Performance Audit, Department of Public Safety, Telecommunications Bureau

              PERFORMANCE AUDIT
DEPARTMENT OF PUBLIC SAFETY
TELECOMMUNICATIONS BUREAU
Report to the Arizona Legislature
By the Auditor General
November 1990
90- 5
DOUGLAS R. NORTON, CPA
AUDITOR GENERAL
STATE OF ARIZONA
OFFICE O F THE
AUDITOR GENERAL
November 26, 1990
Members of the Arizona Legislature
The Honorable Rose Mofford, Governor
Colonel F. J. " Rick" Ayars, Director
Department of Public Safety
Transmitted herewith is a report of the Auditor General, A Performance
Audit of the Department of Public Safety, Telecomunications Bureau.
This report is in response to a June 14, 1989, resolution of the Joint
Legislative Oversight Committee and was conducted as part of the Sunset
Review set forth in Arizona Revised Statutes § § 41- 2351 through 41- 2379.
This report i s the f i r s t in a series of five reports to be issued on the
Department of Public Safety. The report addresses the effectiveness and
efficiency of the Telecommunications Bureau. We found the Bureau is
generally well managed. However, we found that the Department could
generate more than $ 700,000 annually by establishing a fee for conducting
background checks and by modifying the fee for copies of accident
reports. Further, the report recommends that the Department improve i t s
operational audits of agencies which access criminal history data. In
addition, the Department should take steps to upgrade the completeness
and reliabi l i t y of i t s criminal history data. We also report the need
for the Department to strengthen i t s policies and procedures over the
acquisition and assignment of telecommunications equipment, and to
improve i t s cash handling procedures in i t s Arizona Criminal Justice
Information System Division. Finally, we found that most users are
generally satisfied with the Emergency Medical Services Communication
system operated by the Bureau, although a few problems were noted.
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 the report.
Sincerely,
~ o u w sR. Norton
Auditor General
DRN : l mn
Enc l osu re
STAFF: William Thomson Kurt L. Schulte
Deborah A. Klein Sarah J. Auvil
Kimber ly S. Hi ldebrand Mary M. Edmonds
2700 NORTH CENTRAL AVENUE ' SUITE 700 ' PHOENIX, ARIZONA 85004 ( 602) 255- 4385 ' FAX ( 602) 255- 1251
SUMMARY
The O f f i c e o f the Auditor General has conducted a performance audit of
the Arizona Department of Public Safety ( DPS), Telecommunications Bureau,
pursuant to a June 14, 1989, resolution of the Joint Legislative
Oversight Committee. This performance audit was conducted as part of the
Sunset Review set forth in Arizona Revised Statutes ( A. R. S.) $ 541- 2351
through 41- 2379.
This i s the f i r s t in a series of five reports to be issued on the Arizona
Department of Public Safety ( DPS). The report focuses on the functions
of the Telecommunications Bureau, which i s responsible for developing,
operating, and maintaining the State's criminal history data system as
well as the department's other data processing and communications
systems. The bureau has 284 full- time employees and a budget of
$ 14,234,600 for fiscal year ( FY) 1991.
DPS Could Generate More Than
$ 700,000 Annuallv bv Establishing
a Fee for Certain Services and Updatinq
Current Fee Schedules ( see pages 7 through 11)
By establishing a fee for conducting background checks and modifying the
fee for copies of accident reports, DPS could generate more than $ 700,000
annually. DPS s t a f f conduct approximately 6,000 background checks per
month for both criminal and other agencies. However, Arizona i s one of
only seven states that has not established a state processing fee for
completing background checks. I f DPS charged $ 10 for processing state
background checks, the department could raise over $ 600,000 for the
state. ( A. R. S. $ 41- 1750 would have to be amended to al low for the
establishment o f t h i s fee.) In addition, during the course of our audit,
DPS began charging for a l l requests for copies of accident reports-- this
change shou Id generate an add i t ional $ 108,000 annual ly .
DPSs' O~ erational Audit
Proaram Is Deficient ( see pages 13 through 20)
DPS i s not conducting required audits of agencies accessing criminal
history data. A National Crime lnformation Center ( NCIC) policy requires
biennial audits of computer terminal users of the Arizona Criminal
Justice lnformation System ( ACJIS) to ensure these agencies comply with
state and NCIC policy and regulations. ( ACJIS provides Arizona users an
interface with the Federal Bureau of Investigation's NCIC database, which
contains criminal history record information on a national level.)
However, since the audit program was implemented in 1987, only 39 percent
of the terminal agencies have been audited by DPS on a biennial basis.
As a result, deficiencies such as improper dissemination of information,
inaccurate or incomplete data on the system, or system security problems
may go unnoticed.
Further, most of the audits that have been completed by DPS have been
very limited with the entire audit consisting of only a questionnaire.
Few audits have included a v i s i t to the agency s i t e where the terminal i s
housed. In comparison, none of the five states we contacted use a
questionnaire as the entire audit, and a l l of them include a s i t e v i s i t
as part of each audit. During 1990, DPS began v i s i t i n g most agencies
being audited and plans other improvements to the audit program as well.
DPS Needs to lm~ rovet he Completeness and
Reliabilitv of its Criminal Histow Information
( see pages 21 through 33)
DPS needs to take steps to upgrade the completeness and r e l i a b i l i t y of
i t s criminal history data. DPS i s s t a t u t o r i l y required to maintain
criminal history information on individuals arrested by Arizona law
enforcement agencies. This information i s widely used by both criminal
justice and other agencies in completing criminal investigations and
pre- employment and pre- licensing background checks. However, the
i n t e g r i t y of the data i s compromised for several reasons. F i r s t , arrest
data i s not always entered on the computer system in a timely manner due
to slow submission of arrest cards by local agencies. We found that 24
percent of the agencies we sampled submitted arrest cards over 30 days
after the arrest. Also, once DPS receives the arrest card the
information may not be entered into the system for over three weeks.
Second, over 60 percent of the arrest records currently on the system are
missing at least one disposition i e . , resolution of the arrest).
Consequently, the outcome of these arrests ( convicted, acquitted, etc.)
is not captured on the system. Some of these arrests are for serious
offenses including homicide and child molestation. A backlog of over
100,000 disposition forms awaiting entry into the system explains some of
these open arrest records. However, insufficient procedures for
resolving disposition forms rejected by DPS staff may be another cause of
the numerous open arrest records.
Finally, DPS does not routinely verify the arrest and disposition data
entered into the system to ensure data entry accuracy. In contrast, four
of the five model states we contacted verify 100 percent of the data
entry performed by state criminal history records s t a f f .
Other Findina Areas:
Other findings in this report address:
The need for the department to strengthen i t s policies and procedures
over the acquisition and assignment of telecommunications equipment.
Currently, other bureaus may purchase such equipment regardless of
whether the Telecommunications Bureau feels the equipment is
just i f ied. ( See pages 35 through 39)
The need for the Arizona Criminal Justice Information System Division
to improve i t s cash- handling procedures. Cash and checks are not
adequately secured. ( See pages 41 through 42)
Users are generally satisfied with the Emergency Medical Services
Communications system. This system, which provides a radio link
between rescue vehicles and hospital staffs, is a statewide system
operated by the bureau. Although most users feel that the system
meets thei r needs, some problems were noted. ( See pages 43
through 48)
TABLE OF CONTENTS ( Con t ' d )
FINDING IV: DPS NEEDS TO STRENGTHEN
POLICIES AND PROCEDURES OVER
THE ACQUISITION AND ASSIGWENT
OF TELECOUUN lC AT lO NS EQU IP MENT . . . . . . . . . . . . .
Existing Policies and Procedures
Are Insufficient . . . . . . . . . . . . . . . . . . . .
Department Strengthens
Acquisition Policy . . . . . . . . . . . . . . . . . . .
Recommendations. . . . . . . . . . . . . . . . . . . . .
FINDING V: THE ACJlS DIVISION SHWU.
IUPROVE ITS CASH- HANDLING PROCEDURES . . . . . . . . . .
Recommendations. . . . . . . . . . . . . . . . . . . . .
FINDING VI: MOST USERS FEEL THAT THE
EMERGENCY UED l CAb SERV I CES COUllJN l CAT l ONS ( EUSCOU)
SYSTEM ADEQUATELY WETS THEIR NEEDS;
HOWEVER, SOUE PROBLEYS WERE NOTED. . . . . . . . . . . .
Most Users of EMSCOM Are
Generally Satisfied. . . . . . . . . . . . . . . . . . .
Some Problems Noted Cannot
Be Addressed by DPS. . . . . . . . . . . . . . . . . . .
Transmission Quality Problems Stemming
from Insufficient Dispatcher Training
Can Be Addressed . . . . . . . . . . . . . . . . . . . .
Lack of Centralized Control for
Monitoring User Complaints . . . . . . . . . . . . . . .
Recommendations. . . . . . . . . . . . . . . . . . . . .
TABLE OF CONTENTS ( Con t ' d)
Pane
OTHER PERTINENT INFOfUATION. . . . . . . . . . . . . . . . . 49
Automated Fingerprint
Identification System. . . . . . . . . . . . . . . . . . 49
Mobile Digital Terminals . . . . . . . . . . . . . . . . 52
Computer Assisted Dispatch System. . . . . . . . . . . . 53
Phoenix Metropolitan Area 800 MHz
Radio Trunking System. . . . . . . . . . . . . . . . . . 54
ACJlSComputer Upgrade . . . . . . . . . . . . . . . . . 55
AGENCY RESPONSE
LIST OF TABLES
TABLE 1 Department of Public Safety
Telecommunications Bureau
Statement of FTEs and Actual and Budgeted
Expenditures for Fiscal Years 1988- 89,
1989- 90, and 1990- 9 1 ( Unaud i t ed ) . . . . . . . 3
TABLE 2 Other States' Applicant Processing Fees. . . . 9
TABLE 3 Arrest Counts Without Dispositions . . . . . . 28
INTRODUCTION AND BACKGROUND
The O f f i c e o f the Auditor General has conducted a performance audit of
the Arizona Department of Public Safety ( DPS), Telecommunications Bureau,
pursuant to a June 14, 1989, resolution of the Joint Legislative
Oversight Committee. This performance audit was conducted as part of the
Sunset Review set forth in Arizona Revised Statutes ( A. R. S.) 5541- 2351
through 41- 2379. This i s the f i r s t in a series of f i v e reports on the
department.
Backuround
The Department of Public Safety was established on July 1, 1969,
conso I i dat i ng the funct ions and respons i b i I i t i es of the Ar i zona Highway
Patrol, the Enforcement Division of the Department of Liquor Licenses and
Control, and the Narcotics Division of the Arizona Department of Law.
Currently, DPS i s organized into five bureaus: Criminal Investigation,
Highway Patrol, Administration, Telecommunications, and Criminal Justice
Support. The department employs approximately 1,620 full- time employees
( FTEs) with an annual budget of $ 86 mi I I ion.
Telecommunications Bureau
Oversees Data Processina and
Telecommunications Svstems
The Telecommunications Bureau i s responsible for developing, operating,
and mai n ta i n i ng the department ' s data process i ng and commun i cat i ons
systems. Most of these systems operate statewide and provide essential
information services.(') Headed by the Assistant Director for
Telecommunications, the bureau i s composed of the Data Processing,
( 1) The Telecomnuni c a t i ons Bureau a1 so provides comnuni c a t i ons services t o several other
State agencies.
Arizona Criminal Justice Information System ( ACJIS), Technical
Communications, and Operational Comunicat ions ~ i v i s i o n s . ( l ) Each
division's s t a f f i n g level and responsibilities are as follows:
9 Data Processing has 55 FTEs. This division i s responsible for
developing and maintaining computer programs for the administrative,
enforcement, and investigative needs of the department, as well as
for the statewide criminal justice on- line network. The DPS computer
systems are operated 24 hours a day, seven days a week. Technical
assistance i s also provided to county and local criminal justice
agencies regarding linkages to the statewide criminal justice
computer network.
Arizona Criminal Justice Information System has 63 FTEs. The
division i s responsible for the statewide Arizona Computerized
Criminal History ( ACCH) network, which includes criminal arrest and
disposition information. The division also administers the
department's records systems for c i t a t i o n s , warrants, and motor
vehicle accident and offense reports. They also t r a i n DPS and other
agency personnel in the use of ACJIS and audit ACJIS user agencies.
Technical Comnunications has 74 FTEs and i s composed of four
sections: Engineering Services, Carrier Services, Telephone and Data
Services, and Radio Services. Engineering Services i s responsible
for design, development, and i n s t a l l a t i o n of a l l radio system
improvements and enhancements including the Emergency Medical
Services Communications ( EMSCOM) system, as well as for the
development of equipment specifications. Carrier Services maintains
the statewide microwave radio system and a l l mountaintop buildings,
towers, antenna systems, and emergency generators. Telephone & Data
Services i n s t a l l s and repairs the department's statewide telephone
and data communications systems, and maintains the records and
accounts associated with these systems. Radio Services i n s t a l l s and
repairs electronic equipment, including mobile and portable radios,
pagers, and mountaintop repeaters, at radio repair shops throughout
the State.
O~ erational Comnunications has 88 FTEs who provide dispatch services
for the department, for law enforcement personnel from other
agencies, and for the statewide EMSCOM system. Dispatch centers are
located in Flagstaff, Phoenix, and Tucson.
( 1 ) The bureau has four administrative s t a f f positions: Assistant Director. Executive
Secretary, Administrative Services Officer 111, and Major.
Budqet and Staffinq
Currently, the Telecommunications Bureau has 284 full- time equivalent
employees and a general fund budget of $ 14,234,600. See Table 1 for
further detail of the expenditures of the bureau.
TABLE I
DEPARTMENT OF PUBLIC SAFETY- TELECOMMUNICATIONS BUREAU
STATEMENT OF FTEs AND ACTUAL AND BUDGETED EXPENDITURES
FISCAL YEARS 1988- 89, 1989- 90, AND 1990- 91
( Unaudited)
1988- 89 1989- 90 1990- 91
Actual Actual Budqe t ed
FTE Positions 284 284 284
Expenditures
Personal Serv i ces $ 7,698,921 $ 7,943,596 $ 8,310,900
Employee Related 1 ,566,747 1 ,337,307 1,846,500
Professional and
Outside Services 135,039 93,398 105,000
Travel - In State 67,815 77,305 88,400
Travel - Out of State 13,117 16,788 12,300
Equ i pmen t 1 ,836,445 1,612,814 2,513,800
Other operating 2,161,412 2,096,513 905,400
ACJ IS 439.937 447.742 452.300
TOTAL
Source: Arizona Financial Information Systems, Department of Public
Safety financial reporting system and the State of Arizona,
Appropriations Report for the Fiscal Year Ending June 30, 1991.
- A Sco~ e
O u d i t report of the Department of Public Safety's Telecommunication
Bureau presents findings and recommendations i n six major areas:
@ the need for updating and establishing fees for services;
the adequacy of the audits performed by the ACJlS Division;
the adequacy of criminal history data maintained by DPS;
the need for strengthening controls over the acquisition and
assignment of telecommunications equipment;
the adequacy of the ACJlS Division's cash- handling procedures; and
@ the - fectiveness of the EMSCOM system in meeting user needs.
Our , dit scope was limited because we lack the statutory authority to
review criminal history data. Several areas within our audit were
impeded due to our i n a b i l i t y to access this data. For example, although
accuracy of criminal history data is c r i t i c a l , we were unable to assess
the re1 iabi l i t y of this data. In addition, we were unable to physically
inspect arrest cards and dispositions to determine and v e r i f y the extent
of backlogs. Further, although we were able to obtain a data tape of the
arrests on the ACCH system that lacked corresponding dispositions, we
were unable to obtain any identifying information for follow- up
purposes. Finally, we could not observe clerks entering criminal history
data: this impacted our soil i t y to conduct productivity studies.
This report also presents other pertinent information regarding
state- of- the- art technologies which DPS plans to implement ncluding an
automated fingerprint i d e n t i f i c a t i o n system ( see page 49), -< p i le d i g i t a l
terminals ( see page 52), a computer assisted dispatc system ( see
page 53), and an 800 MHz radio trunking system ( see ,& ye 54). In
addition, we present information on the cast of upgrading the Arizona
Criminal Justice Information System ( see page 55).
Th i s aud i t was conducted i n accordance w i t h genera 1 1 y accepted government
auditing standards.
The Auditor General and s t a f f express their appreciation to the Director
of the Arizona Department of Public Safety, the Assistant Director f o r
Telecommunications, and s t a f f for their cooperation and assistance during
the course of our audit.
FINDING I
DPS COULD GENERATE MORE THAN $ 700.000 ANNUALLY
BY ESTABLISHING A FEE FOR CERTAIN SERVICES
AND UPDATING CURRENT FEE SCHEDULES
By establishing a fee for certain services and updating current fee
schedu les, DPS cou Id generate more than $ 700,000 annual ly . DPS shou Id
establish a fee for conducting background checks on applicants for
licenses and employment. Further, DPS's recent modification of i t s
accident report fee schedule w i l l generate additional revenues.
DPS Should Establish
a Fee for Performinq
Backaround Checks
DPS should establish a fee for performing background checks on applicants
for l icenses and employment as do other states. I f DPS establ ished a $ 10
fee, i t could generate over $ 600,000 annual ly for the State.
DPS staff perform background checks for both criminal and other
agencies. For example, law enforcement agencies, school systems, and
State licensing agencies submit the names and fingerprints of prospective
employees and licensees t~ BPS to determine i f they have criminal
records. DPS conducts approximately 6,000 background checks per month
( over 5,000 for other agencies and almost 1,000 for criminal justice
agencies). Approximately 85 percent of these requests include both
federal and State investigations. BPS performs State background checks
at no charge, and then forwards these requests to the FBI for completion
at the federal level.(') The FBI charges other agencies a fee of $ 20
( 1) I n 1989, the ACJIS Division proposed revising A. R. S. $ 41- 1750 to e s t a b l i s h a State fee
wi t h a $ 20 cei 1 i ng; however, DPS never submitted t h i s proposal.
for i t s service, and allows DPS to retain $ 1 from each fee.(') Thus, DPS
receives only limited revenue for 85 percent of the background checks,
and no revenue for the remaining 15 percent.
In general, agencies have required prospective employees or licensees to
pay for the federal background check. The six State ager- les with the
largest number of requests for background checks pass the cast of the FBI
serv i ce on to the prospect i ve emp l oyees or l i censees . The two schoo l
systems generating the most requests also pass the $ 20 FBI service charge
on to prospective employees and licensees.
Other states have established a state ~ rocessina fee - Other states have
implemented a service fee for completing background checks. Arizona i s
one of only seven states that has not established such a fee. The fees
charged by 43 other states and the D i s t r i c t of Columbia are shown in
Table 2 ( see page 9). Table 2 also reveals the following information
a b ~ u tt hese states' fees for completing background checks:
o 21 of the 43 states l i s t e d have fees of $ 10 or more;
9 of the 10 western states charge a fee;
0 a l l states bordering Arizona except New Mexico charge at least a $ 10
fee and ;
the average state fee i s $ 9.35.
In addi t ion, in the past, at least one Arizona agency has charged
applicants a processing fee. For example, even when the FBI fee was oniy
$ 14, the Arizona Department of Education charged applicants $ 20 for
( 1) I n PI 1988- 89, DPS collected over $ 50,000 by retaining t h i s portion of the FBI service
charge. These revenues are used to pay the salaries of three l i m i ted- term employees
hired to perform background checks. According to the ACJIS Division manager, due to
an increase i n salaries and empl oyee- re1 ated expenses, revenue obtained from FBI fees
no longer covers the d i r e c t costs of these three limited- term employees. Also, FBI
revenues have never covered the i n d i r e c t costs, such as space, supervision, and
personal computers.
( 2) These six State of Arizona agencies, i n descending order of the volume sf requests f o r
background checks, are the following: Departments of Economic Security, Education,
Public Safety ( security guard licensing), Health Services, Insurance, and Real Estate.
State
TABLE 2
OTHER STATES' APPLICANT PROCESSING FEES(')
A l abama
A l aska
California
Co lorado
Delaware
Florida
Georgia
l daho
I l l inois
l nd i ana
l owa
Kansas
Kentucky
Lou i s i ana
Mai ne
Mary l and
Massachusetts
Michigan
Mi nnesota
Missouri
Mon t ana
Nebraska
Nevada
New Hampshire
New Jersey
New Mex i co
New Yo r k
North Carolina
North Dakota
Ohio
Ok l ahoma
Oregon ;
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Utah
Virginia
Washington
Washington, D. C.
West Virginia
W i scons i n
Wyom i ng
( a) I n addition to recovering the cost of processing applications, some states use the
application fee to fund special projects. For example, New York's fee includes a $ 10
surcharge to fund i t s Automated Fingerprint I d e n t i f i c a t i o n System.
Source: State of Arizona, Office of the Auditor General staff
compi lation from a 1989 criminal just ice consortium study. For
states with a range of fees, the fee presented is the one that
most closely resembles the processing performed by DPS.
c e r t i f i c a t i o n and retained the difference of $ 6 from each applicant to
cover the cost of processing operations.(')
BY establishina a $ 10 fee, DPS could aenerate over $ 600.000 - I f DPS
charged $ 10 for processing State background checks, the department cou Id
raise over $ 600,000 annually for the State. Since ten states, including
Colorado and Utah, charge $ 10, and 11 states, including California and
Nevada, charge over $ 10, the establ ishment of a $ 10 fee seems
reasonable. A. R. S. 941- 1750 should be amended to allow for the
establ ishment of t h i s fee.( 2)
DPS Has Recentlv Modified Its
Accident Re~ ort Fee Schedule
During the course of the audit, we noted the need for DPS to modify i t s
accident report fee schedule. DPS had been charging only people
requesting accident report copies for commercial purposes ( about 5
percent of a l l requests) a $ 6 fee.( 3) However, statutes allow the
department to charge everyone. Of the five states cited by the Deputy
Chief of the FBI's National Crime Information Center as having model
records operations, minimum fees for accident report copies range from $ 2
to $ lo.(" In addition, the ten largest c i t i e s and two largest counties
in Arizona a l l charge for accident report copies. Late in the audit, DPS
began charging everyone requesting accident report copies a fee of $ 1 per
page. The department expects to generate an additional $ 108,000 for the
State from this fee structure.
( 1) On March 1, 1990, the FBI increased i t s fee from $ 14 to $ 20.
( 2) I n addition, A. R. S. 515- 512, subsection C, regarding noncertified school d i s t r i c t
personnel, and A. R. S. 515- 534, subsection E, regarding the Ari zona Department of
Education applicant c e r t i f i c a t i o n fee, w i l l need to be amended, since they specify $ 20
as the cost of applicant processing.
( 3) A. R. S. 539- 121.03 defines comnercial purpose i n part as '% ny purpose i n which the
purchaser can reasonably anticipate the receipt of monetary gain from the d i r e c t o r
i ndi rect use of such pub1 i c record ."
( 4) Alabama, Arkansas, California, Nevada, and New York.
RECOMMENDATION
The Legislature should consider revising A. R. S. 941- 1750 to allow DPS to
establish a $ 10 fee for conducting other background checks on applicants
for l icenses and employment at the State level. Revenues not needed to
support t h i s processing should then be deposited in the general fund.
FINDING II
DPS's OPERATIONAL AUDIT PROGRAM IS DEFICIENT
Deficiencies are evident in the Department of Public Safety's Operational
Audit Program. DPS is not in compliance with a National Crime
lnformation Center policy requiring biennial audits of computer terminal
users of the Arizona Criminal Justice lnformation System ( ACJIS). In
addition, most of the audits that have been conducted by DPS to date have
been superficial. However, DPS is taking steps to improve the audit
function.
DPS maintains Arizona's Criminal Justice lnformation System which
contains criminal history record informat ion. Through ACJIS, users have
access to drivers license information, stolen vehicle f i l e s , warrants,
and criminal history data such as arrests and dispositions. In addition,
ACJIS provides Arizona users with an interface with the Federal Bureau of
Investigation's NClC database which contains criminal history information
on a national level. Local users enter records d i r e c t l y into NClC
through a computer terminal l inked into the network.(')
To ensure agencies comply with State and NClC policy and regulations
regarding data i n t e g r i t y , NClC adopted a policy in 1985 mandating
biennial audits of those agencies with computer terminal access to NCIC.
DPS i s responsible for conducting these biennial audits on Arizona
agencies. The audits are quality control checks to determine the
completeness and accuracy of records and to ensure that information
contained in those records is disseminated only to those e n t i t l e d by law
to the information.
( 1 M i ssi ng person, wanted person, unidentified person, and property records are placed
d i r e c t l y i n the NCIC by the o r i g i n a t i n g agency ( agency holding the warrant, missing
person report, or t h e f t report) through a control terminal t i e d i n t o the network.
DPS Fails to Com~ lwy ith NCIC
Biennial Audit Requirement
DPS has not audited every terminal agency biennially as required.(')
Since the audit program was i n i t i a t e d in 1987, only 39 percent of the
terminal agencies have been audited by DPS on a biennial basis. Without
these audits, deficiencies may go undetected. The noncompliance appears
to be caused by other tasks taking higher p r i o r i t y and poor management
information.
The maioritv of terminal aaencies in Arizona have not received mandated
biennial audits - Although DPS i n s t i t u t e d a mandated biennial audit
program in 1987, we reviewed DPS audit r - ds and found 107 of the 176
terminal agencies had not been audited ekery other year as required.
Since 1987, DPS conducted 168 audits of terminal agencies.( 2) Although
most agenc i es have been aud i ted at l east once s i nce the program began i n
1987, only 69 of the 176 agencies have been audited biennially as
required. In fact, 13 agencies have never had an audit and 25 agencies
are being audited for the f i r s t time t h i s year. The Phoenix Police
Department, the largest user of the system, w i l l be audited for the f i r s t
time t h i s year.
Problems with local terminal aqencies so undetected - As a result of
inadequate audits, deficiencies are not detected in a timely manner.
When audits are conducted, the department may identify deficiencies such
as improper dissemination of information, inaccurate or incomplete data
on the system, or system security problems. The department can then
require the agency to take corrective action. However, since the
department i s not conducting regular audits o f agencies performance,
deficiencies may not be identified. An example of the types of problems
i d e n t i f i e d in a local j u r i s d i c t i o n audit follows:
( 1) A terminal agency has d i r e c t terminal access t o NCIC data and i n many cases has
authority to enter records on the system.
( 2) The term a u d i t r e f e r s to the operational audits which DPS conducts to comply with the
mandated biennial audit requirement. DPS also conducts directed audits of agencies.
However, these i n v o l v e i n v e s t i g a t i n g s p e c i f i c a1 1 egati ons of system misuse and do not
address a l l areas covered i n an operational audit.
A County S h e r i f f ' s Office was f i r s t audited by DPS in 1989. The
agency scored 45 out of a possible 137 points. Two very c r i t i c a l
areas ( dissemination of data and f i le accuracy) scored very low. The
audit revealed an error rate of 94 percent in the wanted person f i l e
and an error rate of 86 percent in the stolen vehicle f i l e .
Consequently, the agency was directed to do a complete validation of
al I their ACIC/ NCIC entries within 60 days.(') The agency
subsequently sent a l e t t e r t o DPS informing them that they had
validated a l l their records and were now in compliance. DPS
conducted a follow- up audit of the agency 8 months later and found
that although the agency's dissemination procedures had d r a s t i c a l l y
improved, error rates were s t i l l very high in the f i le accuracy
area. Again the agency was advised to do a complete record
validation within 60 days.
The importance of the audit process i s further supported by the fact that
within eight months of the S h e r i f f ' s Office s t a t i n g i t was in compliance,
f i l e accuracy errors were again identified. Inaccuracies on the system
could cause serious legal ramifications especially i f an inaccurate
record resulted in an innocent person being wrongly arrested. In
addition, the i n t e g r i t y of the system i s seriously compromised when data
accuracy i s questionable.
DPS's f a i lure to complv with biennial audit reaui rement i s due to a few
factors - In addition to conducting routine audits, DPS Operat ions Audit
s t a f f have other responsibilities, some of which are viewed as higher
p r i o r i t y than operations audits. In addition, identifying agencies which
require auditing i s hindered due to the lack of adequate management
information.
( 1) The ACIC ( Arizona Crime Information Center) contains the same type of information
( stolen vehicles and wanted persons) as the NCIC. ACIC information i s more
comprehensive since NCIC only accepts offenses which meet i t s c r i t e r i a . For example.
about 98 percent of the warrants i n the ACIC are not on the NCIC data base because
out- of- state e x t r a d i t i o n i s not indicated or the offenses don't meet NCIC c r i t e r i a .
The ACJlS Operations Audit staff have various responsibilities in
addition to conducting routine operational audits of terminal
agencies.(') ACJlS audit staff also inspect the s i t e of terminal
locations, work with local agencies to help them obtain access to the
system, investigate allegations of system misuse, and do various other
tasks. In addition, audit staff spend a considerable amount of time on
administrative tasks.(*) For example, in 1989-- 24 percent of available
staff time was devoted to administrative tasks. In contrast, the unit
has never devoted more than 23 percent of i t s time to routine audits. In
fact, since the audit function began in 1987, DPS has cancelled 77
routine audits to accomplish other tasks.
Furthermore, DPS does not keep a l i s t of agencies due to be audited or
the date the agencies were last audited ( i f they have ever been
audited). This prevents the ACJlS auditors from easi ly identifying those
agencies that need to be scheduled for an audit. In order to determine
exactly how many agencies required audits and whether the agencies had
been audited in compliance with the biennial requirement, we requested a
master l i s t i n g of terminal agencies required to be audited. However,
ACJlS audit staff had no such master l i s t i n g for audit purposes. In
early 1990, ACJlS staff identified 14 agencies which had not yet been
scheduled for an audit. ( Our review subsequently identified three
additional agencies that had never been scheduled for an audit that ACJlS
staff fai led to identify.) Further, the only way DPS can determine
whether an agency has been on the sys tem long enough to even requ i re an
audit is by searching the agency f i l e .
Most: Audits Conducted by
DPS Have Been Superficial
Even when DPS does conduct audits, the audits are very limited and their
effectiveness is questionable. The majority of audits conducted have
( 1) The Operations Audit Unit has 6 s t a f f : f i v e auditors and one supervisor.
( 2) Administrative tasks include research, analysis, correspondence, and audit report
writing.
been questionnaires only with no site v i s i t s by audit staff or review of
agency records. In addition, there are other indications of weaknesses
in the audit program. Other states1 audit programs appear stronger in
comparison.
Most audits have only included questionnaires - Although DPS has
identified three types of audits ranging from a questionnaire audit to a
s i t e v i s i t with extensive data testing, most of the audits conducted by
DPS thus far have been limited to the questionnaire or level I audit. A
level I audit consists of mailing the terminal agency a questionnaire
which an agency representative completes and returns to the audit section
for grading. In a level I audit, the agency s i t e is not typically
visited nor are any agency f i l e s reviewed and compared against data on
the system to check for accuracy and completeness. Of the 168 audits DPS
has conducted in the past three years ( 1987, 1988, and 1989), 148 or 88
percent have been level I audits.
During the same time period, only 20 agencies received more extensive
audits. Nineteen agencies received a level II audit which entails
v i s i t i n g the agency s i t e and compi ling and reviewing a random sampling of
the agency's records. In addition, one agency received a level I l l
audit which is similar to a level I I but involves more extensive data
sampl ing.
Effectiveness of audit procrram i s questionable - DPS1s audit program
appears to be ineffective, especially given the reliance on questionnaire
audits. A recent audit conducted by NClC appears to indicate that the
department's audit program is not adequate. In 1987 and 1989, NClC
audited the Department of Public Safety because of i t s responsibility to
ensure agencies are in compliance with NClC policy. As part of the audit
of DPS, NClC reviewed selected local terminal agencies to ensure
compliance with NClC regulations. Both audits revealed that local
agencies were not complying with the NClC validation policy which
requires that agencies take steps-- such as calling the owner of a
reported stolen vehicle to ensure the vehicle is s t i l l missing-- to ensure
that a record on the system i s s t i l I valid.(') According tc \ JCIC, the
biennial audit i s one way to monitor local agencies' procedures to ensure
they are complying with the NCIC validation requirements. Although the
level I questionnaire audit does request agencies to identify t h e i r
validation procedures, no follow- up i s performed to ensure what i s put on
paper i s what i s done in practice.
In addition, DPS's own data seems to point to a weakness in i t s audit
program. Audit scores show that DPS audit s t a f f discover considerably
more deficiencies when conducting on- site v i s i t s as opposed to relying on
self- reported responses to i t s questionnaires. For example, scores
averaged 90 percent for the 148 agencies receiving level I audits while
scores averaged only 79 percent for the 19 agencies which had a level ! I
audit.
Furthermore, the current audit process f a i l s to take into consideration
various remote sites that also house terminals. Presently, only one
audit i s scheduled per terminal agency even though an agency may have
terminals at various locations. ( For example, Phoenix Police Department
has over 200 terminals located at various substations around the valley
and Department of Corrections has terminals located at i t s prisons across
the State). In the case of a level I audit, one person i s responsible
for f i l l i n g out the questionnaire audit. However, DPS l u d i t s t a f f
recognize that one person may not have knowledge of OPE Dns at the
agency's various remote sites. Even when conducting a lever II or level
I l l audit, only one s i t e is v i s i t e d even though an agency may have
terminals at various locations around the State.
( 1) The v a l i d a t i o n process reqt+- . s the entering agency t o compare i t s support documents
t o the information entered : e system. ( The entering agency i s supplied a p r i n t o u t
of a sample of records tered.) I n addition, the agency must consult with
appropriate sJtrrces ( compla nant, prosecutor, court, etc.) to ensure the accuracy,
completeness, and v a l i d i t y of the record. I n 1987, NCIC found 8 of 12 ( 66.6%)
agencies v i s i t e d f a i l e d to validate t h e i r vehicle records properly, and 3 of 13
( 23.08%) f a i l e d to validate t h e i r wanted person records properly. I n 1989, 7 of 15
( 46.6%) agencies with vehicle records and 4 of 7 ( 57.14%) with wanted person records
f a i 1 ed t o val i date thei r records proper1 y .
Other state's audit Droarams are more in- depth - We contacted five other
states to obtain information on their audit process and found that in
comparison, Arizona's audit program appears weak.(') None of the other
f i v e states use a questionnaire as the e n t i r e a u d i t , although some use a
questionnaire to f a c i l i t a t e the audit process and identify areas that
need more scrutiny. A l l five identified a s i t e v i s i t as part of each
audit. In addition, Nevada, New York, and California also v i s i t a l l
remote terminal sites while Alabama spot checks terminals at remote
sites. Furthermore, these states review a sample of records for accuracy
and completeness as part of each audit conducted.
DPS Is Takinq
Steps to Improve
DPS management i s aware of many of the prob l ems men t i oned above and has
begun taking corrective action. ACJlS audit s t a f f recognized the
inadequacy of the questionnaire audit and have scheduled mainly level I I
audits for 1990. The audit s t a f f plan to conduct 69 audits in 1990; 49
of those are scheduled to be level II audits.( 2) According to a May 1990
memo " the plan i s to audit a l l agencies at a level II to place more
emphasis on f i l e accuracy and the importance of the validation process."
In addition, any level I audits conducted w i l l be done on s i t e as well
" to ensure the most complete and accurate information i s obtained during
the audit." Other planned improvements include:
Implementing an automated database which w i l l provide complete audit
histories sf every terminal agency as well as track and monitor
audits in progress.
Incorporating v i s i t s to remote terminal location areas to ensure
policies and regulations are adhered to at these sites.
Auditing non- terminal agencies which receive confidential criminal
history record information to ensure the information i s being used
only for legitimate purposes.
( 1) We contacted Alabama, Arkansas, California, Nevada, and New York because they were
recomnended as model states by a NCIC representative.
( 2) To be i n compliance with the NCIC biennial audit requirement, 88 audits would need to
be scheduled each year given the current 176 terminal agencies.
RECOMMENDATIONS
1. DPS should take steps to bring the department into compliance with
the NClC biennial audit requirement.
2. DPS should make every effort to conduct on- sit8 ( level I I or level
I l l ) audits when possible.
3. DPS should continue to implement improvements to its audit program
such as visiting remote terminal locations and auditing non- terminal
agenc i es.
FINDING Ill
DPS NEEDS TO IMPROVE THE COMPLETENESS AND
RELIABILITY OF ITS CRIMINAL HISTORY INFORMATION
DPS needs to take steps to upgrade the completeness and r e l i a b i l i t y of
i t s criminal history data. Statutes require DPS to maintain criminal
history record information on individuals arrested by Arizona law
enforcement agencies. However, the integrity of this data is compromised
by the untimely entry of arrest data and missing data on case
dispositions. Adding to the problem, DPS does not routinely verify
arrest and disposition data entered.
DPS Must Maintain Criminal
Historv Information on Individuals
Arrested by Local Asencies
Statutes require DPS to maintain criminal history information on
individuals arrested by Arizona law enforcement agencies. Because this
criminal history data is used by both criminal justice and other agencies
when making c r i t i c a l decisions, i t is important that the information be
complete and accurate.
DPS maintains criminal historv data - Statutes require that DPS maintain
criminal history record information on offenders arrested by Arizona law
enforcement agencies. A. R. S. 541- 1750 states that DPS shall " procure and
maintain records of photographs, descriptions, fingerprints, dispositions
and such other information as may be pertinent to a l l persons who have
been arrested for or convicted of a pub1 ic offense within the state."
The ACJlS Division has the responsibility of maintaining the State's
criminal history system, the Arizona Computerized Criminal History
network ( ACCH).
ACCH contains arrest and disposition data on offenders who have been
arrested in Arizona. By statute, arresting agencies are required to
forward arrest data to DPS, while courts must forward disposition
information to DPS. I n f i s c a l year 1989- 90, a monthly average of 9,000
arrest fingerprint cards and 9,300 disposition forms were received by the
ACJlS Division.
Criminal justice and other aaencies access ACCH when makina important
decisions - ACCH data provides input into decision making at both
criminal justice and other agencies. Criminal history information
assists members of the criminal justice community in making release,
bonding, and penalty decisions, and also assists o f f i c e r s in completing
criminal investigations. For example, many judges w i l l release suspects
on bond i f they cannot review the suspects1 criminal history records.
This provides suspects an opportunity to flee. I f judges could review
the records, however, they might opt to keep the suspects in custody.
Recent legislation allows many other agencies access;. to ACCH data for the
completion of background checks for employment and licensing applicants.
For example, ACCH i s used to determine i f prospective teachers and chi Id
care workers have criminal records. ACJlS Division s t a f f disseminate
a l l criminal history information ( convictions, acquittals, and open
arrests) to most employment and licensing agencies.
A U. S. Department of Justice report on criminal history data quality
notes that the existence of incomplete or inaccurate data increases both
the potential of causing unjustified harm to individuals and the States1
s u s c e p t i b i l i t y t o l i a b i l i t y . ( ' ) Thus, the maintenance of timely,
complete, and accurate criminal history data is essential.
Arrest Data Not Entered
on Timelv Basis
Untimely entry of arrest informat ion compromises the re1 iabi l i ty of ACCH
data. Local law enforcement agencies do not always submit arrest data to
DPS in a timely manner. Further, once arrest data i s received, DPS does
not consistently input the data on a timely basis. In addition,
( 1) United States Department of Justice. Strateqies f o r Improvinq Data Ouali ty. Report
of work performed under Bureau of Justice S t a t i s t i c s grant, awarded to SEARCH Group,
Inc. A p r i l , 1989.
procedures for resolving problematic arrest cards cause delays in
updating the criminal history data base.
Submission of arrest cards not alwavs timely - Some local law enforcement
agencies do not forward arrest cards to DPS on a timely basis.
Establishing a statutory time l i m i t could result in more timely
submission.
Working with DPS s t a f f , we compiled, for arrest cards input during a
7- day period, the dates of arrest and the dates the arrest data arrived
at DPS. Analysis of the data revealed that on average, 24 days had
elapsed between the two dates. Many agencies are timely in their
submission of arrest fingerprint cards, as 44 percent of the agencies
submitted cards within 18 days of the arrest. For example, the Phoenix
Police Department submitted arrest cards on average 6 days after the date
of arrest. However, other agencies are not as timely: 24 percent of the
agencies submitted arrest cards over 30 days after the arrest. For
example, one arrest card arrived at DPS 788 days after the date of
arrest. Another c i t y ' s police department submitted arrest cards on
average 157 days after the dates of arrest.
Staff from the two largest police departments having problems submitting
arrest cards on a timely basis noted that arrest card completion has not
been given a high p r i o r i t y . Consequently, the departments have not
dedicated an adequate number of staff to arrest card tasks, as the
following example illustrates.
Staff from one of these c i t i e s ' police departments attribute the
untimely submission of arrest fingerprint cards to a lack of staff
assigned the task of completing arrest cards; Until March 1989,
county j a i l staff completed a l l arrest cards for this particular
police department. After March 1989 the county decided that i t would
s t i l l take individualsf fingerprints while booking them into j a i l ,
but would not complete the information portion of the arrest cards.
For more than a year after the police department assumed the
responsibility of completing arrest cards, no additional personnel
were assigned to f u l f i l l the responsibility. This resulted in a
backlog of arrest cards which was at one point 10 feet high.
A statutory time l i m i t could result in more timely submission of arrest
cards. The Department of Justice report recommends that states institute
mandatory reporting laws which specify a time period within which local
agencies should submit criminal history information. The report states
that such statutes have proven " highly effective" in some cases.
Last year the ACJlS Division proposed revising A. R. S. § &"- I750 to require
local agencies to submit arrest fingerprint cards w n 10 days of
arrest, but DPS never submitted the legislation. Hor r , many local
agencies may not be able to submit cards within 10 da2-.. For example,
Flagstaff and Tucson Police Department staff indicated that a 15- day
l i m i t would be more reasonable. DPS staff should consult the local law
enforcement agencies in order to determine a reasonable statutory time
limitation. Once a time li t is established, DPS should monitor the
timeliness of agencies' submission of arrest cards in order to ensure
compliance.
Entry of arrest data does not always occur on a timely basis - Once DPS
receives arrest data, i t does not always input the data in a timely
manner. Staff vacancies are a cause of untimely entry.
DPS is untimely in entering arrest data. We analyzed arrival and input
dates of arrest data entered during a 7- day sample period. Although
f i r s t offender arrest entry ( performed by the day s h i f t ) occurred on
average 6 days after arrival of arrest cards at DPS, entry of repeat
offender arrest data ( performed by the evening s h i f t ) occurred on average
23 days after arrival of arrest cards at DPS, ? is turnaround time does
not meet the division's goal of processing arrest cards within 14 days of
receipt.
Our review also indicated that evening s h i f t s t a f f do not consistently
enter arrest cards in the order in which they arrived at DPS. For
example, on June 7 the evening s h i f t entered arrest cards which had
arrived on May 31 ( a 7- day turnaround), and on June b arrest cards which
had arrived on May 14 were entered ( a 24- day turnaround). Because
arrival dates are not reviewed when choosing which arrest cards to enter,
daily turnaround times of repeat offender data entry vary.
Staff vacancies may represent part of the cause of untimely entry. The
evening s h i f t has 6 clerk positions. During the f i r s t half of 1990, two
experienced clerks l e f t . Also, two other positions were f i l l e d with
inexperienced clerks during the same time period.
i n s u f f i c i e n t ~ rocedures for resolvina ~ robiematic arrest cards - Current
ACJIS Division procedures for resolving problematic arrest cards
contribute to the unreliabi l i t y of ACCH data. The procedures place the
responsibility for correcting and resubmitting cards on the local law
enforcement agencies. Other states take more responsibility for seeing
that rejected cards are corrected and the data entered.
DPS rejects arrest cards for several reasons. I f cards contain i l l e g i b l e
fingerprints, they are rejected. Also, DPS rejects arrest cards i f they
contain inaccurate or incomplete information. For example, the year of
arrest i s indicated frequently as the year of b i r t h . In fiscal year
1989- 90 DPS rejected 2,959 arrest cards, or three percent of the cards
received, because the fingerprints were i l l e g i b l e . DPS also rejected
2,113 arrest cards, or almost two percent, because they contained
inaccurate or incomplete information.
ACJIS Division procedures for resolving rejected arrest cards do not
guarantee r e l i a b l e ACCH data since DPS r e l i e s on local agencies to
correct and resubmit rejected cards. When a card i s rejected, ACJIS
Division s t a f f enter the arrest data in an error data base and return the
card to the submitting agency. I f the agency corrects and resubmits the
card, ACJIS s t a f f w i l l delete the entry from the error data base and
enter the arrest data into ACCH. However, the likelihood that the arrest
data w i l l be entered into ACCH in a timely manner i s low. The average
age of arrest records in the error f i l e i s 323 days.(') The ACJlS
Division does not attempt to resolve the problems with these records.
Other states have established procedures which, i f implemented by the
ACJIS Division, could result in improved arrest recording. Four of the
( 1) The 323- day figure i s based on a review of the 6,630 arrest records residing i n the
error f i l e as of June 5, 1990. DPS began recording rejected records i n an error f i l e
July 1, 1988.
f i v e states i d e n t i f i e d as model states by the Deputy Chief of the
National Crime Information Center ( NCIC) do not return cards with
information problems to the local agencies. Rather, these four states
( Alabama, California, Nevada, and New York) telephone the submitting
agencies to obtain the needed information. By implementing similar
procedures, DPS could resolve almost 30 percent of the records in the
error f i l e on a timely basis and ensure that the records were entered
into ACCH.(')
Calling local agencies w i l l not, however, resolve problems with cards
rejected because of i l l e g i b l e fingerprints. In tt- Te cases, DPS should
consider entering the cards into ACCH. Frequently, , y the time an arrest
card having an i l l e g i b l e p r i n t i s rejected and returned to the submitting
agency, the subject i s no longer in custody. Thus, the agency cannot
obtain new fingerprints and resubmit the card. Three of the states
contacted ( California, Nevada, and New York) enter such cards and flag
that the record i s not supported by fingerprints. Any subsequent
searches on the subject w i l l result in a " conditional h i t t t because of the
lack of fingerprints. Currently, DPS has 4,603 arrest records in the
error f i le. Although these records can be searched, they typical ly are
not used because the record i s not supported by fingerprints.
Numerous ACCH Arrest
Records Are Incomplete
Incomplete arrest records also contribute to the u n r e l i a b i l i t y o f ACCH
data. Numerous ACCH arrest records do not have corresponding
dispositions. A disposition backlog explains some of these open
arrests. The existence o f i n s u f f i c i e n t procedures for resolving rejected
( 1) Although t h i s process may require additional s t a f f , because we were r e s t r i c t e d from
working with ACCH data we could not perform any p r o d u c t i v i t y analyses to determine
where s t a f f i n g , i f any, might be required. However, four of the f i v e s t a t e s ' criminal
j u s t i c e agencies we contacted have s t a f f dedicated to correcting problem cards. The
number and sophistication of the units responsible f o r c o r r e c t i n g a r r e s t cards
varies. The New York Division of Criminal Justice Services has two s t a f f devoted to
arrest card resolution. The C a l i f o r n i a Department o f J u s t i c e resolution u n i t has 20
staff; however, these s t a f f do not j u s t process problematic arrest cards. For
example, four s t a f f are devoted to t r a i n i n g local agency s t a f f .
disposi t ion forms may represent another cause of the numerous open arrest
records. Because of the importance of ACCH data, DPS should i n i t i a t e
procedures for identifying and completing open arrest records.
Disposition forms are initiated at the time of arrest and completed by
agencies terminating arrest cases. For example, county or prosecuting
attorneys can decide to modify or not f i l e arrest charges. When cases go
to court, court staff note on the forms whether individuals were
convicted or acquitted of the charges. ACCH arrest records are open
u n t i l dispositions are forwarded to and entered by ACJlS Division s t a f f .
Numerous ACCH arrest records lack correspondina dispositions - Many ACCH
arrest records do not have dispositions. There are approximately 720,000
arrest records in ACCH. Of those arrests occurring before May 1988,
almost 312,000 have at least one count which does not have a
disposition. Of those arrests occurring during and since May 1988,
almost 129,000 have at least one count which does not have a
disposition. Thus, over 60 percent of the arrest records are missing at
least one disposition. Although some of the more recent arrest records
may not have been resolved yet, those more than two years old should have
been resolved and therefore should contain dispositions.
Many of the arrest records missing dispositions represent serious
offenses. Table 3 ( see page 28) shows a compilation of serious counts
that do not have corresponding dispositions in ACCH. Arrests occurring
between 1975 and 1987 are shown. As most arrest cases are resolved
within two years of the arrest, these counts should have been resolved
and thus should have dispositions recorded. As Table 3 shows,
dispositions are not recorded for almost 1,400 homicide counts, over
4,000 sexual assaults, almost 6,000 robberies, and over 2,500 child
molestation counts. The lack of dispositions for these and other counts
could impede criminal investigations. The missing disposition data also
could increase the State's susceptibility to l i a b i l i t y . For example, i f
a person was den i ed emp l oymen t or l i cens i ng because ACCH showed an open
arrest when the person had actually been acquitted, the person could sue
the State.
Year
Total :
TABLE 3
ARREST COUNTS WITHOUT DISPOSITIONS(')
Nuinber of
Homicides
Number
of Sexual
Assau l ts
61
127
82
11 1
209
313
345
304
327
426
533
449
737
Number of
Robber i es
N d er
of Chi Id
Molestations
( a) Data i n table based on ACCH data tape provided to Auditor General s t a f f by DPS.
I d e n t i f y i n g information, such as individual s t names and i d e n t i f i c a t i o n numbers, was
not provided. Thus, Auditor General s t a f f could not v e r i f y the data provided.
Source: Auditor General sta. npilation G 3PS data tape of ? sts
Raving at least onr ~ n tth at does not have a disp~. : ion
recorded.
D i s ~ o s i t i o n backlm - A backlog of disposition fo- rms explains some sf
these open arrest records. As of March 1990, the ACJlS Division had
123,812 backlogged disposition forms awaiting entry into ACCH. This
backlog accumulated between October 1987 and May 1989, a period during
which DPS quit entering dispositions.
The causes of the backlog suggest poor managerial oversight. The ACJlS
Division quit entering dispositions so that i t could address arrest card
backlogs.(') Arrest cards were backlogged because o f s t a f f participation
on a task force overseeing the move to the new DPS building, supervisor
rotation, and special programs. ACJIS Division management should have
controlled these events to prevent backlogs. ~ or'example, during the
l a t t e r h a l f of 1987, two experienced fingerprint technicians devoted 251
hours to the move task force. In addition, DPS management directed the
division to undertake a special project to provide Cal i f o r n i a with over
20,000 arrest cards f o r arrests occurring along the Arizona- California
border. This project was completed at the expense of not entering data
on Arizona's own system. Although the ACJIS Division has i n i t i a t e d
e f f o r t s to eliminate the disposition backlog, as of June 1990 over
107,000 dispositions awaited processing.
Insufficient procedures for r e s o l v i n ~ reiected dispositions - The
existence of i n s u f f i c i e n t procedures for resolving rejected dispositions
may be another cause of the numerous open arrests. In fiscal year
1989- 90, BPS rejected over 30 percent of the disposition forms received.
Like arrest cards, some disposition forms may be rejected i f they contain
inaccurate or incomplete data, such as an incorrect date of b i r t h .
According to the disposition processing supervisor, however, bad data
accounts for only 2 to 3 percent of the disposition forms rejected.( 2)
The majority of the dispositions rejected are denied because they do not
match ACCH arrest records. When these dispositions are rejected, DPS
enters the disposition data into the error data base and returns the
forms to the arresting agencies. ACJIS s t a f f do not attempt to correct
these dispositions or identify what happened to the corresponding arrest
card .( 3)
( 1) Backlogs of arrest cards existed at several stages of the entry process. As of
September 1988, almost 16,000 arrest cards had to be name searched ( the f i r s t task of
the entry process), over 12,000 f i r s t offenders had to be assigned State
i d e n t i f i c a t i o n numbers before entry could occur, over 16,000 repeat offender arrest
cards awaited entry, and over 2,500 arrest cards needed review by f i n g e r p r i n t
technicians before entry could occur.
( 2) The ACJIS Division should also handle these forms by c a l l i n g the agencies t o obtain
the needed information.
( 3) There were 34,174 disposition records residing i n the error f i l e as o f June 5, 1990.
Returning rejected dispositions to the arresting agencies i s
ineffective. Dispositions are returned to the arresting agency in an
attempt to obtain the corresponding arrest cards. However, local law
enforcement agency s t a f f interviewed stated that by the time dispositions
are rejected by DPS and returned to them, the individuals are generally
no longer in custody. Thus, the agencies cannot take a new set of
fingerprints to resubmit corresponding arrest cards. Given t h i s
situation, the local agencies either discard the dispositions or place
them in their own offender f i l e s . In either case, the dispositions are
never returned to DPS for entry into ACCH.
The ACJlS Division should revise i t s procedures regarding rejected
dispositions. F i r s t , DPS could retain the forms and t r y to locate
records of the arrests. For example, it could review arrest records in
i t s error data base to determine i f any match the dispositions. Second,
DPS could i n i t i a t e the use of unique tracking numbers i n order to
f a c i l i t a t e proper linking of arrest and disposition data. Three of the
f i v e states contacted have implemented tracking number systems. Under
these systems, a tracking number is assigned to an arrest at the time of
arrest, and a l l subsequent information regarding the arrest includes the
number. Thus, arrests and dispositions can be matched even i f some of
the information on the arrest cards and dispositions d i f f e r .
Need for proqram to address oDen arrest records - Because of the
imp~ stance of ACCH data in completing criminal investigations and
pre- employment and pre- licensing background checks, DPS should identify
and complete arrest records which do not have dispositions.(') In
addition to ensuring that disposition processing i s not backlogged and
that rejected dispositions are entered into ACCH in a timely manner, the
ACJlS Division should ensure that a l l dispositions are reported. Most
states do not receive dispositions for a l l arrests reported. The U. S.
Department of Justice report recommends that states monitor reporting of
( 1 ) I n 1989 the C a l i f o r n i a Department of Justice was ordered by the court to attempt to
obtain missing disposition data p r i o r to disseminating arrest information f o r
employment purposes.
dispositions. Flagging arrest records over a certain age which do not
have dispositions, that is, arrest records which should have
dispositions, represents the f i r s t step of such a program. Next, the
disposition data must be acquired and entered.
DPS Does Not Routinelv Verify
Data Entered into ACCH
The ACJlS Division's failure to routinely verify the data entered into
ACCH contributes to the u n r e l i a b i l i t y of ACCH data. Although we could
not determine the accuracy of individual ACCH records, we noted a lack of
data entry verification procedures. DPS should implement some of the
verification procedures used by other states1 criminal history records
staffs. In addition, DPS should plan a comprehensive data quality audit.
Data entered into ACCH i s not routinely verified - Currently, arrest and
disposition data entered into ACCH is not routinely verified to ensure
data entry accuracy. Although we were unable to review the accuracy of
individual ACCH records, we noted a lack of mechanisms allowing data
entry errors to be detected.
We were unable to verify the accuracy of individual ACCH records because
access limitation laws do not allow for auditing criminal history
records. State statutes specify that criminal history records are only
available to criminal justice agencies performing criminal investigations
and other agencies completing pre- employment and pre- licensing background
checks. The DPS Assistant Attorney General representative and our
Counsel agreed that statutes do not provide Auditor General staff
authority to access ACCH data.
However, we noted deficiencies in existing data entry quality control
mechanisms which led us to question the accuracy of ACCH records.
Currently, there are three ways in which ACJlS Division staff detect data
entry errors. F i r s t , a supervisor may uncover errors while performing
periodic quality control checks of a clerk's entry. Second, clerks may
identify arrest data entry errors while entering disposition data.
Third, data entry errors can be uncovered i f a subject reviews and
challenges his or her record. In addition, local agencies do not
routinely v e r i f y ACCH data. S t a f f o f local law enforcement agencies
interviewed noted that unless a subject approaches them to review his or
her record or unless some external impetus occurs, they would not detect
a data entry error in an ACCH record depicting an arrest made by one of
thei r o f f i c e r s . ( ' )
DPS should iumlanent v e r i f i c a t i o n ~ rocedures used bv other states - The
ACJlS Division should develop procedures to routinely v e r i f y the data
entry accuracy of arrest and disposition data input into ACCH. Four of
the five states contacted v e r i f y 100 percent of the: data entry performed
by state criminal history records s t a f f . ( One of these states, Nevada,
has used different v e r i f i c a t i o n procedures in the past and feels the
procedure requiring double entry of criminal history data i s most
e f f i c i e n t . ) The f i f t h state contacted has procedures whereby the data
entry supervisor v e r i f i e s a s t a t i s t i c a l sample of records entered.
DPS should orqanize a comprehensive data qualitv audit - In addition t o
improving i t s data entry v e r i f i c a t i o n procedures, the ACJlS Division
should plan a comprehensive data quality audit to determine the accuracy
o f existing data. The U. S. Department of Justice report recommends that
states provide for " baselinew audits of criminal history records
repositories. Audits should include evaluations of data quality control
procedures as well as the completeness and accuracy of criminal history
records. The report recommends, " I f possible, the audit should be
performed by an outside contractor or by an independent agency such as
the state a u d i t o r ' s o f f i c e . " The ACJlS Division has never undergone a
baseline audit, although i t plans to have such an audit done in fiscal
year 1991- 92. In order for this audit t~ be successful, the agency
performing the audit must be granted access to ACCH records.
( 1) An example of an external impetus causing an agency to v e r i f y an ACCH record i s the
FBI informing an agency that a subject has two state i d e n t i f i c a t i o n numbers.
RECOMMENDATIONS
1. The ACJlS Division should obtain input from local law enforcement
agencies and identify an acceptable time l i m i t for submission of
arrest fingerprint cards. DPS should petition the Legislature to
modify A. R. S. $ 41- 1750 accordingly.
2. ACJlS Division management should more closely monitor turnaround
times, backlogs, and staffing levels for entering arrest data.
3. The ACJlS Division should improve i t s procedures for processing
problematic arrest cards and disposition forms. The following should
be incorporated into the revised procedures:
a. The error data base should be used to identify outstanding
arrest cards and disposition forms which have been rejected.
b. ACJlS staff should call agencies to obtain needed information
rather than return the documents to the submitting agencies.
ACJIS Division management should consider dedicating staff to
the correction of problematic documents.
c. DPS should consider entering arrest data and flagging arrest
cards having i l l e g i b l e fingerprints.
d. In cases where a disposition must be rejected because there is
not a corresponding arrest record, the ACJlS Division should
consider obtaining arrest data i t s e l f rather than returning
disposition forms to agencies. DPS should also consider
implementing a unique tracking number system to f a c i l i t a t e the
matching of arrests and dispositions.
4. DPS should implement a system to identify older arrest records which
lack dispositions. Once identified, DPS should take steps to obtain
and enter missing disposition data.
5. The ACJlS Division should implement procedures to verify at least a
s t a t i s t i c a l sample of arrest and disposition records input. Double
entry should be considered as a verification mechanism.
6. DPS should ensure that a comprehensive data qua1 i ty audit, planned
for FY 1991- 92, occurs. The agency performing the audit must be
granted access to ACCH data in order for the audit to be successful.
FINDING IV
DPS NEEDS TO STRENGTHEN POLICIES AND PROCEDURES
OVER THE ACQUISITION AND ASSIGNMENT
OF TELECOMMUNICATIONS EQUIPMENT
A l though DPS spends on average over $ 560,000 annual ly on
telecommunications equipment, the department's policies and procedures
for the acquisition and assignment of telecommunications equipment need
improvement.(') DPS has taken steps to enhance controls, but further
e f f o r t i s needed.
Traditionally, DPS telecommunications equipment has consisted primarily
of mobile/ vehicular radios. However, the technological advances of the
past few years have resulted in the expansion of the department's arsenal
of communications equipment. Portable radios, cellular/ mobile phones,
pagers, facsimile machines, answering machines, and mobile display
terminals are some of these recent additions.
We reviewed the communications equipment purchased in the last three
years ( f i s c a l years 1987- 88 through 1989- 901, and found that the
department purchased approximately 1,000 u n i t s o f new and replacement
equipment at a cost of about $ 1.6 m i l l i o n . The equipment purchases we
reviewed did not include large or unique items such as radio consoles or
telephone switches, but rather the communications equipment most l i k e l y
to be assigned to individual DPS employees.
Existina Policies and Procedures
Are lnsuff icient
Given the amount of money DPS spends annual l y on te l ecommun i cat ions
equipment, current acquisition and assignment policies and procedures are
( 1) This amount i s f o r comnunications equipment most l i k e l y t o be assigned t o individual
employees. I f large or unique i terns were included, the t o t a l annual figure would be
over $ 1 m i l l i a n .
inadequate, and fai l to address the assignment and use of al l types of
equipment, both new and used. This lack of adequate policies also allows
the responsibility for selecting and purchasing equipment to be assumed
by various personnel. Finally, DPS does not identify whether equipment
is really needed, and j u s t i f i c a t i o n for equipment assignments is not
always evident.
Policies reaardina telecomnunications eauiprnent assiament and use are
inadequate - Present polices do not address the assignment and
u t i l i z a t i o n of equipment. The department divides policies into two
categories--" General Orders," directives a l l bureaus must follow, and
" Bureau Orders," directives developed by one of the five bureaus that
apply only to those employees in the specific bureau that promulgates the
pol icy.
Some General Orders discuss communications equipment; however, there are
no department or bureau policies that address the newer equipment the
department u t i l i z e s ( i. e., cellular phones, facsimile machines, answering
machines, or pagers). Additionally, there is no department or individual
bureau policy that specifies which types of equipment w i l l be issued to
personnel within each bureau or at various organizational levels within
the department.
For example, only three orders address communications equipment. One
identifies required and optional equipment for patrol vehicles; however,
the l i s t does not mention any communications equipment that might be in a
vehicle, ( e. g., mobile radio or cellular phone). In addition, there are
two policies on the installation and use of police radio scanners and
citizen band radios. NeZher specifies which officers should be assigned
this equipment. The absence of policies addressing these issues can
result in inequities in the assignment of equipment.
The lack of policies promotes inconsistency - Due to a lack of policies,
various personnel select and purchase equipment. Although the
Telecommunications Bureau is responsible for developing, operating, and
maintaining the department's communications systems, the bureau is not
solely responsible for the purchase of telecommunications equipment, and
any DPS bureau with the funds can purchase communications equipment. At
least 28 percent of the 1,061 units of communications equipment purchased
in the past three years was purchased by other bureaus in the department.
The Highway Patrol Bureau's use of scanners is one example of
inconsistent acquisition and u t i l i z a t i o n . A survey of Highway Patrol
d i s t r i c t offices indicated that the use of scanners varied from a single
unit in a d i s t r i c t office to several units in each d i s t r i c t ( one scanner
in each patrol vehicle). The Assistant Director of Highway Patrol told
us that the use of scanners is discretionary, and the bureau has no
universal policy governing their use. He also noted that the d i s t r i c t s
have the funds to purchase additional equipment.
We surveyed three neighboring states, Utah, New Mexico, and California,
to determine whether they had a centralized policy within their public
safety departments for purchasing communications equipment. In a l l three
states, the equivalent of Arizona's Telecommunications Bureau purchased
a l l equipment for a l l departments, rather than the individual departments
purchasing equipment for their operations.
Needs assessments are limited - In addition to inadequate equipment
policies and procedures for purchasing equipment, DPS does not conduct
comprehensive needs assessment before purchasing new equipment. When we
attempted to obtain copies of needs assessments for the communications
equipment recently acquired by the department ( including, but not limited
to, mobi le display terminals, cel lular phones, portable radios, pagers,
facsimile machines, and answering machines), we were able to locate only
two written assessments-- one for facsimile machines and the other for
cel lular phones.(')
( 1) Several DPS managers indicated to us that assessments had been performed, but had not
been documented. Also, the assessment for mobile display terminals written i n the
early 1980s had been sent to the archives and, due to retention schedules, has since
been destroyed.
A survey of bureau s t a f f was used in both of these assessments to
determine whether the equipment was needed, and for what types of
a c t i v i t i e s .
Therefore, of the assessments we reviewed, no c r i t e r i a had been
established against which to evaluate equipment implementation or perform
cost analysis. Estimates of implementation costs should include not only
the i n i t i a l purchase price, but also estimated maintenance and equipment
replacement costs.
J u s t i f i c a t i o n f o r certain eaui~ ment assianments i s lacking - We found no
written j u s t i f i c a t i o n s within needs assessments, p o l i c i e s , o r other
administrative documentation to indicate the reason certain types of
equipment are assigned to particular s t a f f . For example, each assistant
director and a l l command s t a f f above the rank of lieutenant are assigned
a c e l l u l a r phone, in addition to a mobile radio, portable radio, and a
pager .
The Department's Information Analysis Section conducted a limited study
to develop c r i t e r i a for the evaluation and approval of requests to
purchase mobile telephones. The c r i t e r i a given highest p r i o r i t y in the
report was " Is there a need which cannot be met by existing means of
communications ( portable or car radios, stationary telephone o r d i r e c t
contact) and which can only be met by using portable telephones?"
However, we found no documentation to v e r i f y that t h i s c r i t e r i a had been
considered by the department in assigning this equipment.
Department Strenqthens
Acquisition Policv
In July 1990, an executive s t a f f committee tentatively approved revisions
to an existing General Order that addresses the procurement of goods and
services. These revisions require the Assistant Director of the
Telecommunications Bureau to endorse a l l communications equipment
requisitions p r i to review and approval by the purchaser's chain of
command. The d- ~ rtment i s currently f i n a l i z i n g these revisions, and
hopes to d i s t r i b u t e them to the individual bureaus in the very near
future. However, in order to prevent unnecessary or excessively costly
equ i pmen t purchases , DPS shou I d a I so cons i de r conduct i ng needs
assessments before pu rchas i ng new equ i pmen t .
RECOMMENDATIONS
1. The department should establish written c r i t e r i a upon which to base
equipment assignments. The c r i t e r i a should include appropriate
j u s t i f i c a t i o n for the assignment of each type of equipment to a
particular position.
2. The department should conduct more comprehensive needs assessments
before acquiring equipment. These assessments should include
measurable c r i t e r i a against which to evaluate equipment usage.
Assessments should also be written to allow for review and
documentation.
3. The department should ensure telecomrnunications equipment purchases
and assignments conform to the c r i t e r i a established by the department.
FINDING V
THE ACJlS DIVISION SHOULD IMPROVE
ITS CASH- HANDLING PROCEDURES
The Arizona Criminal Just ice lnformat ion System ( ACJIS) Division should
improve i t s cash- handling procedures. Current procedures present the
potential for cash loss. Many measures, i f implemented, would reduce
this potential .
Currently, the ACJlS Division charges fees for several services. For
example, during fiscal year 1989- 90, approximately 954 parties paid $ 6 to
obtain copies of accident reports for commercial purposes. As DPS begins
charging fees for additional services, suck as providing accident report
copies for noncommercial purposes or conducting background checks, the
ACJIS Division w i l l receive as much as $ 700,000 in additional cash.
Thus, the need for adequate cash- handling controls w i l l only become more
important in the future.
Current procedures resent ~ o t e n t i a l for cash loss - The ACJIS Division's
current cash- handling procedures present the potential for cash loss.
Currently, when cash or a check i s received for a service, a clerk writes
a receipt, stamps the check " for deposit only to the Arizona State
Treasury," and places the cash or check in an envelope. The envelope i s
kept in a drawer which i s l e f t unlocked during the day and locked at
night. The key to the drawer i s kept on a supervisor's desk.
There are several deficiencies in these procedures resulting in the
potential f o r cash loss. For example, many ACJIS Division s t a f f work the
evening and night s h i f t s , and thus could access the envelope where cash
and checks are kept. Also, no d i v i s i o n s t a f f reconcile the receipts
against the cash and check intake. These deficiencies could allow
undetected cash loss to occur. For example, a clerk could receive a
check for a service rendered, not issue a receipt, and take cash equal to
the amount of the check. Even i f a receipt were issued, the clerk could
take the same amount of cash, and, given the lack of reconciliation, the
shortage would not be detected.
Many measures would reduce the ~ o s s i bIii t v of cash loss - The
implementation of several measures would decrease the potential for cash
loss. F i r s t , although current revenues do not warrant the need, i f DPS
begins charging additional fees the ACJlS Division should obtain a cash
register. With a cash register, DPS could both issue receipts and have a
master tape of transactions that a supervisor could easily review.
Second, next to the cash register DPS should place a sign stating that
customers are to be issued receipts and requesting customers to ask for
receipts i f they do not receive them. Finally, DPS should require the
form of payment, cash or check, to be indicated on receipts, and a
supervisor should reconcile daily both the cash and the check intake
against the receipts.
RECOMMENDATIONS
1. I f DPS revenues increase due to new fees, it should purchase a cash
register to issue receipts and provide a master tape of transactions.
2. BPS should place a sign next to the cash register requesting
customers to ask for receipts i f they do not receive them.
3. DPS should require the form of payment, cash or check, to be
indicated on receipts, and a supervisor should reconcile the cash and
checks received against the receipts.
FINDING VI
MOST USERS FEEL THAT THE EMSCOM SYSTEM
ADEQUATELY MEETS THEIR NEEDS,
HOWEVER. SOME PROBLEMS WERE NOTED
Although most users of the Emergency Medical Services Comunications
( EMSCOM) system are generally satisfied, some system problems do exist.
A survey of EMSCOM users revealed that the system does meet most of their
needs. Most concerns raised by users address problems that DPS has
limited a b i l i t y to correct. However, one problem noted is correctable
and appears to stem from inadequate dispatcher training. In addition,
when problems occur there is no centralized means for ensuring that these
concerns are forwarded to the appropriate EMSCOM personnel for resolution.
Arizona's EMSCOM system provides a radio communication link between f i e l d
Emergency Medical Services ( EMS) personnel and medical f a c i l i t i e s
throughout the State. The EMSCOM system allows ambulance, f i r e
department, and other rescue vehicle personnel to talk with physicians at
hospitals regarding patient care. EMSCOM users access the system using a
radio mounted in their vehicle. They c a l l i n and speak with an EMSCOM
dispatcher located in DPS's Phoenix Communications Center. This
dispatcher w i l l then connect the user with a specific hospital to receive
medical advice. These radio connections are called " patches."
DPS is statutorily designated as the EMSCOM system manager and works with
the Department of Health Services ( DHS) to provide an effective
communication network. DHS is statutorily required to coordinate the
establishment of a statewide EMS system, and consequently provides much
of the EMSCOM system funding. DHS also provides radio equipment to users.
Most "-. ers of EMSCOM
Are .. rerally Satisfied
A l~ rnited survey of EMSCOM users conducted by Auditor General staff
revealed that most appear to be satisfied with the system. We conducted
a telephone survey of 45 different EMSCOM users. Fifteen of the 78
hospitals in the system were surveyed, along with 30 of the 164 f i e l d
users ( p r i mar i l y ambu lance compan i es and f ; re departments) . Respondents
within each user category were selected based on amount of system usage
as well as geographic location.
Results showed an overall satisfaction with the system. Most f i e l d and
hospital users surveyed remarked that the system was good and that EMSCOM
personnel were helpful and professional. In addit. ion, many users f e l t
that the system adequately met their needs. Further, users were
generally satisfied with the quality of EMSCOM radio transmission.
Some Problems Noted
Cannot Be Addressed bv DPS
Despite a high level of satisfaction with EMSCOM, users s t i l l identified
some areas of concern. However, DPS has limited a b i l i t y and resources to
remedy these problems. These areas are discussed below.
Comnon calling - Many f i e l d users do not like a recently implemented
procedural change for accessing EMSCOM because i t affects their
a b i l i t y to use their portable radios. This procedure requires a l l
users to call a dispatcher on the same channel, and then switch to
another channel for comnunicat ion with the hospital . ( ' I However,
many f i e l d users have portable radios which do not have channel
switching capability. Thus, when they are away from the ambulance
t reat i ng a pat i en t , the Emergency Med i ca l Techn ic i an ( EMT) must l eave
the patient and return to the vehicle to switch channels on the
mobile radio mount~ d in the ambulance to obtain hospital
communication.
( 1) The c * ? procedure allowed the f i e l d units to comnunicate with the hospital on the
channel they a l l e d i n on. No channel switching was required.
Several f i e l d users have also complained that the common calling
procedure increases their time on EMSCOM as well as time needed to
establish hospital patches. This can be detrimental i f they are
close to the hospital and don't have sufficient time to establish a
patch before arriving at the hospital. They are also concerned that
the dispatchers do not monitor the patch as closely, and thus are not
as available to assist with transmission problems.
DPS o f f i c i a l s claim that having one common channel is an accepted and
efficient means for users to access the system, and they implemented
this change to respond to the increase and growth in EMSCOM radio
t r a f f i c . This new procedure should minimize the amount of time the
dispatcher spends establishing patches and consequently allow the
dispatcher to handle more calls.
Access delays - Despite the implementation of a common calling system
to help free up the dispatcher, users are s t i l l experiencing delays
getting on the system. Since there is only one call- in channel for
a l l users statewide, backup can occur i f more than one user calls in
at the same time. In fact, 8 of the 32 f i e l d units surveyed
responded that they cannot get through to a dispatcher 25 percent of
the time.
Obtaining quick access to the EMSCOM system is important. For
example, communication delays give the hospital less time to prepare
for the patient's arrival. Also, patient care may be compromised i f
f i e l d units have to wait for medical advice from the hospital
regarding treatment of c r i t i c a l l y i l l patients. For example, a
paramedic was transporting a patient who needed an IV. The paramedic
could not get through on EMSCOM, so he had to start the IV without
consulting a doctor.
We attempted to identify the demands on the system to validate user
concerns. Although data regarding system use is limited, i t does
provide an indication of the demands on the system.(') Analysis of
the available data indicates that system use is intermittent and
sporadic. We found that f i v e c a l l s may come in within ten minutes of
each other while at other times only one c a l l may come in within an
hour period. We also observed EMSCOM console operations during a
period of normally heavy patching a c t i v l t y and found a low level of
a c t i v i t y . The unpredictability of system use makes i t d i f f i c u l t to
determine whether thefe is a need to open another console to handle
EMS ca l Is. DPS has a second EMSCOM console that they purchased
several years ago which i f manned may ease some of the access
delays. Although we could not determine the need for an additional
console, DPS plans to request three additional dispatcher positions
to man the second EMSCOM console.
( 1) There i s no system i n place to monitor the number and times of radio transmissions.
The only data available i d e n t i f i e s the number of patches created per user; however,
there can be several radio transmissions r e l a t i n g t o each patch.
Some areas of the state lack EUSCOM coveraqe - Some users complained
that there are too many areas or " deadspots" which do not have EMSCOM
cove rage. These deadspo t s ex i st because mount a i ns and va l l eys b lock
the transmission, or the mountaintop radio towers are too distant
from the user trying to access the system.
DPS is aware of these areas of limited coverage, and is taking steps
to improve coverage where feasible. The department is adding two
radio towers to the EMSCOM system in fiscal year 1990- 91 at a cost of
approximately $ 60,000.(') In addition, an existing tower, which is
unnecessary at i t s present urban location, w i l l be transferred to a
remote location to improve coverage in that rural area. However,
these improvements w i l l place the current system at capacity due to
the limitations on the number of channels the dispatching console can
hand l e .
The only solution i s to purchase a new console with increased channel
capacity. Although additional towers could then be handled by the
console, DPS o f f i c i a l s state i t would not be feasible or cost
effective to provide 100 percent coverage for the entire State.
There are areas where the population is so small that the probability
for an EMSCOM transmission would be very small on a yearly basis, and
would not j u s t i f y the major expense of installing a radio tower.
Transmission Qualitv Problems Stemrninq
from lnsuff icient Dispatcher Training
Can Be Addressed
Although DPS has limited a b i l i t y to address the concerns noted
previously, usersf concerns regarding poor transmission quality i n
certain areas of the state can be addressed by improving dispatcher
training . Users have exper i enced poor t ransmi ss ion qua l i ty which appears
to stem from the dispatcher's lack of knowledge regarding system
capabilities. BPS is aware of the problem and is taking steps to correct
i t .
Several users have expressed concerns regarding dispatchersf i n a b i l i t i e s
to create quality patches. These concerns focused on the dispatcher
t e l l i n g the f i e l d user to use the wrong tower or channel to establish the
patch. Inadequate patching as a result of dispatcher error can cause
static or even cause the patch to break o f f . For example, there are
areas in the northern ( Flagstaff and Sedona) and southern ( Bowie and Casa
Grande) parts of the state where the dispatcher must have a good
( 1 ) DHS funded these towers through the Emergency Medical Services Revolving Fund.
46
knowledge of the area and available towers to establish a good quality
patch. EMSCOM dispatchers create many " cross patches" for users in some
of these areas because they require the use of more than one tower to
establish the patch. I f the dispatcher attempts to patch the user by
using only one tower or i f the dispatcher selects the wrong towers for
transmission, the user may experience heavy static or lose the patch
altogether. The following case examples i l l u s t r a t e some of the problems
users in these areas have experienced:
CASE 1
An ambulance user requested a dual cross patch with the destination
hospital as well as with the hospital from which the company receives
medical control. The user requested a specific channel for the cross
patch because one of the hospitals only had a particular channel.
However, the EMSCOM dispatcher told the user to u t i l ize another
channel. The base hospital could not hear the transmission on this
other channel, so the user switched to the channel originally
requested, and then the transport hospital could not hear the
transmission.
There was l i t t l e time to establish a new patch before arriving at the
transport hospital. The dispatcher was unaware that there were
problems with this patch and could not assist the users in
reestablishing i t . As a result, the patch was not completed.
CASE 2
An intermediate level emergency medical technician ( IEMT) requested a
patch with his base hospital, and the dispatcher replied that the
patch could not be made. When the dispatcher said that the patch
could not be done, the IEMT interrupted patient care to explain to
the dispatcher how the patch could be made. Once he did t h i s , the
patch was completed without further problems.
According to DPS management, these situations should not occur i f the
proper towers are used for the cross patch. Consequently, inadequate
dispatcher training appears to be the cause of these problems.
DPS is addressing these concerns by stressing dispatcher training on the
EMSCOM system. Special attention w i l l be given to cross patching
procedures. Dispatchers w i l l also be made more aware of the towers and
repeaters available in each area, so that they know which ones can be
used from certain locations to improve transmission quality. DPS
communications staff are also looking into developing a l i s t that would
show what towers would normally be needed to create patches for each user.
Lack of Centralized Control
for Monitorina User Com~ laints
The DPS Telecommunications Bureau currently lacks a formal means of
tracking and resolving complaints submitted by users regarding EMSCOM.
In many instances, resolution of these complaints is delayed because they
are i n i t i a l ly sent to either DHS or the State EMS Counci I. We identified
four EMSCOM complaints that were i n i t i a l l y 3nt to a member of the State
EMS Council before being forwarded to DPS. SS needs to coordinate w : h
both of these entities as well as with users to ensure that a l l
EMSCOM complaints are sent to DPS.
Once these complaints are forwarded to DPS, there is no procedure to
ensure that they are tracked and resolved in a timely manner. For
example, a complaint received by DPS in March 1990 was not formally
responded to u n t i l four months later. The Assistant Director of the
Telecommunications Bureau is in the process of formulating a new policy
that would ensure timely resolution of these complaints. The policy
would require that these complaints, when expressed in ting, i n i t i a l l y
be routed through the Assistant Director's o f f ice. In s way, he w i l l
be aware of the complaints and w i l l also be able to t;, their progress
toward resolution to ensure i t is timely.
RECOMMENDATIONS
1. DPS should continue to improve dispatcher training regarding EMSCOM
operations.
2. DPS should develop a formal procedure for the handling and
disposition of written complaints submitted by EMSCOM users.
OTHER PERTINENT INFORMATION
During the audit we obtained information on state- of- the- art technologies
which DPS plans to implement including an automated fingerprinting
identification system, mobile d i g i t a l terminals, a computer assisted
dispatch system and a 800 MHz radio trunking system for the Phoenix
Metropolitan area. In addition we obtained information on the cost of
upgrading the Arizona Criminal Justice Information System.
Automated Finaerprint Identification Svstem ( AFIS]
Senate B i l l 1001, passed by the Legislature and approved by the Governor
during the spring of 1990, allowed for the implementation of an Automated
Fingerprint ldent i f icat ion System ( AFIS).
The use of AFlS w i l l allow DPS staff to analyze crime scene fingerprints
and generate l i s t s of suspects for crimes for which there are no
suspects. Generally, DPS staff do not perform searches to determine i f
latent prints match any of the 830,000 offender fingerprints on f i le.( l)
However, i f law enforcement o f f i c i a l s have 3 to 4 parties suspected of
committing a crime, they can compare the suspects' prints with the
fingerprints obtained at the crime scene. Otherwise, DPS f i l e s but does
not analyze the latent prints. When DPS has attempted to generate l i s t s
of suspects by matching latent prints with offender prints, the e f f o r t
has been inefficient and time consuming. For example, DPS staff served
on a task force formed to solve the Tucson prime time rapist case. One
DPS latent print examiner reviewed 500,000 offender fingerprints over an
8- month period before identifying the rapist.
AFlS w i l l allow DPS staff to identify suspects in a more efficient
manner. AFlS digitizes and stores both latent and offender
fingerprints. When a latent print is run against the offender data
base, AFlS w i l l provide a candidate l i s t of possible matches. Final
( 1) A l a t e n t p r i n t i s a f i n g e r p r i n t obtained a t the scene of a crime. Latent p r i n t s
generally do not consist of clear p r i n t s of 8 fingers and 2 thumbs and, consequently,
cannot be used to i d e n t i f y suspects.
verification of the match w i l l be made by a fingerprint examiner.
According to the supervising latent print examiner, i f AFlS had been
operational during the prime time rapist investigation, the
identification would have been made sooner and some of the crimes could
have been prevented. DPS plans to have AFlS operational in May 1992.
As an alternative to purchasing i t s own AFIS system, DPS could have
joined the Western Identification Network ( WIN). WIN is an AFlS system
which is being operated j o i n t l y by some of the western states. ACJIS
Division staff indicated that they had prepared a cost benefit comparison
of joining WIN versus purchasing an independent AFlS system. The
comparison revealed that purchasing an independent AFlS system was a
better solution for Arizona because DPS can have management control and
i n s t a l l more remote stations for almost the same cost as joining WIN.
Searching against the FBI fingerprint f i l e is not an alternative because
the f i l e is not automated.
Senate Bi l l 1001 created an AFlS fund and transferred $ 2 mi l lion from the
general fund to the AFlS fund for the lease- purchase of equipment and
services. DPS w i l l seek additional annual appropriations over the next
four years to fund the remainder of the lease- purchase agreement. The
f i r s t year of the AFlS implementation w i l l cost DPS approximately
$ 2,100,000. Annual costs for the second through f i f t h years are expected
to be approximately $ 2,865,000.(') Thus, the total cost for the f i r s t
five years w i l I be approximately $ 13.56 mi l l ion. In order t~ generate
revenues to fund the operation and maintenance costs of AFIS, the b i l l
raised the rate of assessments on fines from 37 to 40 percent and
modified the distribution of Arizona Criminal Justice Enhancement Fund
( ACJEF) monies. As of October 31, 1990, 7.5 percent of the total monies
collected for ACJEF w i l l be deposited in the AFlS fund. DPS anticipates
collecting $ 1,195,000 annually from this distribution, based on fiscal
year 1988- 89 ACJEF revenues.
( 1 ) The annual $ 2,865,000 cost i s comprised of $ 1,935,000 f o r hardware ( including
conversion, software, and s i t e preparation), $ 15,000 f o r communications 1 ines,
$ 150,000 f o r other operating costs, and $ 765,000 f o r maintenance. The maintenance
component i s not included i n the f i r s t year's cost.
Many other states and c i t i e s have acquired AFlS systems and reported
increases in the number of crimes solved.(')
California - According to the system manager, 30,000 crime scene
identifications have been made since AFlS became operational in
1979. For example, 75 of 400 unsolved murders have been solved.
San Francisco - S t a t i s t i c s for the f i r s t 4 years during which AFlS
was operational ( 1984- 1988) reveal the following numbers of crimes
solved: over 2,000 burglaries, 98 homicides, 200 robberies, 224 auto
thefts, 39 rapes, 32 narcotics offenses, 29 h i t and run incidents,
and 43 assaults. According to a San Francisco Police Department
crime scene investigations inspector, the increased number of
burglaries solved resulted in a 28.6 percent decrease in the number
of burglaries in the c i t y .
St. Louis - Since AFlS was installed in 1988, the AFlS manager
indicated that there have been 760 crimes solved that would otherwise
not have been solved.
In addition, during our audit, the Phoenix Police Department was
negotiating with a vendor in order to buy an AFlS system. One Phoenix
Police Department s t a f f member noted that Phoenix needs i t s own AFlS
system because the department maintains data on certain misdemeanors
which DPS does not maintain. Also, the Phoenix Police Department has
such a large volume of p r i n t s t o process that i t would monopolize the DPS
AFlS system i f i t did not have i t s own system, thus preventing other law
enforcement agencies in the state from using the DPS'system.
DPS anticipates the need for additional latent p r i n t examiners as a
result of the i n s t a l l a t i o n of AFIS.(~) Additional examiners w i l l be
needed to address both the anticipated workload increase ( DPS expects the
volume of latent cases to double after AFlS becomes operational) as well
as previously f i l e d latent p r i n t s . C a l i f o r n i a and San Francisco
( 1 ) According to the Assistant Director of the Telecomnunications Bureau, 26 other states
have AFIS systems. The FBI does not have an AFIS system.
( 2) DPS could not provide a range of the number of positions which might be needed, but
c e r t i f i e d l a t e n t p r i n t examiners w i l l be needed to f i l l the positions. C e r t i f i e d
l a t e n t p r i n t examiners obtain fingerprints a t crime scenes, t e s t i f y i n court, and
perform contpl ex fingerprint comparisons.
increased their latent print staff as a result of implementing their AFlS
systems, and Phoenix anticipates a similar staff increase when i t s AFlS
system becomes operational.
Mobile Digital Terminals
Mobi le d i g i t a l terminals ( MDTs) are portable computer terminals that are
mounted in vehicles and enable Highway Patrol officers to access computer
data bases to run warrant checks on people and vehicles without having to
communicate with the dispatcher. Presently, the officers transmit these
requests to the dispatchers, who runs these checks for them. This slows
the response time to the officers, and congests voice radio channels. The
implementation of MDTs and their interface with the computer assisted
dispatch ( CAD) system should enable DPS to obtain significant savings of
time and a i r t r a f f i c .
Once interfaced with the CAD system, the MDTs w i l l allow units to check
in and out on cal Is for service without voice transmissions. Units w i l l
be able to indicate their arrival on the scene through a single key
stroke, and can check back in and update CAD records with c a l l status
information and other pertinent data without broadcasting this
information to a l l listeners. MDTs w i l l enable f i e l d units to maintain
their status with the dispatcher and dispatch center without voice
contact.
Implementation schedule - The timetable for having an MDP system
operational in Phoenix and Tucson is f a l l 1990. DPS has ordered 119 MDTs,
along with the equipment to build up mobile d i g i t a l repeater sites in
both Maricopa and Pima county for a cost of $ 796,281.(') These MDTs w i l l
be distributed along geographic boundaries associated with major
metropolitan areas, specifically Phoenix and Tucson. It is intended that
30 MDTs be installed in Highway Patrol vehicles in Tucson, and 70 in
( 1) Criminal Justice Enhancement Funds ( CJEF) and Racketeer Influenced Corrupt
Organizations ( RICO) funds were used t o implement the MOT system.
vehicles in the Phoenix area. The remaining 19 w i l l be installed in
Criminal Investigations specialty vehicles.
There is no funding available for expanding this system during fiscal
year 1990- 91: the number of terminals w i l l remain at 119. Future
expansion is dependent upon available funding. The ultimate goal is that
a l l appropriate vehicles in the Phoenix, Tucson, and Flagstaff areas w i l l
eventually receive these terminals for an approximate total of 250
terminals.
Dis~ atcher workload - DPS anticipates that the MDTs w i l l cause dispatcher
workload to remain constant, but have a different focus. Officers w i l l no
longer be asking dispatchers to run records checks for them. However,
since the officers w i l l have the a b i l i t y to do this themselves, they
should come across more " hits" ( instances where a drivers license is
suspended or revoked, or a vehicle is listed as stolen), resulting in the
dispatchers having to c a l l more wreckers, issue more department report
numbers, and handle other tasks associated with processing a h i t . Since
the amount of radio t r a f f i c in the Phoenix and Tucson areas is near the
saturation point, i t is DPS's intention that MDTs w i l l allow dispatchers
to maintain the same level of service for the next two to three years.
Com~ uter Assisted Dis~ atch System
The Operational Communications Division is presently implementing a
Computer Assisted Dispatch ( CAB) system to improve the efficiency of i t s
Highway Patrol dispatch function. CAD is an automated system which
assists the dispatcher in the i n i t i a l receipt of information, the
assigning ~ f appropriate resources, and the tracking of desired
information such as response times, type of event, location, and call
assignment.
CAD w i l l eliminate many currently manual tasks and in the process improve
department record keeping as well as response times to calls for
service. For example, CAD w i l l eliminate much of the manual records and
procedures used by the dispatchers such as rolodexes, telephone books,
wrecker rotation l i s t s , incident cards, locator/ status sheets, and radio
hand logs. In addition, CAD w i l l help the dispatcher identify the
appropriate resources to respond to calls. For example, when a f i e l d
unit calls in to dispatch, a dispatcher w i l l enter the location using
highway number and mile marker, and CAD w i l l display a l l the possible
assisting agencies for that service area on the screen. Agencies
identified include the appropriate f i r e department, police department,
tow trucks, ambulance company, and hospital in that area. The
dispatchers currently retrieve this information from maps, phone l i s t s ,
and other source documents. This can be time consuming. Furthermore,
the system w i l l have the a b i l i t y to generate management reports which
should be useful in assessing dispatches workload, unit a c t i v i t y , types
of events, locations of events, response times, and other a c t i v i t i e s .
CAD w i l l also improve tracking of officer status and may increase officer
safety. Unit status w i l l be displayed by committed units and available
units. In addition, pending calls w i l l be displayed on the CAD system
according to six p r i o r i t y classifications, with p r i o r i t y one calls
( emergency calls) in red on the screen. Furthermore, there w i l l be an
alarm bui It into the system to alert the dispatcher that an officer is
overdue to cal I- in, Timers w i l l cause the o f f i c e r ' s status line to turn
to red from white * hen the specified time has elapsed.
The CAD system w i : st $ 217,000, a l l of which w i l l come from state
appropriations. Ha . re has been installed in Phoenix and Tucson
Communications Centers, and CAD is expected to be f u l l y operational by
September 199Q. The Flagstaff Communications Center was originally
scheduled for implementation in fiscal year 1990- 91: however, funding was
not approved. This request w i l l be resubmitted for fiscal year 1991- 92.
Phoenix Metro~ olitan Area
800 MHz Radio Trunkina System
Due to the increased use of existing radio channels, DPS plans to
implement an 800 MHz radio trunking system. The department currently
uses a 460 MHz lfl tra High Frequency ( UHF), two- way mobi le radio system
for i t s voice j i o communications that allocates one channel to each
user.(') According to DPS, voice radio channels in the Phoenix
metropolitan area are already extremely congested during peak hours, and
waiting time to broadcast over the system is considered excessive for a
law enforcement agency ( i . e., greater than a ten- second wai t 1. W i th the
projected expansion of the freeway system in metropolitan Phoenix, the
demand for radio communications w i l l be increased. Additionally, there
are no UHF radio frequencies available in the metropolitan area that are
compatible with DPS radios. Anticipating the need for additional radio
channels, DPS has obtained licenses from the FCC for eight 800 MHz radio
channels to al low for lltrunkedll radio use.
An 800 MHz radio trunking system allows one radio channel to act as a
" t r a f f i c cop" to direct a mobile or portable radio to any vacant voice
channel on the trunked radio system. A computer keeps track of the
assignments of radio users to prevent one group of users from talking to
another group or accidental ly being overheard. Computerized cal l
direction allows approximately twice as many units to be placed on a
trunked system with a given number of channels as compared to a
conventional channel radio system. A three- year departmental plan
proposes implementing an 800 MHz trunked radio system for the Phoenix
metropol itan area at a cost of $ 610,000. In fiscal year 1991- 92,
$ 130,000 w i l l be required for equipment for the White Tanks Mountaintop
Center ( i n the western Ph~ enix metropolitan area), as we1 1 as for a small
number of radios to use the system. In fiscal year 1992- 93, an
add i t i ona l $ 480,000 w i l l be needed for equ i pmen t at the Thompson Peak
Mountaintop Center ( to cover the eastern and northeastern Phoenix
metropolitan area), for additional computer controls to make the system
more effective, and for mobile and portable radios.
ACJIS Com~ uter U~ arade
DPS maintains the State central repository for criminal history record
information. Federal, State and local law enforcement agencies u t i l i z e
( 1) The department also provides radio comnunications services f o r several other State
agencies including the Department of Transportation, Game and Fish, State Land, State
Parks, Liquor Licenses and Controls , Emergency Services, Corrections, Agricul ture and
Horticulture, and the Capi to1 Pol ice.
the Arizona Criminal Justice Information System ( ACJIS) to identify,
investigate, and apprehend criminals. This information is available to
local agencies at no charge.(') Due to expanded usage of the system and
to comply with National Cr ime lnformat ion Center ( NCIC) user response
times, DPS has requested additional funding to upgrade the current
computer system housing the criminal h i s t ~ r y information. The cost to
upgrade the system ranges from $ I mi IIion to $ 3 million depending on
whet her the department purchases a used or new computer .
Due to the cost of the expansion, we surveyed s i x other states to
identify how these states fund the hardware costs associated with their
criminal justice information system and specifically whether users of the
system shared the cost of the system.(" We found that each of the six
states require some type of cost sharing from the users although only two
of the states require the users to help support the cost of the
mainframe. For the most part, users are responsible for the cost of
terminals, printers, and telephone lines used to access the system.
However, Michigan and Alabama both charge their users a monthly fee to
help fund the system. Alabama charges a $ 450 monthly user fee while
Michigan is charging users $ 100 per month over the l i f e of a five- year
lease/ purchase agreement to upgrade their system.
( 1) The DPS computer system was purchased with state funds. I n addition. $ 452,300 i n
general funds are appropnated to the ACJIS l i n e fund each year. This l i n e fund
provides the user agencies with one terminal, one p r i n t e r , the cost to hook up t o the
system, and pays the monthly telephone b i l l s f o r accessing the system. Users pay f o r
any additional terminals and the month1 y telephone costs associated with the
addi ti onal terminal s .
( 2) DPS1s request of $ 301,000 i n t h e i r f i s c a l year 1990- 91 budget to upgrade the computer
system under a mu1 ti year installment purchase approach was not approved. Therefore,
on July 12, 1990, DPS requested $ 200,000 i n Criminal Justice Enhancement Funds to
upgrade the system by December 1990 i n order to maintain satisfactory user response
time.
( 3) Arkansas, New York, California, Alabama, and Nevada were recomnended by the Depllty
Chief of the National Crime Information Center as states with model criminal j u s t i c e
information programs. Michigan was surveyed as we had learned that i t was charging
users and we wanted t o obtain additional information regardsng i t s charging.
ARIZONA DEPARTMENT OF PUBLIC SAFETY
2102 WEST ENCANTO BLVD. P. O. BOX 6638 PHOENIX, ARIZONA 85005- 6638 ( 602) 223- 2000
ROSE MOFFORD F. J. " RICK" AYARS
GOVERNOR DIRECTOR
November 1 6 , 1990
Mr. Douglas R . Norton
A u d i t o r General
2700 North C e n t r a l Avenue, S u i t e 700
P h o e n i x , AZ 85004
Dear Mr. Norton:
Enclosed i s t h e r e s p o n s e from t h e A r i z o n a Department of
P u b l i c S a f e t y t o t h e r e v i s e d d r a f t r e p o r t of p e r f o r -
mance a u d i t of t h e T e l e c o m m u n i c a t i o n s Bureau.
S i n c e r e l y ,
November 13, 1990
ARIZONA DEPARTMENT OF PUBLIC SAFETY'S
RESPONSE TO
REVISED DRAFT REPORT OF PERFORMANCE AUDIT
OF THE TELECOMMUNICATIONS BUREAU
FINDING I
DPS COULD GENERATE MORE THAN $ 700,000 ANNUALLY BY ESTABLISHING
A FEE FOR CERTAIN SERVICES AND UPDATING CURRENT FEE SCHEDULES
As noted in the Audit Report, the Department of Public Safety has
begun charging everyone requesting accident reports at a rate of
$ 1.00 per page. This is expected to generate $ 108,000 annually.
RECOMMENDATION
The legislature should consider revising A. R. S. 41- 1750 to allow DPS to
establish a $ 10 fee for conducting other background checks on applicants
for licenses and employment at the State level. Revenues not needed to
support this processing should then be deposited in the general fund.
RESPONSE
DPS agrees with this recommendation and will take steps in the next
legislative session to introduce changes to A. R. S. 41- 1750 to allow
DPS to begin charging a fee for background checks on applicants.
FINDING I1
DPS'S OPERATIONAL AUDIT PROGRAM IS DEFICIENT
RECOMMENDATIONS
1. DPS should take steps to bring the department into compliance with the
NCIC biennial audit requirement.
2. DPS should make every effort to conduct on- site ( level I1 or level
111) audits when possible.
3. DPS should continue to implement improvements to its audit program
. such as visiting remote terminal locations and auditing non- terminal
agencies.
RESPONSES
1. Although the audit report indicates DPS is not in cornpliance with
NCIC policy, the NCIC 1989 audit report states, " ACJIS
Operations Audit Program is in compliance with the NCIC policy."
Although there is a difference of opinion between the state
auditors and the NCIC auditors regarding the degree of this
compliance, it should be noted this is a federal NCIC audit
requirement and if the state of Arizona is meeting the
requirements to the satisfaction of NCIC ( federal government)
note of this should be made.
DPS concurs with the recommendation of the audit team that
every effort should be made to conduct level I1 or level I11
audits when possible. However, limitations on human resources
oftentimes prevent this from being feasible. It should be noted,
the audit program has never received any authorized FTEs
through the appropriations process. When these programs were
mandated by NCIC it became necessary for DPS to utilize existing
resources in an effort to comply with these federal
requirements. Due to legislative cuts in authorized positions in
FY 1982183 the total process of auditing criminal history records
and maintaining the DPS1s records system is being done with 3
less FTEs today then it was prior to FY 1982183. During this
same time, arrest cards received have gone from approximately
50,000 annually to over 111,000 annually. Applicant fingerprints
have increased from 37,000 annually to over 70,000 annually, and
dispositions of cases has increased from 60,000 to 113,000
annually.
DPS agrees that possibly all programs could use some level of
improvement, however, considering there has been a decrease in
available human resources over the past 10 years while at the
same time, existing programs have increased by as much as 60%
and many new programs have been implemented which had to be
complied with with existing financial and human resources.
3. It is agreed that improvements to audit programs could be made
by more visits to terminal locations and every attempt will be
made to make more personal visits to terminal locations,
however, the auditing of nonterminal locations is not feasible
with the existing amount of FTEs.
FINDING I11
DPS NEEDS TO IMPROVE THE COMPLETENESS AND RELIABILITY OF ITS
CRIMINAL HISTORY INFORMATION
RECOMMENDATIONS
1. The ACJIS Division should obtain input from local law enforcement
agencies and identify an acceptable time limit for submission of arrest
fingerprint cards. DPS should petition the Legislature to modify ARS
41- 1 750 accordingly.
2. ACJIS Division management should more closely monitor turnaround
times, backlogs, and staffing levels for entering arrest data.
3. The ACJIS Division should improve its procedures for processing
problematic arrest cards and disposition forms. The following should
be incorporated into the revised procedures :
a. The error data base should be used to identify outstanding arrest
cards and disposition forms which have been rejected.
b. ACJIS staff should call agencies to obtain needed information
rather than return the documents to the submitting agencies.
ACJIS Division management should consider dedicating staff to the
correction of problematic documents.
c. DPS should consider entering arrest data and flagging arrest
cards having illegible fingerprints.
d. In cases where a disposition must be rejected because there is
not a corresponding arrest record, the ACJIS Division should
consider obtaining arrest data itself rather than returning
disposition forms to agencies. DPS should also consider
implementing a unique tracking number system to facilitate the
matching of arrests and dispositions.
4. DPS should implement a system to identify older arrest records which
lack dispositions. Once identified, DPS should take steps to obtain
and enter missing disposition data.
5 . The ACJIS Division should implement procedures to verify at least a
statistical sample of arrest and disposition records input. Double
entry should be considered as a verification mechanism.
6. DPS should ensure that a comprehensive data quality audit, planned
for FY 1991- 92 occurs. The agency performing the audit must be
granted access to ACCH data in order for the audit to be successful.
RESPONSE
The Department of Public Safety concurs with these
recommendations and steps will be taken to implement some of the
suggestions with existing hurnan resources. However, programs
such as calling agencies regarding rejected dispositions, rejected
fingerprint cards, and etc., in addition to double entry
verification of disposition records are all programs which would
be exceptionally difficult to implement with existing resources.
FINDING IV
DPS NEEDS TO STRENGTHEN POLICIES AND PROCEDURES OVER THE
ACQUISITION AND ASSIGNMENT OF TELECOMMUNICATIONS EQUIPMENT
RECOMMENDATIONS :
1. The department should establish written criteria upon which to base
equipment assignments. The criteria should include appropriate
justification for the assignment of each type of equipment to a
particular position.
2. The department should conduct more comprehensive needs assessments
before acquiring equipment. These assessments should include
measurable criteria against which to evaluate equipment usage.
Assessments should also be written to allow for review and
documentation.
3 . The department should ensure telecommunications equipment purchase
and assignments conform to the criteria established by the department.
RESPONSES
1. DPS concurs with this recoinmendation and is currently in the
process of drafting a general order which will identify the
telecommunications equipment required for each critical position
within the department.
2. DPS also agrees with this recommendation and steps are being
taken to establish procedures to ensure assessments of major
procurements are made prior to making major telecommunications
purchases.
3 . Increased emphasis will also be placed on existing policies and
procedures to ensure they are being adhered to as it relates to
the procurement of telecommunications equipment.
FINDING V
THE ACJIS DIVISION SHOULD IMPROVE ITS CASH- HANDLING PROCEDURES
RECOMMENDATIONS
1. If DPS revenues increase due to new fees, it should purchase a cash
register to issue receipts and provide a master tape of transactions.
2 . DPS should place a sign next to the cash register requesting
customers to ask for receipts if they do not receive them.
3. DPS should require the form of payment, cash or check, to be
indicated on receipts, and a supervisor should reconcile the cash and
checks received against the receipts.
RESPONSE
The Department of Public Safety agrees with all three recommendations
and is in the process of implementing all three recommendations. By
January 1991, the cash register, a receipt log and procedures for
handling payments received on a daily basis will all be implemented.
FINDING VI
MOST USERS FEEL THAT THE EMSCOM SYSTEM ADEQUATELY MEETS THEIR
NEEDS; HOWEVER, SOME PROBLEMS WERE NOTED
RECOMMENDATIONS :
1. DPS should continue to improve dispatcher training regarding EMSCObl
operations.
2. DPS should develop a formal procedure for the handling and
disposition of written complaints submitted by EMSCOM users.
RESPONSES
1. DPS is presently implementing a revised advanced EMSCORl
dispatcher training program which should greatly enhance the
ability of DPS EMSCOM dispatchers.
2. A formal procedure has been developed within the
Telecommunications Bureau to handle not only EMSCOM written
complaints but all complaints received by the Telecommunications
Bureau. This procedure is in addition to established general
orders and will supplement the complaints and discipline
procedures manual utilized by DPS.
SUMMARY
In summary, there are a number of good suggestions in this audit report
from which DPS can make a number of improvements to provide better
service to both the criminal justice community and the citizens of the
State of Arizona. As noted in the responses, those programs which can
be implemented without any additional human or financial resources are
already being undertaken by DPS. Those changes which require
legislative changes, or additional funding, will be addressed in the coming
fiscal years.
We would like to thank the auditors from the Auditor General's Office for
their cooperation and their professionalism in attempting to minimize the
day- to- day operational impact to the Telecommunications Bureau while this
audit was being performed.