improperly, making i t
d i f f i c u l t to determine i f key rings were missing. Key rings often were
not properly soldered to deter the removal of keys from them. Emergency
keys were sometimes not marked in such a way as to f a c i l i t a t e their use.
In addition, emergency keys were not regularly tested to ensure that they
worked properly. A t one institution, emergency keys failed to provide
access to the complex detention unit.
Body Alarms - The consultant team found that the Department does not
issue body alarms to staff at any DOC f a c i l i t y . This equipment, which
can be connected to an o f f i c e r ' s radio, is designed to sound an alarm in
the control center when an officer i s i n trouble. The consultants
believe body alarms are a " cost effective method of providing personal
protection for staffn and should be issued to security staff in certain
locations. Body alarms should f i r s t be issued to staff in locked units
and those on the more remote assignments.
( 1) DOC has established Departmental policies on hostage situations and use of force.
However, the consultants be1 ieve these policies should be incorporated i n the post
orders to ensure they are reviewed frequently by s t a f f .
RECOMMENDATIONS
The Department of Corrections should revise several of its policies
and procedures to reduce risks to both inmates and staff. Policies
that require revision include security inspections, post orders,
inmate counts, tool control, and key control.
The Department should determine the amount of additional lock- up
space that is needed at DOC institutions and request additional
funding for the construction of these facilities.
The Department should assess the need for body alarms for security
staff and request additional funding for this equipment.
FINDING II
DOC HAS IMPLEMENTED A MORE OBJECTIVE
INMATE CLASSIFICATION SYSTEM
DOC has improved inmate c l a s s i f i c a t i o n since the last audit. A new
system was implemented in 1986 to resolve problems with the previous
c l a s s i f i c a t i o n system. Since implementation, DOC has reviewed and
modified the system twice and plans to continually evaluate the system.
Our 1985 review of DOC found that inmates were often misclassified
causing increased security risks and other problems. Problems with
c l a s s i f i c a t i o n occurred because of the subjective nature of the
c l a s s i f i c a t i o n methods then in use. The Department's c l a s s i f i c a t i o n
system lacked a formal model for making decisions regarding an inmate's
custody level and i n s t i t u t i o n a l assignment. For instance, although s t a f f
were required to review the results of psychological and s k i l l s tests
p r i o r to making c l a s s i f i c a t i o n decisions, there were no guidelines for
assigning weights to the test results or considering them in a structured
manner. A t the time of the audit, DOC was in the process of developing
an objective system.
The Offender Classification System ( OCS) was i n i t i a t e d by the Department
in August 1986 to resolve problems with the c l a s s i f i c a t i o n methods that
were in use at the time. Developed with the assistance of an outside
consultant, the c l a s s i f i c a t i o n system i s based on the Correctional
Classification P r o f i l e , which involves the development of a series of
numerical scores for ten factors related to inmates' risk- potential and
needs. These factors include public and i n s t i t u t i o n a l r i s k ; medical and
mental healthcare needs; work s k i l l s , educational and vocational training
needs; alcohol, drug, and sexual offense treatment needs; and proximity
to residence. Classification guidelines and procedures were outlined in
the OCS Operating Manual. DOC s t a f f received training in the new
c l a s s i f i c a t i o n procedures once the system was ready for implementation.
DOC Has Reviewed And Modified The
Current Classification Svstem Twice
The Department has reviewed and modified the classification system twice
since the 1986 implementation. The Department f i r s t reviewed i t s new
classification procedures in March 1987. The OCS Operating Manual was
modified to address problems identified in the months following the
implementation of the system. For instance, DOC decreased the number of
risk score levels from ten t o f i v e to reduce the complexity of the
system. However, this change affected the distribution of the inmate
population, rapidly increasing the proportion of the inmate population in
the lower public- risk custody levels.
In 1989, DOC again reviewed i t s inmate classification system. The
evaluation was prompted by concerns that too many inmates had been
classified below the appropriate risk levels. An independent consulting
agency was asked to assist with the evaluation. The consultants, at that
time, also validated the system. A number of recommendations were made
to further improve the Department's classification system, including
increasing the amount of time an inmate must serve before being
considered for a lower public- risk score.
DOC Plans To
Evaluate The Svstem
On An Onuoina Basis
The Department plans to continually evaluate the v a l i d i t y of i t s
classification system. DOC staff recently attended a National Institute
of Corrections training program on validating inmate classification
systems. The information obtained from this training w i l l assist DOC in
constantly reexamining the system to ensure i t s effectiveness.
While DOC has implemented a more objective classification system, proper
classification of inmates w i l l continue to be hampered by limitations of
bed a v a i l a b i l i t y . Many DOC institutional units are currently operating
near capacity. In addition, the prison population continues to
increase at a rate of approximately 75 inmates per month.(') The
pressures of the increasing prison population make i t d i f f i c u l t for the
Department to place inmates in the f a c i l i t y most appropriate for their
needs and r i s k potential.
RECOMMENDATION
DOC should continue i t s e f f o r t s to evaluate i t s c l a s s i f i c a t i o n system to
ensure i t s effectiveness.
( 1) DOC projects that the prison population w i l l increase a t a rate of 75.1 inmates per
month through December 1992. However, the Department projects that the long- term
growth rate at DOC f a c i l i t i e s w i l l be 89 inmates per month.
FINDING Ill
DOC HAS MADE IMPROVEMENTS
IN ITS PERSONNEL SYSTEM
In addition to addressing security deficiencies, DOC has also made
improvements in several other areas since the l a s t audit. The s t a f f
recruitment and selection process has been strengthened and l i t t l e more
can be done to speed up the h i r i n g process without the authority for
advanced h i r i n g . The Department also has reduced s t a f f turnover.
DOC Has Done Much
To Speed UD Hirinq
The implementation of a centralized h i r i n g unit in 1985 has vastly
improved the efficiency and effectiveness of CSO recruitment, selection,
and hiring. DOC'S Recruitment Unit for Selection and Hiring ( RUSH) has
replaced the previously i n e f f i c i e n t method of h i r i n g Correctional Service
Officers. DOC can do l i t t l e more to hasten the h i r i n g process without
the authority and funding for advanced h i r i n g .
In our 1985 audit, we reported that DOC'S h i r i n g process was inadequate
to meet i t s h i r i n g needs. A t that time, the responsibility for
developing procedures governing the recruitment, selection, and h i r i n g of
CSOs was the responsibi l i ty of the Department of Administrat ion's ( DOA)
Personnel Division. However, DOC had established RUSH and was to assume
r e s p o n s i b i l i t y f o r the process. The establishment of RUSH in 1985 has
greatly improved the process for recruiting, selecting, and h i r i n g a
large number of personnel on an ongoing basis. Through an
intergovernmental agreement with DOA, DOC now has direct authority to
r e c r u i t , select, and h i r e entry- level CSOs.
DOC begins the h i r i n g process by forecasting annual s t a f f i n g needs. RUSH
u t i l i z e s a forecasting plan to predict when and where vacancies w i l l
occur. Through analysis of h i s t o r i c a l data on CSO a t t r i t i o n , RUSH, in
conjunction with the Correctional Officer Training Academy ( COTA), i s
able to anticipate workload, institutional hiring needs, and the timely
flow of cadets into training.
Efforts to recruit staff to f i l l forecasted positions are more aggressive
and innovative now than in the past. In addition to basic newspaper
advertising, the Department also sponsors job fairs. Additionally, DOC
has acquired the services of an advertising agency that specializes in
employee recruiting. Specifically, the agency assists DOC in conducting
needed research and design, writing, and advertising. Advertising
methods u t i l i z e d include newspaper advertising, public service
announcements, and posters. This new approach is aimed at increasing
both the number and quality of CSO applicants.
Application processing and candidate screening are also more thorough and
e f f i c i e n t . When RUSH receives an application ( which is directly
forwarded from the institutions, the DOA Personnel Office, or any DES job
service office), the application is reviewed to determine i f the
applicant meets the minimum qualifications. The applicant is then
notified within 3 to 4 days regarding his or her qualifications and is
scheduled to appear for testing. Testing is much more comprehensive than
i t was prior to RUSH. Applicants must take a written exam, a
psychological exam, a medical exam, undergo a background check, and have
an interview.(') Once applicants have successful ly completed the enti re
screening process, their names are placed on hiring l i s t s awaiting notice
of vacancy by the institutions. As vacancies occur, candidates are
selected for pre- service training at the academy.
In addition to the establishment of RUSH, there have been other
improvements that have increased the efficiency of the hiring process.
In our 1985 audit, it was noted that background checks of applicants are
untimely, resulting in delays of more than 30 days. Decentralization of
the background check responsibility was found to contribute to the
de lays . Add i t i ona l l y, i t was noted that app l i cant process i ng was
unnecessar i l y delayed wh i l e background checks were conducted. S i nce that
audit, DOC has improved the timeliness of background checks. The process
( 1) Prior to RUSH, the screening process included only a background check and an interview
a t ' the i n s t i t u t i o n .
is now centralized and takes an average of 5 work days.(') However, to
avoid any further delays, RUSH continues the hiring process during the
background check.
Without the authority and funding to implement an advanced hiring
program, DOC can do l i t t l e more to speed up the hiring process.
Currently, i t takes approximately 11 weeks to f i l l a position.( 2) Seven
weeks are required for pre- service training at the academy. The
remaining four weeks include about two weeks to process the position
request and another two weeks for the applicant to report to the academy.
Advanced hiring would allow DOC to f i l l vacancies almost immediately by
hiring and training CSOs before vacancies occur. Although DOC has
continually requested funding for advanced hiring, the high cost of the
program has prohibited annual appropriations from the Legislature. DOC
estimates that i t would cost approximately $ 5.5 mi l l ion annual ly to fund
230 positions for advanced hiring.
CSO Turnover Rate
Has Decreased
Steps have been taken to reduce turnover in recent years. However, DOC
o f f i c i a l s remain concerned about turnover rates and the factors that
could increase turnover in the future.
During our last audit, DOC was experiencing high turnover of Correctional
Service Officers. In fiscal years 1984 and 1985, the turnover rate for
CSO positions was approximately 32 and 31 percent, respectively. In
contrast, the turnover rate for a l l State employees was 20 percent and 18
percent for the same years. Additionally, the CSO turnover rate was also
higher than the average rate for a l l positions within DOC ( 26 percent).
( 1) DOC also implemented a policy i n 1988 that established guidelines f o r d i s q u a l i f y i n g
CSO applicants. RUSH i s responsible f o r determining qualifications based on a
background check by the Inspections and Investigations Division.
( 2) DOC s t a f f report that i t took up to 20 weeks or longer to f i l l vacancies p r i o r t o the
establishment of RUSH.
Since the release of our last audit, some steps have been taken that have
contributed to a decrease in CSO turnover.
DOC has established an additional level for CSOs. In fiscal year
1986, a CSO I I position was establ ished to provide more of a career
ladder for Correctional Service Officers.
A centralized promotional system was established in 1986 that allows
for more objectivity in the promotion process.
Approximately five years ago, CSOs were given an automatic salary
increase of 2.5 percent once they graduated from COTA.
In 1986, CSOs were offered the opportunity of participating in a
newly established correctional officer retirement plan that provided
more of a career incentive for CSOs because i t al lows them to r e t i r e
after 25 years of service, regardless of their age.
Based on concerns noted in an employee survey, DOC provided basic
supervisory training for i t s supervisors to help improve working
relationships between supervisors and their s t a f f .
Since these steps were taken, turnover has dropped to 20 percent for
CSO Is and is about 8 percent for CSO I l s . Additionally, as of
December 31, 1990, this rate had dropped even further to an annualized
rate of 15 percent for CSO Is and 6 percent for CSO I Is, which is close
to the current State service turnover rate of 11 percent.
Despite reductions in turnover, DOC o f f i c i a l s remain concerned about
factors that may contribute to a future increase in turnover. One factor
that has always been a concern is CSO salary. Exit questionnaires
completed by CSOs reveal that many s t a f f , even though they enjoy their
job, resign due to the low salary. As shown in Table 2 ( see page 191,
current starting salaries for CSOs are not competitive with similar
positions both in- state and out- of- state. Concerns about salary have
also been expressed by the Department of Administration. DOA has
recommended that the Legislature fund an alternative salary plan for CSO
Is that, beginning January 1, 1992, would increase the starting salary
after graduation from the academy.
TABLE 2
COMPARISON OF CSO SALARIES WITH THOSE OF
SIMILAR POSITIONS IN- STATE AND OUT- OF- STATE
California Dept. of Corrections $ 29,436 ( after academy)
Pima County ( Detention Officer) 22,233 ( entry level)
Federal Bureau of Prisons 21,023 ( after 6 months)
Maricopa County ( Detention Officer) 20,966 ( after probat ion)
Nevada Dept. of Prisons 20,005 ( after 1 year)( a)
Arizona Dept. of Corrections 18,199 ( after academy)
( a) Nevada also pays 100 percent o f the employee's retirement contribution.
Source: Office of the Auditor General February 1991 survey of CSO
salaries.
Two other factors -- transportat ion and housing -- have emerged as
factors that can and do impact turnover. DOC has found that, in addition
to salary, transportation is a reason some CSOs resign. For instance,
according to a DOC o f f i c i a l , almost 500 of Florence's 1,700 employees
commute round t r i p approximately 140 miles a day from Tucson. Some
employees feel the cost of commuting is too high, especially when the
salary is considered to be insufficient. In addition, inadequate housing
in some prison locations contributes to the transportation problem. In
some areas of the State, such as Winslow, housing is limited. According
to the warden in Winslow, the biggest problem facing the prison is the
lack of housing for the s t a f f . The d i f f i c u l t y i n attracting staff to
some of these locations is evidenced by the incentive pay provided for
some pos i t ions . DOC provides an add i t i ona l 10 percent " i ncen t i ve payt' to
attract Correctional Service sergeants and captains. However, similar
incentive pay is not offered to CSO Is and CSO I l s .
RECOMMENDATION
1. The Legislature should consider increasing the starting salary of
Correctional Service Officers to allow DOC to be more competitive.
FINDING IV
DOC NEEDS TO DO MORE TO DEFINE
SECURITY STAFFING NEEDS
DOC needs to do more to define security staffing requirements. DOC needs
to periodically recalculate i t s formula for staffing to reflect current
conditions. Additionally, while DOC has taken some steps toward
identifying post positions, more needs to be done.
Security staffing needs in a correctional i n s t i t u t i o n depend on two
factors: 1) the number of posts required to maintain adequate security,
and 2) the number of people required to cover each post. Specific posts
are generally established on the basis of the need to monitor and control
inmate movement and a c t i v i t y . Thus, the number and types of posts vary
throughout the day and also from day to day, reflecting changes in the
level and type of inmate a c t i v i t y . Coverage for a particular post ranges
from 24 hours a day, seven days a week, to eight hours a day, five days a
week or less. Once necessary posts are determined, the number of people
needed to cover those posts is computed from a staffing formula. A
staffing formula determines the number of people required to cover a post
and takes into account the amount of time employees are expected to be
away from the post due to vacations, illness, etc.
In January 1986, the Office of the Auditor General issued a performance
audit report on DOC's Security Staffing ( Report # 86- 1). That report
concluded that DOC could not accurately determine i t s security staffing
needs for two reasons. F i r s t , DOC's formula for determining the number
of pos i t i ons needed to cover a 24- hou r post was not accurate and had not
been. updated for nine years. Second, a comprehensive post analysis had
never been performed. This type of analysis is necessary to assess the
impact of the size, nature, and design of each f a c i l i t y , the inmate
custody level, and several other factors on staffing needs. We
recommended that DOC review the staffing formula annually and undertake a
comprehensive staffing study to identify post needs and periodically
assess changes in conditions affecting staffing requirements.
DOC Should Review
Its Staffina Formula
Our analysis of current security staffing reveals that a periodic review
of DOC'S staffing formula is essential. Conditions have changed since
our 1985 report, and we found that staffing needs per post now are not as
high. For example, DOC has made improvements in reducing staff turnover
( as noted in Finding I I I) that have consequently reduced the turnover
factor in the formula. DOC should periodically review i t s formula to
account for changing factors.
DOC reauests for staffina levels have not reflected chanaes in staffinq -
Although DOC has continually requested the staffing levels recommended by
the Office of the Auditor General in the 1985 audit, these requests do
not reflect the current level of security staffing. Since our last
audit, DOC'S requests to the Legislature for funding for security
staffing have been based on the formula of 5.3 people to cover one
24- hour post. However, this staffing has not been authorized by the
Legislature.
We recalculated DOC'S current staffing formula and found that staffing
requirements have changed since the last audit. We reviewed a
s t a t i s t i c a l sample of payroll records for 341 Correctional Service
Officers ( CSO Is and I Is) to determine present post staffing needs.(')
The results of this recalculation showed that the number of staff needed
to cover a 24- hour, seven- day post has declined from 5.3 to 5.1. Table 3
( see page 23), i l l u s t r a t e s the changes in various categories since 1985
that have affected staffing. For example, DOC has been able to reduce
turnover for CSO Is and I Is from 31 percent five years ago to an average
of 18 percent. This reduced turnover has impacted the length of time
needed to f i l l vacancies and thus has had an impact on the results of the
calculation. While some factors have changed, DOC has never recalculated
i t s staffing formula. As stated in our last audit, because the formula
is sensitive to unstable factors such as turnover, i t should be regularly
updated to reflect current staffing needs.
( 1) Using a population of 3,095 CSO I s and 11s employed a t DOC between October 1989 and
October 1990, we selected a sample s i z e o f 341 employees based on a 95 percent degree
of confidence and a f i v e percent margin o f e r r o r .
TABLE 3
COMPARISON OF 1985 AND 1990
OFFICE OF THE AUDITOR GENERAL STAFFING STUDIES
Catesorv
Regular days
Annual leave
o f f
Sick leave
Holiday leave
Compensatory time taken
M i l i t a r y leave
Absence without approval
Administrative leave w/ pay
Administrative leave w/ o pay
Bereavement leave
Civic duty
Industrial leave
Education leave
Time to f i l l vacancies
Total number of days
employee was away from post
Avera~ e Days Per Position
1985 Studv 1990 Study
104.00 days 104.00 days
14.61 13.72
6.66 8.16
5.68 4.90
2.75 4.75
.39 1.12( a)
.15 .13
.04 .01
.26 1.21
.16 .38
. O1 .05
1.05 1.22
5.00 3 . OO( b)
18.24 8.37( c)
Calculation of Staffina Formula
365 days per year 365 365
Minus number of days
emp l oyee away f rom post - 159 - 148.9
Equals number of days
employee available to
cover post 206 216.1
Employees required:
Seven- day , 8- hou r sh i f t
Seven- day, 16- hour s h i f t
Seven- day , 24- hou r sh i f t
( a) Payroll records reviewed did not include the time period of the Gulf War. Therefore,
current m i l i t a r y leave time may have increased and would have had a greater impact on
the s t a f f i n g formula than was reflected i n our review.
( b) I n the 1985 audit in- service t r a i n i n g was based on the ACA standard of 40 hours a year
because a t that time DOC had no o f f i c i a l t r a i n i n g policy. While our review of payroll
records revealed an average of .85 days per position taken f o r training, DOC o f f i c i a l s
indicated that payroll records do not always capture time taken for in- service
training. Our review also showed some t r a i n i n g occurs on days o f f and i s charged as
compensatory time. Because actual t r a i n i n g time i s not available, we have
incorporated the current DOC requirement of 24 hours i n t o the 1990 s t a f f i n g formula.
( c) This category i s a combination of time spent during pre- service t r a i n i n g ( i. e .. COTA)
and the combined turnover rate f o r CSO I s and 11s. To a r r i v e at the 1990 calculation,
the time required to f i l l vacancies ( 9.3 weeks) was m u l t i p l i e d by 5 ( the number of
regular work days per week ) to a r r i v e at the number of days l o s t for each position
each time turnover occurs. This number was then mu1 t i p l i e d by the turnover rate ( 18
percent).
Source,: Office of the Auditor General review of 341 DOC payroll records
for CSO Is and CSO I l s during fiscal year 1989- 90; DOC staff
interviews; Auditor General Report No. 86- 1; and a review of
DOC'S process for f i l l i n g vacant CSO positions.
23
Post Positions
Should Be Reviewed
In addition to a periodic review of i t s staffing formula, DOC needs to do
more to identify the number of posts needed at each i n s t i t u t i o n . While
DOC has performed some staffing studies, these studies have been based on
internal estimates of work a c t i v i t y and justifications for existing
posts. A more c r i t i c a l review is needed.
Post wsitions have been identified - Since our 1985 audit, DOC has
identified i t s security post positions. The 1985 audit revealed that DOC
did not have an accurate l i s t of the Department's posts on which to
evaluate i t s security post requirements. Recently, each warden
identified a l l security post positions in his institution, and t e s t i f i e d
about post functions to the House Subcommittee on Prison Staffing. This
information was designed to j u s t i f y existing posts.
More can be done to quant i fv post needs - Wh i l e secur i ty posts have been
identified, DOC has not completed a comprehensive analysis of
Departmental post staffing requirements that includes the c r i t i c a l
elements recommended in our 1985 report. The 1985 audit recommended a
comprehensive study of staffing to determine the number of security posts
that were needed, based on workload measures, detailed task analyses,
s t a f f u t i l i z a t i o n , and other factors affecting security staffing.
Although DOC has conducted some staffing studies, the Department has not
performed a comprehensive study. We examined several DOC staffing
studies such as a 1985 DOC Staffing study,(') and a Security Staff
Analysis of ASPC- Douglas in July 1987. However, these studies rely on
estimations of workload and institutional j u s t i f i c a t i o n of post positions
rather than a c r i t i c a l review of each post. Some states, such as
California and Florida, have a central office function that performs
staffing studies and validates post positions on a routine basis. For
example, Florida has a full- time staffing coordinator to perform staffing
studies and review the functions of each post task and position in each
i n s t i t u t i o n .
( 1) This study was conducted at the same time as our 1985 audit and, according to DOC
s t a f f , i t was performed as a r e s u l t of the recornnendations i n that audit.
Additionally, our security consultants in their observations of security
staffing, noted that some housing units appeared to be understaffed. As
a result, they have strongly recommended conducting an independent
staffing study involving the analysis of post positions, physical plant
and ongoing program a c t i v i t i e s , a review of staff assignment records, as
well as the use and efficacy of management systems. Our consultants
estimate the cost of such a study would be approximately $ 60,000.
Chanaina conditions im~ act staffing - The need for a staffing review
assumes greater importance when other factors impact the staffing of
security posts. For example, in 1990 DOC installed a telephone
monitoring system that requires the on- line monitoring and recording of
inmate telephone calls. DOC supervisors state that this task requires a
considerable amount of CSO staff time. One i n s t i t u t i o n reports that two
CSOs are required t o s t a f f the phone lines six hours a day. Because this
function is not defined as a post position, the duties are absorbed by
current staff positions, thereby impacting the staffing of both the
defined and undefined posts. Further, the u t i l i z a t i o n of security staff
in nonsecurity positions impacts post position staffing. We found, for
example, that CSOs are being u t i l i z e d as locksmiths, timekeepers and
training officers. A comprehensive analysis of staffing would measure
the workload of each post and identify the type sf staff best suited to
performing each task.
RECOMMENDATIONS:
1. DOC should recalculate the staffing formula periodically.
2. DOC should request funding from the Legislature to have an
independent comprehensive staffing analysis performed to determine
the following:
the number of security posts needed, basedon workload analysis;
and
staff qualifications to determine i f CSO staff are appropriately
ass i gned .
3. DOC should also review post positions periodically to assess changes
affecting staffing.
FINDING V
WHILE DOC HAS MADE STRIDES TO INCREASE
INSERVICE TRAINING. MANY CSOs STILL
DO NOT MEET REQUIREMENTS
A1 though DOC has made attempts to provide sufficient in- service training
for i t s CSOs, further improvement is needed. While the average number of
training hours has increased, many CSOs s t i l l have not met current DOC
in- service training requirements. The reasons in- service training has
been inadequate may be lack of commitment by institutions to ensure
training is received and staff shortages.
Since the duties of correctional officers can impact their safety, as
well as the safety and well- being of inmates and sometimes the public,
proper training is essential. To receive proper training, the American
Correctional Association ( ACA), in cooperation with the Commission on
Accreditation for Corrections, recommends that correctional officers
receive 40 hours of training each year after their i n i t i a l year of
training. The ACA recornends that the training program include areas
such as security procedures, inmate supervision, detection of the signs
of suicide risk and precautions against suicide, the use of force
regulations and tactics, f i r e and emergency procedures, firearms
training, key control, and f i r s t aid.
In our 1985 audit, some CSOs had not received pre- service training prior
to assignment to an institution. In addition, others did not receive
sufficient in- service training to meet DOC'S training requirement of 20
hours. For fiscal years 1984 and 1985, CSOs employed by DOC for more
than two years had received an average of less than 13 hours of
in- service training.
While More Trainina Is Beina Provided,
Manv CSOs Still Do Not Meet Requirements
Although problems with pre- service training have been addressed, and
in- service training for CSOs has increased since our 1985 audit, many
CSOs s t i l l do not meet DOC training requirements. Furthermore, even when
staff do receive training, it may not be enough.
2 7
Pre- service trainina i s now provided to a l l 0 s - Since the 1985 audit,
DOC has implemented i t s Correctional Officer Training Academy ( COTA),
which provides pre- service training to a l l new CSOs. Prior to that time,
training was l e f t to individual institutions, and some CSOs received
l i t t l e or no pre- service training. DOC'S current pre- service training
consists of a 288- hour, seven- week program encompassing nine major
subjects: ethics and professionalism, inmate management, legal issues,
communication s k i l l s , officer safety, security, custody and control,
emergency procedures, conflict and c r i s i s management, and medical and
mental health. The f i r s t COTA class was held in January 1986. Since
that time the academy has conducted about 130 classes, and approximately
4,500 cadets have graduated and been assigned to institutions throughout
the Department. Once assigned to an institution, COTA graduates are
required to attend an additional 40 hours of on- site orientation.
More trainina i s beina ~ rovided - In addition to ensuring that staff
receive pre- service training, i t also appears that DOC has increased the
amount of training provided to i t s s t a f f . In 1989 DOC adopted i t s
current policy of requiring that CSOs receive a mandatory 24 hours of
in- service training annually. To meet this requirement, DOC provides
annual CSO refresher courses. These " refreshers" are generally taught in
blocks of 3 to 5 days and are provided throughout the year. In addition
to refresher courses, staff are provided with the opportunity to obtain
training in other areas as well. Also, as new issues develop, training
in these areas becomes mandatory.
Our review of 1989- 90 training records(') for Correctional Service staff
reflects an improvement in the area of staff training. For 1989- 90,
training received by Correctional Service staff employed more than two
years in each i n s t i t u t i o n on average ranged from 20.7 hours to 71.5
hours. Overall, in 1989- 90, those security staff employed more than two
years received an average of over 33 hours of advanced training. In
addition, while correctional staff are required to obtain 24 hours of
training each year as part of thei r refresher courses, some institutions
are providing more than 24 hours.
( 1) Because DOC training requirements are based on an Employee Performance Appraisal
System ( EPAS) year, our review of 1989- 90 records i s based on the most recently
completed EPAS year ( May 1, 1989 to April 30, 1990).
hiany s t a f f have not received s u f f i c i e n t training - Although the average
number of training hours received has increased, and most staff attend
training in excess of the 24- hour requirement, many staff s t i l l do not
meet the training requirements. Our analysis of the training hours
received by 1,970 security staff during 1989- 90 revealed the following:
Forty- four percent of a l l CSOs did not receive the required 24 hours.
Over ten percent of a l l CSOs received no training.
Many higher ranking security staff also failed to receive the
mandatory 24 hours of training. Table 4 i l l u s t r a t e s by employee
classification the percentage of security staff that received
training during 1989- 90. Further, Table 5 ( page 30) i l l u s t r a t e s by
i n s t i t u t i o n the percentage of security staff that received training
in 1989- 90.
In addition, DOC may not be able to identify the specific subjects in
which individual CSOs have received training. CSOs that attend the
refresher courses are given credit for receiving a " CSO Refresher" on
their training records in the Arizona Personnel Payroll Leave Education
System ( APPLES). We found that the type of training provided in the
refresher courses varies by institution. While some subjects such as
c e l l area searches, report writing, f i r s t aid, staff/ inmate relations,
and the use of force are offered at most institutions, not a l l
institutions include these subjects in their annual refresher courses.
TABLE 4
DEPARTMENT OF CORRECTIONS
PERCENTAGE OF IN- SERVICE TRAINING HOURS RECEIVED
BY CORRECTIONAL SERVICE STAFF IN 1989- 90
BY EMPLOYEE CLASSIFICATION
Classification 0 Hours 1 - 23 Hours 24+ Hours
cso I
cso I I
Sergeant
Lieutenant
Captain
Major
A l l Classes 12.9% 30.7% 56.4%
Source: Office of the Auditor General staff analysis of training hours
received by 1,970 DOC security s t a f f i n 1989- 90.
TABLE 5
DEPARTMENT OF CORRECTIONS
PERCENTAGE OF IN- SERVICE TRAINING HOURS RECEIVED
BY CORRECTIONAL SERVICE STAFF IN 1989- 90
BY INSTITUTION
Institution 0 Hours 1 - 23 Hours 24+ Hours
Doug l as
F lorence
Fort Grant
Perryvi l le
Phoen i x
Saf ford
Tucson
Winslow
Yuma
A l l Institutions 12.9% 30.7% 56.4%
Source: Office of the Auditor General staff analysis of training hours
received by 1,970 DOC security staff in 1989- 90.
DOC'S 24- hour annual trainina requirement may not be enoush - Although
DOC has increased the required number of hours for in- service training
since our 1985 audit, additional training requirements may be necessary.
As noted earlier, DOC'S annual training requirement is 24 hours each
year, while the most recent information available from ACA standards
suggest that correctional line staff receive 40 hours of training each
year. Additionally, according to the National Institute of Corrections,
26 states provide their correctional officers with 40 hours of annual
training. Four states provide more than one week of training.(') One
DOC o f f i c i a l stated that the Department recognizes the need for
increasing i t s training requirements and would eventually like to
increase the requirement to 40 hours.
Lack Of Commitment And Resources
Mav Have Hindered Traininu Efforts
Regardless of the number of training hours that are mandated, i f
institutional s t a f f are not f u l l y committed to ensuring that a l l staff
receive training, then i t is unlikely most staff w i l l receive adequate
( 1) Information obtained as of November 1988.
30
training. A lack of commitment to staff training may be one reason many
correctional s t a f f a t DOC have not met training requirements. In
addition, because of staff vacancies i t may be d i f f i c u l t for CSOs to find
the time for training. A lack of resources to implement an effective
tracking system may also hamper training efforts.
Lack of comnitment to staff trainina mav be one reason that many staff do
not meet reauirements - A lack of commitment on the part of some
institutional staff may be a reason that many CSOs have not obtained the
necessary training. According to a DOC o f f i c i a l , in order for CSOs to
meet the requirement of 24 hours of training each year, the warden and
staff at each i n s t i t u t i o n must be committed to ensuring that a l l security
staff receive the required hours of training. Without this commitment,
i t is not l i k e l y that the required number of training hours w i l l be met.
For example, a training officer at one i n s t i t u t i o n recently noted that
since the arrival of a new warden, attendance at training classes has
greatly improved. He attributed the increase to the new warden's
commitment to training. In fact, he noted evidence of how the
importance of training has been stressed recently -- some officers were
now coming in for weapons training who had not been c e r t i f i e d for three
or four years.
Staff shortages mav prevent some s t a f f from attendina training - Even
when there is a commitment to training, i f there are a high number of
vacancies in an institution, attendance at staff training may be low.
One training officer noted that correctional supervisors t e l l him that
they cannot afford to send CSOs to training because of staff shortages.
Another training officer noted that the only reasons cited for lack of
attendance at training are staff shortages in the units or due to illness.
Lack of resources may have also hampered trainina efforts - A lack of
funding to improve DOC'S system for monitoring training may also
contribute to the problem of inadequate training. Since the 1985 audit,
DOC has implemented an automated system to track training hours for a l l
of i t s s t a f f . DOC'S i n i t i a l attempt to automate used a personal computer
that quickly ran out of space and capacity to handle training information
for approximately 6,000 employees. As a result, the training records
management function was added to the mainframe system DOC was developing
at the time, APPLES, for payroll and personnel information. APPLES was
established in January 1988. However, according to a DOC o f f i c i a l , the
Department receives no funding for the personnel needed to run the
system, although it does receive funding for processing time. DOC
o f f i c i a l s indicate that two programmers and a training person are needed
to upgrade and maintain the system. Without sufficient funding for the
APPLES system, the system's a b i l i t y to generate adequate management
reports on staff training is limited.
DOC should ensure that each of i t s Correctional Service staff receive
the required 24 hours of training by stressing the importance of
advanced training on an ongoing basis even when there may be staff
shortages.
DOC should review i t s refresher courses and determine which subjects
are c r i t i c a l to a l l institutions. Those identified as c r i t i c a l
should then be taught at a l l institutions.
DOC should review i t s training monitoring system to determine what
changes can be made to make the system more effective in tracking
staff training.
OTHER PERTINENT INFORMATION
During our audit, we compiled data on the growing impact of compensatory
time accumulated by security s t a f f .
Growina impact of camensatory time - The impact of the use of
compensatory time on security staffing is increasing. During our
calculation of the security staffing formula ( Finding I V , page 23) we
found that, as compared to the 1986 formula, CSOs are taking nearly twice
as much compensatory time o f f now. In fiscal year 1989- 90, CSO Is and
CSO I l s averaged 4.75 days o f f for compensatory time compared to the 1985
average of 2.75 days. In addition to taking time o f f , we found that many
CSOs are receiving pay for compensatory time. In our sample review of
341 CSOs for the staffing formula, 200 ( approximately 60 percent) were
paid for compensatory time during fiscal year 1989- 90. According to a
DOC o f f i c i a l , in fiscal year 1989- 90 the Department paid $ 834,786.70 for
compensatory hours to s t a f f a t the adult institutions, the majority of
whom are Correctional Services s t a f f . Additionally, DOC paid over $ 2
million during the f i r s t months of fiscal year 1990- 91. This was a
one- time payoff of a l l accumulated compensatory time and holiday leave
time. DOC budget staff stated that this was the f i r s t time they had ever
paid o f f the balances.
These large payments for compensatory time very l i k e l y w i l l continue to
grow. A court decision in June 1989 found that certain employees who
work overtime have the option of receiving overtime payment or earning
compensatory time. The employer and employee must reach an agreement or
understanding prior to the performance of any overtime work. This
decision applies to certain employees covered by the Fair Labor Standards
Act ( FLSA). The Department of Administration was notified by the
Attorney General's Office that this ruling would apply to Arizona. This
decision w i l l have a significant effect on agencies, especially those
that operate on a 24- hour basis. Agencies such as DOC w i l l have to give
a employees the option of taking compensatory leave or receiving pay for
' compensatory time. In fact, agencies w i l l not be able to allow
employees, such as CSO Is and CSO I l s , who are subject to the FLSA to
work unless the agency is able to pay time and a half or unless these
employees have elected to receive compensatory leave. This policy w i l l
be implemented by DOC in May 1991. Over half of a l l e l i g i b l e s t a f f at
the institutions have indicated that they would accept either
compensatory leave or cash payment. An additional 20 percent elected
cash payment only. DOC has requested $ 2.8 million to cover the cost of
the compensatory time payments.
Security staffing w i l l be impacted by this policy in several ways.
F i r s t , CSOs are needed to work overtime during emergency situations, such
as escapes, disturbances, and staff shortages. Should the agency be
without funds to cover these overtime expenses, i t is unlikely that some
CSO staff would be available to work during emergencies or, i f they must
be used, DOC w i l l have to make budgetary adjustments afterwards to cover
the cost. DOC previously had the option to provide time o f f in lieu of
overtime pay. Second, this policy could affect calculation of the
staffing formula. Since employees may elect to receive a cash payment
for overtime rather than time o f f , the amount paid for compensatory time
may be increased and the amount of compensatory time taken as time o f f
may decrease. In such a case, the staffing formula could be affected,
potentially reducing the number of staff needed.
FIFE SYMINGTON
GOVERNOR
1601 WEST JEFFERSON
PHOENIX, ARIZONA 85007
( 602) 542- 5536
SAMUEL A. LEWIS
DIRECTOR
May 17, 1991
Douglas R. Norton, Auditor General
Office of the Auditor General
2700 North Central Avenue, Suite 700
Phoenix, " irizona 85004
Dear Mr. Norton:
Thank you for the opportunity to review your revised draft as a result of our
meeting on hlay 9, 1991. The Department of Corrections acknowledges and agrees
to the changes of your preliminary draft.
Please include the Department of Corrections' comments where appropriate as
submitted on May 7, 1991, in response to your previous draft dated April 30, 1991.
Thank you for your cooperation. If additional information is required, please
advise.
-
Director !
FIFE SYMiNGTON
GOVERNOR
1601 WEST JEFFERSON
PHOENIX. ARIZONA 85007
( 602) 542- 5536
SAMUEL A. LEWIS
DIRECTOR
Donald X. Norton, Auditor General
Office of the Auditor General
2700 North Central Avenue, Suite 700
Phoenix, Arizona 85004
Dear Mr. Norton:
Attached i s t h e preliminary response to the Performance Audit- Draft on Institutional
Security and Staffing.
With minor exceptions, the Department of Corrections agrees and accepts the findings of
this performance audit. Your office should be commended for producing a report t h a t
accurately describes performance deficiencies and recommendations on methods for
improvement. The security consultants, Mr. James D. Henderson and David E. Musacchio
were an excellent choice. Their knowledge and expertise was very evident. They o f f e r e d
valuable suggestions to our s t a f f f o r improvement of security operations and always
demonstrated their professionalism and leadership qualities.
Please be assured that this Department will make every effort t o achieve our goals t o
improve security and maximize the use of staff as addressed in the attached. response to
your Performance Audit.
This audit will serve as a valuable resource and should prove significant in supporting our
goals t o operate efficient and safe prisons.
Attachment
RESPONSE TO AUDITOR GENERAL PERFORMANCE AUDIT
ON SECURITY AND STAFFING OF THE
ARIZONA DEPARTMENT OF CORRECTIONS
The Department of Corrections is appreciative of the Auditor General's recognition of
the significant improvements made by this Department since the two previous Audit
reports issued by your o f f i c e i n 1985 and 1986.
The Department of Corrections agrees with most of the findings as submitted and offers
the following comments and comrnittment to improve the noted deficiencies.
TOOL CONTROL: pg. 6,9
ADC policy requires that prisons maintain a complete and current inventory of all tools
and materials t h a t may a f f e c t t h e safe, secure and orderly operation of each facility.
Wardens are again being directed t o improve t h e i r procedures t o properly monitor this
most important issue. Annual review of all tool control operation will be conducted to
ensure compliance with policy. Recent efforts have improved the tool control practice
at most prisons. However, some institutions s t i l l are not in compliance.
INMATE COUNTS: pg. 6,9
The Department's current policy requires a minimum of three formal counts ( one each
shift) at every institution. It's believed t h a t increasing the number of counts t o f i v e i s
not necessary, costly, and would not significantly improve inmate accountability.
Our belief is t h a t the Auditors may have been misled by the term " informal count".
Informal counts are random periodic checks, established by ADC policy, t o determine if
- all inmates a r e present and to monitor their well being. The number of informal counts
is dependent on custody level but in all instances, occurs at least once between each
formal count, Instances where inmates have not been a c c u r a t e l y accounted for in a
timely manner are almost non- existent except where human error o r f a i l u r e t o follow
established procedure occurs.
DETENTION AND ISOLATION CELLS: pg. 7
Double bunking Detention C e l l s i s not good practice and limits the Department's ability
t o effectively manage and discipline inmates. Ideally, every unit would have a
percentage of cells for Isolation and disciplinary problems. However, the costs are
prohibitive. A inore realistic goal would be t o reduce the inmate population via
alternatives t o incarceration and o t h e r e f f e c t i v e programs t o make available more room
in our current prisons so t h a t they can once again become single cell operations.
Response t o Auditor General Performance
Audit on Security and Staffing of the
Arizona Departrnent of Corrections
Page 2
TEMPORARY STRUCTURES: pg. 6
The Department agrees with the Audit findings that quonsets, trailers, modulars and
some dormitory inmate housing are substandard. However, replacement c o s t s f o r t h e s e
units are conservatively estimated at $ 25,000 thousand per bed and with over 2,000
substandard beds or approximately $ 50,000,000 - the replacement costs are prohibitive.
Realistically speaking, the state has not even been successful in keeping pace with i t s
current prison population growth, l e t alone replacing existing inadequate facilities.
Replacement of t h e s e s t r u c t u r e s will become a priority as other priorities deminish or
where there is a turn of unfavorable events that heightened the awareness of operating
under such precarious conditions.
FACILITY DESIGN: pg. 6, pg. 8
The rapid growth and limited resources to m e e t t h e increased demands for additional
prison space of the last 20 years have not kept pace. Temporary emergency beds with
life expectancy of a few years somehow always seem to become permanent. Conditions
which caused development of inadequate f a c i l i t i e s are often forgotten. Fortunately the
Legislature has begun t o recognize the importance of good prison design and appropriated
funds accordingly, examples of which a r e t h e Rynning Unit, SMU, ASPC- Winslow, ASP-Yuma
and ASPC- Tucson. Once construction and inmate population growth level off, and
the legislature appropriates funds, remodeling o r replacement of inadequate facilities
can become a reality.
Additionally, while the Department agrees with the comments regarding the South Unit's
design, i t should be noted t h a t exceptionally good inmate management skills exist by
staff at the South Unit to accommodate the successful placement of high medium
custody inmates placed there, and because of this success i t is not being recommended
that a change in custody level for South Unit be made at this time.
KEY CONTROL: pg. 9
A Departmental policy is being promulgated to establish and monitor key control. The
Department acknowledges t h a t current procedures are inadequate. ADC policy will
require t h a t procedures be established 90 days after the e f f e c t i v e d a t e of the Key
Control policy. Full compliance is expected by January 1, 1992.
0 Response to Auditor General Performance
Audit on Security and Staffing of The
Arizona Department of Corrections
Page 3
PERSONAL BODY ALARMS: pg. 8,10
Installation, monitoring equipment and costs f o r the personal body alarms will be
examined; as will the effectiveness of the alarms. Based on the results of this
investigation a decision will be made as whether to request funding for this additional
safety equipment.
SECURITY INSPECTIONS: pg. 8
A policy requiring each institution to establish written procedures that require regularly
scheduled and consistently documented security inspection for all aspects of prison
security operations is being developed. Current ADC Policy 304.3, " Inspections of
Department Facilities" shall be revised t o adhere to Auditor's recommendations.
Institution will have 90 days from the effective policy date t o establish their
procedures. Full compliance i s expected by January 1, 1992.
POST ORDERS: pg. 9
Hostages and use of force situations are issues that have already been established by
Departmental policy and therefore should not be p a r t of specific post orders.
However, the Department does agree t h a t each job assignment or post should have post
orders and signature s h e e t s f o r employees t o record t h a t they have read and understand
their job responsibilities. The wardens will be directed t o meet full compliance through a
directive from this office. A review for compliance will be part of the Security
Inspections Policy.
INCREASE STARTING SALARY FOR CSO: pg. 20
The Department endorses the recommendation to increase the starting salary for
Correctional Service Officers. The higher the starting salary the more a t t r a c t i v e is the
job, therefore, drawing a better qualified career oriented individual t h a t helps to reduce
employee turnover and improved job performance.
Response to Auditor General Performance
Audit on Security and Staffing Of
Arizona Department of Corrections
Page 4
POST POSrrION REVIEW: pg. 24, 25
The Department welcomes a comprehensive study of our staffing requirements.
...
However, given the current economic condition in Arizona i t may be a waste of the
proposed $ 60,000 for the study unless the Legislature is prepared to allocate funds for
additional positions as determined by the study. Previous studies completed internally all
indicate a shortage of personnel. We believe any responsible analysis of staffing
requirement would meet with the same conclusion.
WHILE ADC HAS MADE STRIDES TO INCREASE IN- SERVICE TRAINING, MANY CSO5
STILL DO NOT MEET REQUIREMENTS: pg. 27
" The reasons In- Service Training has been inadequate may be lack of commitment by
institutions to ensure training is received, and staff shortages."
The lack of commitment may be accurate for one to two institutions according to data
reported, but overall, the commitment to all aspects of training, in my opinion, is very
high.
The following factors, have greater irnpact on lack of In- Service Training than
institutions not committed to training:
1. Staff shortages at most institutions have had the greatest impact on ensuring
staff receive the required and mandatory training. This also includes in many
instances the lack of a Training Officer to direct and coordinate the required
training program.
Historically, filling vacant Training Officer positions has taken three to six
months. Additionally, at two institutions, the legislature has not committed to
funding Training Officer positions ( i. e., ASP- Y uma and ASP- Saf f ord). As new
institutions are constructed and staffed, i t is imperative that Training Officer
positions be recognized as an important area to be funded. Also, it is essential
that support staff ( i. e., clerical) and operating resources be properly funded so
that in- service can achieve the goals and mission of ADC.
Response to Auditor General Performance
Audit on Security and Staffing Of
Arizona Department of Corrections
Page 5
2. Reporting of Training information may be a factor t h a t contributed to the lack
of required training. Entering data from institutions into the APPLES computer
system is slow and in the past provided inaccurate information. To date,
institutions and bureaus have not been provided the necessary information to
indicate training deficiencies or problem areas to be corrected.
Each fiscal year ( starting with FY ' 91) Staff Development and Training will
audit employee training records by conducting an employee review of their
training records with the expectation of correcting any data e n t r y e r r o r s or lack
of training information. Then monthly, each Training Officer and institutional
Administrator will be provided a status report on employee training. Quarterly,
Staff Development will analyze employee training data and submit a report to
ADC Assistant Directors, as t o the status of employee training categories such
as: Employee Orientation, Refresher Training, Weapons Qualification,
Professional Development, etc. Improving data records and increasing the
distribution of training data will hopefully eliminate the problem of staff who
lack or do not meet training requirements.
MANY STAFF HAVE NOT RECEIVED SUFFICIENT TRAINING: pg. 29
The average of 33 hours In- Service Training i s an acceptable figure.
The 44 percent of CSOts who did not receive the required 24 hour In- Service Training is
not acceptable.
Vany higher ranking security s t a f f failed t o receive 24 hours required training. The data
indicates approximately 45% of security supervisors and managers did not achieve the 24
hours required training. The new Professional Development Program should eliminate
this problem. It is projected t h a t 220 supervisors will receive PDP Level I training and
160 managers will receive PDP Level I1 training during FY ' 92, assuming t h a t the funding
to conduct t h i s Professional Development Program is available.
Arizona Department of Corrections may not be able t o identify the specific subjects in
which individual CSO's have received training.
1. This is not an a c c u r a t e s t a t e m e n t as institutional refresher training i s entered
into APPLES as a single entry but each specific course t h a t comprised the
refresher training is detailed in a separate section of APPLES documentation.
It is true t h a t ADC needs t o standardize as much as possible the CSO Refresher
Training and other types of refresher training. A Training Officer has been
assigned the task of coordinating a l l refresher training and proposing a core
curriculum for all institutions to comply with when conducting the required
refresher training.
Response t o Auditor General Performance
Audit on Security and Staffing Of
Arizona Department of Corrections
Page 6
ADC'S 24 ANNUAL TRAINING REQUIREMENT MAY NOT BE ENOUGH: pg. 30
Arizona Department of Corrections recognizes that 24 hours of In- Service Training may
not be sufficient but additional training will have a fiscal impact that will be difficult t o
increase. The Arizona Legislature and J. L. B. C. will have to understand the importance
of adequate training for staff and commit the necessary appropriations to expand the
present In- Service Training.
Presently, most institutions are conducting the 40 hours of refresher training for CSO1s.
This includes the 24 hours training required and approximately 8 hours of weapons
qualifications. The other eight hours is determined by the institutional Training Advisory
Committee and institutional Warden. Four of the nine institutions offer the 24 hours In-
Service Training and one day for weapons qualification. It would be more accurate to say
t h a t ADC has a 32 hour In- Service refresher training program for Correctional Service
series employees.
LACK OF COMMITMENT MAY BE ONE OF THE REASONS THAT MANY STAFF DO
NOT MEET REQUIREMENTS: pg. 31
This reference is t o ASPC- Florence where the refresher training was previously only 20
hours annual and s t a f f were not mandated to attend, but were encouraged. Since Warden
Roger Crist arrived at ASPC- Florence, he has committed t o training and is in cornpliance
by mandating 24 hours of In- Service Training and eight additional hours for weapons
qualification.
RECOMMENDATIONS: pg. 32
Concur with all three recommendations.
Numbers I and 2 a r e in progress while recommendation number 3 will require time t o
research the problems with APPLES and the corrective measures needed f o r a quality
training records system. Initial review indicates some fiscal funding will be necessary t o
revise and improve APPLES t o m e e t t h e needs of Staff Development and Adult
Institutions.
Response to Auditor General Performance
Audit on Security and Staffing Of
Arizona Department of Corrections
Page 7
90% OF ALL ELIGIBLE STAFF WILL CHOOSE THE PAY OPTIONS: pg. 34
Initial data does not support t h i s high percentage. I would suggest a lesser figure such as
50% of staff will choose the pay option.
This compensatory issue has a great impact on the COTA operation because i t reduces
the flexibility in designing and implementing the 288 hour curriculum. Additionally, the
compensatory time issue will have a negative impact on staff who are able to attend In-
Service and professional training programs.
RECOMMENDATIONS:
I would recommend t h a t ADC pursue a longer pay period similar t o law enforcement
agencies instead of using the standard 40 hour work week. The 40 hour work week is very
restrictive and causes compensatory time and overtirne pay which could be greatly
reduced by increasing the pay period to 80 or 169- hours.
Concur with recommendations on Page 20 t o increase the starting salary t o Correctional
Service Officers to be competitive.
Recommend additional funding t o provide transportation f o r staff to commute to
institutions in r e m o t e a r e a s such as ASPC- Florence and ASP- Fort Grant. ADC would
define what i s considered a remote location and in conjunction with state transportation
( van pooling) staff would be supplemented for their daily commute.
Recommend ADC establish a centralized s t a f f position control system on computer so all
data elements can b e e n t e r e d to a c c u r a t e l y r e f l e c t the f a c t o r s f o r proper staffing.
The Correctional Officer Training Academy ( COTA) should be recognized as a 288 hour
pre- service training program for Correctional Service Officers with an additional 40
hours of On- The- Job Training ( OJT) for graduated cadets at their assigned institution.
Marginal staffing has been created by the legislative appropriation's process and
Representative Evans1 sub- committee. This has resulted in:
1. Increased compensatory time especially at the line staff level.
2. Lack of commitment t o training ( and other lower priority activities) because of
the importance t o operate the institutions in a safe and secure manner is the
highest priority.
3. Increased sick leave.
Response to Auditor General Performance
Audit on Security and Staffing Of
Arizona Department of Corrections
Page 8
Therefore, i t is imperative t h a t t h e legislature recognize the importance of properly
staffing institutions in order to meet t h e mission and goals of ADC. Continued marginal
staffing will only result in continued areas such as training that will not be able to
achieve the required objectives and possibly creating a liability situation.
Historical turnover rates for CSO 1' s indicates a higher percentage than the 18% stated
. on Page 23. Presently, the r a t e is 18% but over the past seven years, the figure would
indicate a 20% turnover rate for calculating staffing needs. Additionally, the 18% figure
could quickly increase due to economic recovery and other changes in the Arizona
economy. If the Arizona Legislature does not improve the salary for Correctional
Service Officers, the turnover rate will certainly increase, and will probably increase
beyond the 20% average.
CSO I TURNOVER RATE
1984 - 32% 1988 - 17%
1985 - 30% 1989 - 20%
1986 - 19% 1990 - 20%
1987 - 20% To Date - 18%
GROWING IMPACT OF COMPENSATORY TIME: pg. 33,34
The Department of Corrections requested $ 2.3 million dollars for payment of overtime,
unfortunately i t was not funded. Therefore, the Department's. position will be to only
work those staff who choose to work for compensatory time.
STAFFING FORMULA: pg. 34
The Department agrees that several factors e f f e c t t h e staffing formula and should be
addressed when opening new institutions and annually reviewed to ensure existing
facilities are properly staffed.
The Office of the Auditor General contracted with James D. Henderson and
Associates to assist with a follow- up review of security at the
Department of Corrections' adult i n s t i t u t i o n s . Listed below are the
names and backgrounds of the consultants.
James D. Henderson, a retired warden and regional director for the
Federal Bureau of Prisons, who i s now a criminal justice consultant.
Mr. Henderson has served as a consultant on projects involving prison
operations, management, security, and s t a f f i n g in 16 other states.
He has participated in numerous special projects and training
assignments, and co- authored American Correctional Association
publications about r i o t s and disturbances, and the development of
secur i ty programs.
David E. Uusacchio, a former director of corrections for local
government correctional f a c i l i t i e s in Kentucky and Louisiana has
worked in criminal justice consulting since 1980. Mr. Musacchio has
served as a consultant on more than 150 projects involving criminal
justice planning, and prison design and construction.
A copy of the Executive Summary from their report i s attached.
Technical Assistance Visit
Arizona Department Of Corrections
James D. Henderson
David E. Musacchio
Criminal Justice Consultants
Executive Summary
This report summarizes the results of a technical assistance visit conducted during : he
period Sovember 1990 through January 1991 for the Arizona Auditor General's Office 51
criminal justice consultants James D. Henderson and David E. Musacchio. The projec!
involved on- site visits to sixcorrectional complexes operated by the Arizona Department or
Corrections: Perryville, Phoenix, Florence, Tucson. Douglas, and Winslow.
In general, this project resulted in a findrng that substantial progress has been made toward
upgrading institutional security in the five years since an earlier consultant review of the
Department's operations by N. R. Cox kssociates, Inc. From the results of this survey. I r
appears that master planning and other management improvements that have been the
focus of Department activity in recent years are beginning to show appreciable returns.
These efforts should be continued and expanded.
The Consultants generally found that the physical perimeter security of the institut~ ons
surveyed is adequate. While there remain individual areas that will require additional
attention, and continual vigilance is necessary, overall, perimeter security was not a major
issue, as evidenced by the reduced number of escapes annually.
The Department's inmate classification system generally appears to be functional, although
increased cousistency in procedures would be desirable. While there were signs of minor
imbalances at'rome locations, considering the level of crowding the system is experiencing.
the Department appears to be doing a satisfactory job in balancing the population at the
various units. More creative program use of existing facilities couid effect a reduction In
the use of inmate transportation and ultimately impact the need for Cellblock Six and
Special Management Unit assignments. This would, however, require more locked unlt
space in the various units.
The system as a whole does a very good job in employing inmates. In particular, the use of
properly classified inmates on community service jobs is commendable, as are the efforts to
occupy higher security inmates through Arizona Correctional Industries. As cro~ cd~ ng
continues, so should the efforts to expand job opportunities. a
High security operations in the Special Management Unit and Cellblock Six, two ot the
most critical units in the system, appeared to be quite good.
Inmate accountability procedures in most locations appeared to be satisfactory. This 1s a
critical factor in overall security, since inmates who are not properly controlled throughout
the day will be able to exploit even minor weaknesses in an othexwise sound facility.
a
Procedural issues were somewhat more of a concern. Staff have undertaken to upgrade
critical areas such as key and tool control, inmate accountability, security inspections, and
post orders. However, system- wide, these are common denominator problem areas that
will require additional direction by the Central Office and effort by institutional staff. ( I
Virtually every institution displayed these problems in one or more units. A system- wtde
security evaluation and monitoring process should emanate from the Central Office. This
appears to be a new concept in the Department, but it could also improve communications,
and the sharing of functional procedures and other improvements, throughout the system. ( I
Major findings on those points are enumerated in the body of the report.
In addition to these broad arcas, there are several specific points of concern that were
observed throughout the system.
Some units had little or no locked unit capaclry: others had locked units that were double-bunked.
Based on guidelines accepted by r he .& nerican Correctional Association and
generally accepted correctional practice, the re does not appear to be a sufficient number of •
either isolation or segregation beds available in the system, Increases in this type of
security housing in the individual complexes would improve overall system effectiveness.
Double- bunking is not desirable in any med~ um or high security prison, and certainly not in
special management or other locked- type umu. Understanding that virtually every •
correctional agency in the Nation is double bunking to a sipficant degree, the Anzona
system should make a conarted effort to eliminate it in all locked units.
Although hand- held radios were well- distributed throughout most institutions, there were
no personal body alarms. This shouid be a high priority, starting with locked units and the
more remote assignments. Body alarms should definitely be considered as a part of any
radio equipment replacement program. as it is the most cost effective method of providing
personal protection for staff.
Several of the housing units displayed less than ideal design features. This can only be
improved with major renovations. Some detention units had doors that opened inuard,
making barricading easy. Other housing areas were so arranged as to make impossible
proper supervision by any reasonable number of staff. In a similar vein, the use of trailer
and other modular units, although necessitated by population pressures, is generally
undesirable and inevitably will lead to high maintenance and repair costs.
Staffing was not a focus of this project, but there were clear indications of the need for a
full study of the staffing requirements of the system. The Consultants strongly recommend
that another consultant or consultant firm be engaged for this purpose. Such a project
would entail, in each Department facility, a detailed review of perso~ el asslgnrnenr
records, extensive staff interviews, visits to posts, analyze the physical plant and ongoing
program activities, as well as the use and efficacy of any unit management system. Such a
study would be a major undertaking; it would involve not less than a week for each major
institution like Central Unit. The use of arbitrary relief factors should be abandoned, slnce
they do not reflect the true staffing situation in any institution or the system as a whole.
Overall, the Arizona Department of Corrections has made sipficant progress toward
rectrfylng the conditions identified in the Cox report, and is continuing in its efforts to
upgrade procedures and policies that will further strengthen internal inmate supervision
and perimeter security. In addition to the specific questions and areas that were part of the
project, some additional areas were examined. In addition to the above comments relatlve
to tool control, key control, and inmate accountability, the following areas were observed:
security inspections, policy implementation, post orders, and audit oversight.
Security inspections generally were not carried out in an organized manner. When and
where they were conducted, there is no standardized method of recording and reporting
that would allow for routine review by security supervisors.
Policy and audit oversight as described above. there are department level policies and
directives on almost every subject. However, there currently is no process in place that
reviews staff understanding of those documents or that monitors compliance. Such a
process is recommended. ( I
Post orders are inconsistent throughout the system. These should be standardized. and
each should include a signoff sheet for staff manning the post, to sign. indicating the!
reviewed and understand the orders.
The Arizona Department of Corrections should be commended for their efforts to improbe
operations and security. The problems found in this survey are more in line with a need for
streamlining, rather than a total re- design of the system. With the implementation of these •
recommendations, significant improvements can be expected.