State and Federal

Corrections Information Systems


An Inventory of Data Elements and an Assessment of Reporting
Capabilities


A joint project:
Association of State Correctional Administrators
Corrections Program Office, OJP
Bureau of Justice Statistics
National Institute of Justice



August 1998, NCJ 170016

Acknowledgments

This report was prepared by the Urban Institute under the
supervision of Allen Beck, Ph.D., of the Bureau of Justice
Statistics.  Laura Maruschak was the project monitor at BJS.

The project is sponsored by the following:

Corrections Program Office, Larry Meachum, Director

Bureau of Justice Statistics, Jan Chaiken, Ph.D., Director

National Institute of Justice, Jeremy Travis, Director



Project assistance was provided by the Association of State
Correctional Administrators (ASCA).

The project was supported by BJS grant number 97-MU-MU-K007. 
Principal staff for the project at the Urban Institute were
William J. Sabol, Ph.D., Barbara Parthasarathy, Katherine
Rosich, Mary Spence, and Mark Braza.  Charles Friel, Ph.D.,
served as consultant to the project.  Dave Williams and O. Jay
Arwood designed and produced the report.  Tom Hester of BJS
provided editorial review.  Marilyn Marbrook of BJS administered
final publication.

The project staff acknowledges the cooperation and support of
the ASCA State-Federal Committee, especially the contributions
they made at the Committee meeting of June 16, 1998, in St.
Louis, Missouri.  Committee members and other meeting
participants, including Chairman Joseph Lehman, Kathy Hawk
Sawyer, Robert Bayer, Mike Sullivan, Harold Clarke, Morris
Thigpen, Ari Zavaras, Richard Lanham, Elaine Little, Larry
Meachum, Phil Merkle, Camille Camp, and George Camp provided
excellent comments after reviewing the draft report.  The
project thanks the members of the project's Advisory Committee
for their assistance and thanks all of the staff of the
corrections departments who responded to our surveys and
participated in interviews.

An electronic version of this report and the data analyzed in
the report may be found on the Internet at the folllowing
address:  http://www.ojp.usdoj.gov.bjs/

The Corrections Program Office, the Bureau of Justice
Statistics, and the National Institute of Justice are components
of the Office of Justice Programs, the U.S. Department of
Justice.  The Association of State Correctional Administrators
is a nonprofit membership organization dedicated to the
improvement of correctional services and practices.  The
contents of this report do not necessarily reflect the views or
policies of these organizations

Introduction

Study objectives, methodology, data, and report organization



Objectives of the study

Methodology

Content of the questionnaires

Organization of the Inventory items

Table I. Stages and dimensions of offender-based data elements

Measures used to analyze Inventory survey responses

Rationale behind the availability indicators

Full-availability scores and ratings

Framework for defining and using the common core



Study objectives, methodology, data, and report organization

In a series of meetings, members of the State-Federal Committee
of the Association of State Correctional Administrators and
representatives from the Corrections Program Office (CPO), the
Bureau of Justice Statistics (BJS), National Institute of
Justice (NIJ), the National Institute of Corrections, and the
Federal Bureau of Prisons identified the need to assess the
current status of offender-based information systems in
corrections.  Correctional administrators expressed the need for
a set of performance indicators that could be used to describe,
measure, and compare management outcomes among departments of
corrections.  Administrators also expressed that they often lack
basic information needed to formulate new policies or to defend
existing practices.  Researchers highlighted the difficulties of
conducting comparative studies in the absence of basic agreement
on concepts and definitions, and the diversity in the quality
and coverage of data elements in these systems.

In response, CPO, BJS, and NIJ sponsored a project to conduct an
inventory and assessment of more than 200 data elements in State
and Federal corrections information systems.   An advisory
committee, including representatives of the State-Federal
Committee, other corrections officials, corrections researchers,
and representatives of the sponsoring agencies and the Urban
Institute, was formed to guide the design of the inventory and
to identify priority information areas for attention.

The Inventory of State and Federal Corrections Information
Systems is built around the six priority information areas
identified by the advisory committee:  offender profile,
internal order, program effectiveness, public safety,
recidivism, and operational costs.  The Inventory reports on the
status of information systems in adult State and Federal
departments of corrections.  Its purpose is to provide a basis
for improving the quality of corrections data, enhancing
electronic sharing of information, and improving the capacity of
corrections departments to provide comparable data for
corrections performance measures, and for cross-jurisdictional
research.  

Objectives of the study

The study has several objectives.  The first is to determine
what data elements departments of corrections collect and
maintain in their adult prisoner information systems and whether
most departments maintain data elements about a common core of
information that is roughly comparable across jurisdictions. 
The second objective is to assess the capabilities of
corrections information systems to generate and report
statistical information about offenders.  The third objective is
to organize the many data elements that departments collect into
substantive categories that describe major stages of corrections
processing and to develop a set of dimensions that measure
events in these major phases of the corrections process.  The
fourth objective is to describe the information that corrections
departments maintain about facilities management, medical care,
and costs and revenues.

Methodology

The Inventory uses surveys and interviews to ask two questions
about information systems:  What data on most adult sentenced
prisoners do departments collect and maintain in electronic
form? and To what extend can departments use these data to
respond to requests for statistical information about groups of
offenders?  These questions are asked in two structured
questionnaires about the data elements and usability of the data
systems, and a telephone interview collecting background
information from key individuals to provide an overview of the
architecture, capacities and capabilities of each corrections
information system.

Questionnaires.  The first questionnaire, the "Inventory of Data
Elements in State and Federal Corrections Information Systems,"
collects information on data elements officials maintain in the
information systems they use to manage adult, sentenced
prisoners (Appendix A).  The Inventory contains 242 questions
about data elements and capacities of information systems.  Of
these, 207 were offender-based data elements; 15 are about
facilities; and 20 are about capacities to link data.  For each
of the 207 questions about offender-based data elements, the
Inventory asks officials whether they maintain the data element
and if so, whether it is in electronic or paper form, and the
percentage of offenders for whom it is maintained.  The second
questionnaire, "Survey of Retrieval and Query Capacities of
Corrections Information Systems" (Obstacles survey) collects
information on barriers or obstacles these officials encounter
in producing statistical information in response to queries
about offenders (Appendix B).  This 25-question survey is
organized into 5 categories of obstacles or barriers related to:
 legislative and institutional matters, staffing, software,
hardware, and data. 

During January 1998, both questionnaires were mailed to
information officers in 50 State departments of corrections, the
District of Columbia and the Federal Bureau of Prisons (Appendix
C).  Urban Institute staff collected the responses to the
surveys, verified and coded them, and prepared them for data
entry and analysis.  Urban Institute staff also analyzed the
data, prepared the tables, and did the analysis reported here. 
All 52 departments responded to the Inventory and telephone
interviews, and 51 returned the Obstacles survey.

Telephone interviews.  With one to several staff members in the
corrections information system, respondents included research
staff, programmers, database administrators, system design
specialists, and other personnel involved in designing,
operating or maintaining corrections information systems
(Appendix C).  The interviews were open-ended but structured,
focusing on the following themes:  the institutional, political
and legal context of information systems, nature of technology
used, information system structure and business functions, data
collection procedures, data system linkages and data sharing,
and overview of how well systems meet current needs.  Urban
Institute staff conducted these telephone interviews during the
summer of 1997 and were responsible for transcribing them and
analyzing the content.  In general these discussions with
departmental staff provided vital background information
relevant for the design of the Inventory questionnaires, and
also supplemented data on the capabilities of corrections
information systems.

Content of the questionnaires

Both questionnaires were designed in collaboration with the
advisory committee. The advisory committee identified six
priority information areas for attention: offender profile,
recidivism, program effectiveness, internal order, public
safety, and operational costs.  In developing the Inventory
survey, the first five of the committee's priority areas were
organized into four stages of offender processing through the
corrections systems.  These stages included (1) profiling
offenders, (2) committing offenders, (3) managing offenders, and
(4) supervising offenders.  The sixth priority area of the
committee, operational costs, was used to guide development of
questions related to facility management issues.  A final
component of the Inventory survey was developed to describe the
capabilities of corrections information systems to extract and
link archived data electronically.  The links between the
committee's priority information areas and the development of
the two questionnaires are described further below.

Profiling offenders.  This priority area led to the development
of the data elements in the first two stages of corrections
processing: (1) profiling and describing offenders in
corrections, and (2) committing offenders into prison.  In the
profiling offenders information area, the advisory committee
identified a wide range of information concepts that included
offenders demographic characteristics and risk potential, as
well as their offenses, criminal histories, sentences, types of
admissions, and releases from prison.  For clarity, this
priority area was divided into one stage that described
demographic and social characteristics of offenders (profiling
offenders), and another that described both the behaviors and
decisions leading to commitment to prison and the assessment and
placement decisions made upon entry into prison (committing
offenders).

Program effectiveness and internal order.  These two information
areas were combined into the third stage of corrections
processing: (3) managing offenders in corrections facilities.
This stage includes such information concepts as program
participation by offenders, treatment, medical problems (e.g.,
HIV and TB) and medical care—as well as information related to
offender misconduct, violations of rules, safety considerations,
use of restraint, and drug and alcohol use.

Recidivism and public safety.  These two areas, in which the
advisory committee identified concepts such as the re-arrest,
re-conviction, and return to prison of released offenders, the
harm to the public, and notification of victims, were combined
into the fourth stage of corrections processing: (4) supervising
offenders on release into the community.

Operational costs.  This priority area includes
non-offender-based data, such as those that measure staffing
ratios, program effectiveness, and costs of operating
facilities.  Since these questions are about information at a
different level of analysis from the offender-based data
elements, they are organized into the fifth area of the
Inventory:  facilities management.

Many of the concepts derived from the priority information areas
relate to events that may be repeated through an offender's
career in corrections.  For example, a single offender may have
several admissions into and releases from prison; a single
offender may be involved in several hearings related to
misconduct; a single offender may have several medical tests. 
The challenge is to capture the nature of the repeats.  This may
be represented in information systems in several ways:  Each
separate record of an event may be added to an information
system as a new record.  Only information about the most recent
event may be maintained in the information system.  Records of
events may be repeated as blocks or segments to the end of an
individual's record.  Or past events may be archived on tape or
other medium, with only the current event retained in the
information system.  Regardless of the method used to record
these events, an Inventory of data elements alone cannot provide
answers to important questions about whether departments can
provide data on the number of times events happened.  To address
this concern, the Inventory included 20 questions about whether
departments maintain archival records of repeatable events and,
if so, whether they have the capability to retrieve and link
these records to current records by electronic means.

Thus, the Inventory questionnaire contains three sets of
questions.  The vast majority of the questions (207) apply to
the four stages of offender processing and ask about the
existence of offender-based data elements in corrections
information systems.  Fifteen questions ask about facilities
management, medical care, and costs.  Twenty questions ask about
the capabilities of information systems to extract and link
electronically archived data.

The Inventory project also addresses the reporting capabilities
of information systems.  Reporting capability is measured by the
extent to which the information systems can report on requests
for statistical information.  The Obstacles survey asks
departments to rate the severity of obstacles they face in
providing this information.  It asks about five areas of
potential problems: legislative and institutional, hardware,
software, staffing, and data.  

Organization of the Inventory items

In this report, the 207 offender-based data elements are
organized into 4 stages of corrections processing.  Each stage
contains several dimensions, which are relatively homogenous
groupings of data elements that together define a given stage
(table I).  The stages and categories of dimensions include:

* Stage 1, profiling and describing offenders, contains
dimensions that describe offenders' demographic characteristics,
socio-economic status, and family characteristics and living
arrangements;

* Stage 2, committing offenders, contains dimensions that
describe  offenders' commitment offenses, sentencing
information, and assessment and confinement decisions;

* Stage 3, managing offenders, contains dimensions that describe
routine offender management, methods of release from prison, and
internal order and security; and

* Stage 4, supervising offenders, contains dimensions that
describe offender behavior after release, and details about new
crimes committed and the victims of these crimes.

Measures used to analyze Inventory survey responses

The report uses several indicators to summarize responses to
questions in the Inventory survey.  The indicators provide
information about the overall completeness of the department of
corrections information systems, and point to areas where gaps
exist.  The gaps may be at the level of a data element, of the
extent of the population covered, or of the form in which
elements are maintained (electronic versus paper).  Identifying
the gaps helps identify areas where improvements to information
systems can be made—such as adding data elements, increasing the
number of data elements maintained electronically, or expanding
the coverage of populations.



The indicators of corrections information systems capacities are
based upon the following concepts:

* High availability.  This response shows that an information
system has a data element in electronic form for more than 75%
of offenders.  This high percentage indicates extensive coverage
on an element. The electronic form indicates the data
potentially can be extracted, linked, and easily shared
electronically.

* Medium availability.  This response shows that an information
system has a data element in electronic form but for less than
75% of offenders.  It indicates a medium level of availability
because the scope of coverage is less.  It also indicates that
information about a comparatively large percentage of offenders
is more likely to be missing than in the high-availability
indicator.

* Low availability.  This response indicates that a data element
is available only in paper form.  Data elements available in
low-availability form cannot be extracted, linked, and shared
electronically.  For the purposes of using offender-based data
elements to generate statistical information, low-availability
data elements present large obstacles for departments'
capacities. 

* No availability.  This response indicates that a department
does not collect a data element in any form. 

* Unknown availability.  For some elements, departments
indicated that they maintain the element but did not indicate in
which form, or the scope of coverage.

* Missing.  For some elements, departments did not answer, were
not sure if they collected it, or indicated it did not apply to
their system.

Rationale behind the availability indicators

The high- to no-availability indication implies a continuum of
availability.  The distinctions between high, medium, and low
availability reflect information system management priorities.  

High availability.  Maintaining data elements in a
high-availability form indicates that the information may be
used for on-going and day-to-day management concerns, or that
the information is used to produce regular reports about
corrections systems.  

Medium availability.  The medium-availability form indicates a
problem of scope of coverage—that the electronically maintained
data element does not apply to or is not collected on most
offenders.  There may be many reasons for this.  For example, in
decentralized systems in which data on offenders are collected
at facilities and then submitted to a central information
system, scope of coverage may relate to the absence of elements
in the local information system or to problems transmitting
data.  Partial coverage in electronic form may also reflect
changes in sentencing policy that necessitates the introduction
of new variables or data elements that apply only to specific
classes of offenders.  Or, it may reflect the retiring of old
data elements that may consequently cover smaller and smaller
percentages of offenders.  Regardless of the reasons or
explanations for it, the partial coverage of the
medium-availability format presents a problem if the objective
is to create statistical information on commonly defined
concepts. It is also a problem for which there is unlikely to be
a single solution.

Low availability.  The low-availability format indicates that
departments do not collect the data element in electronic form. 
This suggests that departments may not consider the element
among those needed for day-to-day management or for use in
regular reports.  But it does not  imply that a data element is
unimportant to the departments, or even that it is less
important than a data element maintained in high-availability
form.  For example, parole decisions may be based on information
that is maintained in a low-availability form.  Such information
is crucial for corrections management decisions, but it is not
necessarily crucial for day-to-day decisions.  

Data elements are maintained in paper form for many reasons. 
Some, such as medical records, may not lend themselves to easy
transcription and entry into computers.  Others may be highly
confidential.  Still others may be used intermittently in
decisionmaking about individual offenders.  Alternatively,
important data elements may be stored in paper form because of
information system deficiencies.  For example, some departments
need to calculate data about average length of sentence or
average time served manually, simply because the data elements
are maintained only in paper form.

No availability.  At least maintaining data elements in paper
form indicates that they do exist in a given corrections
information system.  This means that the system has developed
rules and procedures for defining, collecting, and maintaining
the data element—putting that system at a distinct advantage
over other systems that do not maintain it in any form.  In the
case of no-availability elements, departments have not even
defined the element, let alone established rules and procedures
for collecting and maintaining it.  The no-availability format
reflects the judgment that the system in question does not use
the data element for making corrections processing or management
decisions. 

Full-availability scores and ratings

The concepts of high, medium, low, and no availability are used
to create full-availability scores and ratings that measure the
extent to which departments have the availability to maintain
data elements electronically for the large majority of
offenders.  In this report, full-availability scores and ratings
are used to assess data availability in departments for groups
of data elements, although they can also be used to assess
availability for individual elements.  

A full-availability score is created by assigning a value to
each data element based on the level of availability at which a
department maintains it.  The values of the individual elements
are then summed across groups of elements to achieve a
department's availability score for each group.  The
availability indicators are scored as follows: 

* High availability elements are given a score of 3 points;

* Medium availability elements are given a score of 2 points;

* Low availability elements are given a score of 1 point; 

* Unknown availability elements are given a score of 1 point;
and 

* No availability elements are given a score of 0 points. 

For example, in a department with 10 data elements of which 4
are maintained in high availability, 2  in medium availability,
and 4 in no availability, its full-availability score equals 16,
which is derived by assigning the 4 high-availability elements 3
points each, the 2 medium-availability element 2 points each,
and the 2 no-availability elements 0 points each.  The sum of
the availability indicator scores for these 10 data elements
equals the full-availability score of 16 points.

Full-availability scores are a function of the number of data
elements in a group.  Thus, department's relative availability
across groups of data elements are difficult to assess by
reference to their full-availability scores alone.  For example,
if 1 dimension has 8 data elements and another dimension has
only 5 data elements, departments may receive a higher
full-availability score for the 8-element dimension by virtue of
the additional number of elements.  To standardize for the
differences in the number of elements among dimensions, a
full-availability index and full-availability rating are
created.  

The full-availability index is the score that a department would
receive if it maintained all of the data elements in a group in
a high-availability form.  A full-availability rating is the
percentage of the full-availability index that a department
achieves for a given group of data elements.  For example, if a
dimension of corrections processing contained 9 data elements,
the full-availability index would equal 27 (3 full-availability
points times 9 data elements).  If a department receives a
full-availability score of 20 (for any combination of high-,
medium-, low-, unknown-, and no-availability data elements) the
department's full-availability rating would be 74% (20 divided
by 27 times 100% equals 74%).

Tables of availability scores for each department are generated
for each stage of corrections processing and for the common core
of data elements (i.e., the dimensions) within each stage. 
Comparisons may be made between departments to obtain an
indication of relative availability, although not the reasons
for differences.  Comparisons also may be made between stages or
dimensions within departments to obtain an indication of the
relative availability among information areas of a given
information system.   

Included in some of the tables of availability scores are the
distributions of the number of data elements at each level of
availability for each department.  This permits some analysis of
the reasons for the overall availability scores.  For example,
some departments may receive relatively higher scores than other
departments because they maintain more data elements in a
dimension, even though they maintain a smaller proportion of the
data elements in a high-availability form.  For example,
Colorado and Iowa both rate relatively high on the availability
index (83% and 80%, respectively).  Colorado receives most of
its score from the 168 data elements that it collects in
high-availability form.  Iowa maintains fewer data elements in
high-availability form (147), but it also maintains 55 data
elements in paper form or low availability.  The distribution of
the number of data elements that a department maintains at each
level of availability points out the strengths and weakness of
departments' capacities for any group of data elements. 

Framework for defining and using the common core

The framework for analysis, which largely follows the format of
the Inventory and Obstacles questionnaires, aims to:  (1)
identify commonly collected data elements; (2) measure the
availability of data elements in a "high-availability format,"
that is, in electronic form for a large majority (more than 75%)
of offenders; and (3) measure the availability of data in the
areas where data are more commonly collected.  

The approach taken to identify commonly collected data elements
relies on delineating the major dimensions of corrections
processing that comprise each of the four stages of corrections
outlined above.  The term "common core" of data refers to the
data elements in specific dimensions of processing.  To say that
a common-core of data exists for a given dimension does not mean
that every data element is maintained in a high-availability
format by every department.  What it does mean is that, within a
dimension, enough data elements are maintained by a majority of
departments in a high-availability format to constitute a
meaningful core of information measures for a given dimension.  

Specifically, the common core consists of the data elements that
are in core dimensions of corrections processing.  Core
dimensions are identified separately for each stage of
processing; a core dimension is one in which more than an
average number of departments maintain data elements in
high-availability form (Footnote:  These are identified in the
following manner.  First, for each stage, the average number of
departments that maintain data elements in high-availability
form is calculated as the full-availability score divided by the
number of departments reporting data for the stage.  Second, a
similar measure is calculated for each dimension within a stage.
 The dimensions for  which the average number of departments
collecting data in high availability exceeds the average for the
stage are defined as the core dimensions.  Finally, all data
elements within core dimensions are included in the core. ) 

This approach is based on several assumptions.  First,
identifying the "common-core" areas in terms of dimensions
recognizes a limitation of the Inventory:  Departments collect
data elements that are similar to those identified in the
Inventory but may be defined in a somewhat different manner.

Second, within any common-core dimension, some specific data
elements may not be collected as commonly as others.

Third, the common dimensions may obscure the fact that specific
data elements are commonly collected even if the dimensions to
which they belong are not.  However, to the extent that
corrections concepts are measured by several data elements,
analysis of single data elements does not reveal the common
information areas.

Fourth, underlying the "common core" is the assumption that
departments have the capability to use computer methods to
generate statistical information about the common-core areas.

To use the common core or any data elements to generate
comparable statistics or indicators of corrections performance,
standard definitions for the statistics are needed.  Departments
may not be able to meet standard statistical definitions in all
cases.  For example, differences in definitions of offense
classes, types of offenders, and coverage of the corrections
system may vary among departments.  These may lead to
differences in definitions of data elements and, consequently,
in statistics.  These differences need to be understood. 

To begin to ascertain the reporting capabilities of departments,
the study also describes the types of queries for statistical
information to which departments are asked to respond; describes
and analyzes the obstacles that departments face in responding
to statistical queries; and describes departments' capabilities
for sharing and linking data externally and linking data
internally.  These objectives are achieved by analyzing the data
from the telephone surveys about the queries on statistical
information; by analyzing the Obstacles survey to identify and
rank obstacles for producing data; and by analyzing the
Inventory survey about capabilities to link data.



Chapter 1

Profiling and describing offenders



Highlights     
* All 52 departments maintain data electronically on the race
and sex of offenders entering prison; 51 do so for their date of
birth.

* At least 40 departments collect data on the occupation,
military discharge, marital status, and education level of
offenders, but some of these maintain only paper records on this
information.

* Twenty departments can not report on whether inmates have and
support children.

* Data describing offenders' demographic characteristics are
more commonly collected than data describing socio-economic
status or family relationships.

------------------------------------------------------------
Profiling and describing offenders
------------------------------------------------------------

The Inventory includes 29 data elements that can be used for
describing and profiling offenders under correctional authority.
This stage is comprised of three dimensions, each containing
several elements that describe:

* Demographic characteristics such as age, race, sex (11
elements); 

* Socio-economic status such as offenders employment, education,
and related experiences prior to prison admission (13 elements);
and

* Familial relationships (5 elements).

------------------------------------------------------------
Demographic characteristics of offenders 
------------------------------------------------------------

In the profiling offenders stage of corrections processing, data
elements that describe the demographic characteristics of
offenders are collected in any format-- paper or electronic -by
more departments than are the elements that describe offenders'
socio-economic status or familial relationships. 

Most departments maintain data elements on demographic
characteristics of offenders, such as their age, race, sex,
Hispanic origin, and residence in high-availability form
(Footnote:  High-availability format is defined as maintaining
data electronically for more than 75% of offenders table 1.1).
Specifically, of the 52 departments reporting, 51 meet this
criterion for maintaining data on the sex and race of offenders;
50 do so for the date of birth of offenders.  In addition, 39
departments have high-availability data elements describing
Hispanic origin, while 40 meet it for data on State of birth and
on country of birth.

Other demographic variables are less well represented according
to this high-availability measure.  Thirty-three departments
collect data elements on citizenship at a high-availability
level, and 31 departments do so on religious affiliation.
Information on immigrant status (i.e., whether legal or illegal)
is maintained in high-availability form by 18 departments, and
29 departments have high availability on offenders' residence
(table 1.1).

------------------------------------------------------------
Socio-economic status of offenders
------------------------------------------------------------

Relatively few departments maintain data elements for
substantial segments of their population on aspects of military
service, employment status, sources of income or financial
obligations at commitment in a high-availability form (table
1.2).  However, 29 departments maintain data in high
availability form on education level and 23 do so for whether
the offender served in the U.S. Armed Forces.  For the most
part, departments do not maintain data elements on income and
financial obligations.  For example, 17 departments do not
collect employment data and 40 do not collect data on income of
offenders at the time of commitment.  Also, about three quarters
of the departments do not collect data on the type or amount of
financial obligations.

---------------------------------------------
Familial relationships of offenders
---------------------------------------------

Many of the departments do not collect or maintain in
high-availability form data elements on the family circumstances
and living arrangements of offenders prior to admission (table
1.3).  With the exception of the 35 departments that maintain
data elements on marital status in high-availability form, no
other data element about familial relations is maintained in a
high-availability form by more than half of the reporting
departments.  Less than one-third of the departments have a high
availability to provide data on children and dependents.

In general, data elements on family and domestic circumstances
are not collected at all.  Thirty departments do not collect
data on the relationship of the offender to others in the
household -- although eleven departments collect such data in
electronic form at varying levels of coverage of correctional
populations.  Similarly, most departments do not collect data
that describes the residential status of offenders (e.g.,
whether they rent or own their residences, or are homeless).

---------------------------
Summary
---------------------------

Among the 29 data elements in the profiling offenders stage of
corrections processing, the 11 that comprise the dimension of
demographic characteristics of offenders have a higher
concentration of elements maintained in common and
high-availability form than do elements in the other dimensions
-- socio-economic status or family relations.

Data elements on race and sex are maintained by all departments,
and by all departments but one on age and marital status.  Data
elements on education, employment, occupation, and military
experience are maintained by a majority of departments.  Data
elements on residential status are maintained by very few
departments.  Each of the 29 elements in the offender profile
category is collected by at least some of the departments,
although some of these data elements are collected by only a few
departments.



-------------------------------------------------------
Chapter 2

Committing offenders into correctional authority
-------------------------------------------------------

Highlights

* All 52 departments maintain data for the type and date of
commitment into prison and the length of sentence imposed.  More
than 45 maintain several detailed data elements that describe
offenders commitment offenses and their expected release dates.

* At least 48 departments collect data about sentencing:  date
of sentencing, county of conviction, total length of sentences
imposed, and whether sentences are concurrent or consecutive.  A
few maintain these elements in paper form only.

* Thirty-seven departments maintain data about the date of the
criminal incident underlying the conviction; 27 do so about
whether a weapon was involved; 21 about the number of victims in
the incident; and 16 about victim injuries.  In most cases,
these departments maintain this information on criminal
incidents in paper form. 

* Most departments (between 39 and 47) maintain data
electronically about offenders needs, their security
classifications, and units in which they are housed.

* Data describing conviction offenses, sentences imposed,
current commitment, expected time to be served, risk assessment
classification decisions, and confinement characteristics are
more commonly collected than data describing other areas of
committing offenders.

The Inventory includes 70 data elements that describe the second
stage of corrections processing: committing offenders into
correctional authority -- specifically, committing offenders
into prison.  This stage includes elements that describe the
offenses and sentencing decisions leading up to the commitment
into prison and elements describing the assessment and placement
of offenders upon commitment.  The 70 data elements in this
stage are organized into three broad categories that provide
information about the offenses leading to the conviction and
sentences, about the sentences imposed by the court, and about
the assessment and confinement decisions made by corrections
officials upon receipt of an offender from the court or other
authorities.  Among the categories, the data elements are
further divided into 10 dimensions that describe more finely
defined aspects of this stage.

In describing offenses leading to the conviction underlying a
commitment, the 29 data elements are organized into 3 dimensions:

*Criminal incident underlying the conviction offenses, including
the data elements that describe the victims of crimes (14 data
elements);

*Conviction offenses (7 data elements); and

*Offenders criminal histories and records of prior arrests,
convictions, and criminal justice supervision status at the time
the conviction offense was committed (8 data elements).


The 19 data elements about sentencing outcomes and type of
current commitment are organized into 3 dimensions that
describe: 

* Sentences imposed by the court (13 data elements);

* Type of current commitment (3 data elements); and

* Upon commitment, expected time to be served until release from
prison (3 data elements).



Finally, the 22 data elements that describe assessment,
classification, and confinement decisions made by corrections
officials are organized into 4 dimensions: 

* Risk assessment data elements describing the characteristics
of offenders leading to placement decisions (4 data elements);

* Needs assessment describing the needs of offenders for
treatment or placement (6 data elements);

* Classification decisions including 9 data elements that
describe an offender's security level at commitment; and

* Confinement characteristics describing the location and
housing into which an offender is placed (3 data elements).


------------------------------------------------
Offenses leading to commitments
------------------------------------------------

Within this area, the data elements are organized into three
dimensions, those that describe the criminal incident, the
conviction offense, and the offender's criminal history.

Few departments maintain any of the data elements on the
criminal incident in a high-availability form (table 2.1).  This
includes data elements about the incident itself, victim-related
information, and other damages.  The exception to this is the
data element on the date of the criminal incident, which is
collected by 21 of the departments.  About one quarter of all
departments reporting maintain data elements about the criminal
incident in paper form.  Few retain descriptive data about the
victims of offender crimes.  Overall, the majority of
departments do not maintain data elements about victims.

Forty-six departments have conviction offense information in
high-availability form (Footnote:  High-availability format is
defined as maintaining data electronically for more than 75% of
offenders.  table 2.2).  Most of the departments (42) obtain
their conviction offense information from court commitment
orders and maintain them as high-availability data elements and
most (43) maintain high-availability data elements about the
severity of the offenses (e.g., felony or misdemeanor).  A
majority of departments have specific information about criminal
codes and written descriptions of offenses; 31 departments
maintain some data elements with detailed offense descriptions
in high-availability form.

More than half of the departments (31) maintain
high-availability data on the criminal justice status of the
offender at commitment (table 2.3).  But, a large majority of
departments do not capture other data elements on criminal
history in high-availability form.  

Nearly half (25) of the departments maintain data elements on
offenders' prior record of arrests and convictions in a
high-availability form.  Twenty-five maintain high-availability
data elements on the severity level of prior offenses, and 24
departments do so on the actual number of prior convictions. 
Relatively few departments have high availability to produce
data on arrests, and a sizable number (23) do not maintain data
elements on the number of prior arrests.  Only 20 departments
collect data elements that describe whether offenders were
habitual offenders in high-availability form.  Finally, many of
the data elements about criminal history are collected in paper
form, or in electronic form for smaller segments of the
population.  For example, data on the number of prior arrests is
collected by an additional 20 departments in either electronic
form for smaller segments of the population or in paper form.

-------------------------------------------------
Sentencing information
-------------------------------------------------

Most departments collect data elements about the sentences
imposed by courts in the high-availability format (table 2.4). 
More than 44 departments maintain high-availability data
elements on the date of sentencing, the number of sentences
imposed, whether the sentences imposed were concurrent or
consecutive sentences, the length of sentence for each offense,
and the total length of sentence.  In addition, 47 departments
maintain high-availability data elements on the county in which
the offender was sentenced, and 34 maintain elements that
identify the sentencing judge.

The number of departments maintaining high-availability elements
on mandatory sentences, combinations of sentences, and
supervisory sentences is lower than the number having
high-availability elements on the basic sentencing information,
but a majority or near majority of departments maintain
high-availability elements on these other aspects of sentencing
(table 2.4).

In general, departments are less likely to have data elements
about sanctions other than prison sentences than they are about
the prison sentences.  More than half of the departments (30)
have a high availability to produce data on length of community
supervision, but only 23 departments have this availability for
monetary sanctions, and even fewer (19) have it for the amount
of the monetary sanction.

All departments can provide data on type and date of commitment
in electronic form, and all but two can provide both of these
data elements in high-availability form (table 2.5).  Fewer
departments maintain data on the agency having the authority to
release the offender (28).

As with data elements about sentencing, most departments also
maintain high-availability data elements about expected release
dates of offenders.  Of the 52 reporting departments, 48
maintain high-availability elements on the expected date of
release, 43 on the expected parole release date, and 45 on the
date of expiration of sentence (table 2.6).

------------------------------------------------
Assessment and confinement decisions
------------------------------------------------

About half of the departments maintain data elements that are
used in assessing offenders' risk in high-availability form. 
More specifically, 28 departments maintain high-availability
data elements on offenders' history of violence, 22 on the use
of a weapon, and 16 on gang membership.  Additional departments
maintain these data elements as in medium availability or paper
form, but relatively sizable numbers of departments do not
maintain elements on these aspects of offenders' behavior (table
2.7).

On the data elements that measured needs assessment, departments
were split.  For the data element on the type of needs that
offenders had, 34 departments maintained it in high-availability
form.  For offenders' psychological and medical histories, 20
and 26 departments, respectively, maintained elements in
high-availability form.  Conversely, for program participation
prior to imprisonment and drug testing upon admission, most
departments did not collect data elements to measure these
activities (table 2.8).

Despite the interest in medical conditions of offenders, and the
concern over the increasing incidence of tuberculosis infection
and HIV-infected populations in prison, only 26 departments
maintain high-availability data elements on medical conditions. 
However, an additional 17 departments maintain data elements in
some other format on medical conditions.  In addition, 15
departments do not collect data on psychological history at time
of admission.  Twenty departments have psychological history
data for large segments of their populations.

In general, most departments maintain as high-availability
elements, the data about classification decisions (table 2.9). 
For example, 45 departments have a high availability to produce
data on security level, 43 have a high availability on date of
initial classification, and 45 have a high availability to
produce date of classification change.  The outcomes of these
procedures, however, are produced in a high-availability range
at slightly lower levels: 40 departments have data representing
a classification index or score, and 30 departments collect data
identifying the agency making the decision on classification. 
Four departments (Alaska, District of Columbia, Idaho, and New
Mexico) do not collect data on security level (not shown in a
table).

Relatively few departments have a high availability to produce
data on various types of scores and indices related to
classification.  Only 18 departments collect data at a
high-availability level on a psychological index or score, and
23 departments collect data at this level on a medical
classification index or score.

Similarly, all departments collect data on the type of facility
to which the offender is placed at admission (table 2.10). 
Fifty departments maintain it as a high-availability data
element, and only one (New Mexico) does not collect data on type
of facility at placement (not shown in a table).  Slightly less
than all departments collect information about the type of
housing into which offenders are placed at admission (table
2.10).  Thirty-eight departments have a high availability to
provide data on the type of housing unit in which the offender
is placed, and 40 departments have a high-availability level on
data on the type of special housing which offenders are placed. 
Seven departments (Alaska, Indiana, Idaho, Louisiana, New
Mexico, Minnesota, and Wisconsin) do not collect data on the
type of placement housing at admission and five (Alaska,
District of Columbia, Idaho, Minnesota, and Wisconsin) do not
collect data on special units housing  the offenders.

-----------------------
Summary
-----------------------

Among the 70 data elements in the committing offenders stage of
corrections processing, the 42 data elements that measure
conviction offenses, sentences imposed, current commitment,
expected time to be served, risk assessment, classification
decisions, and confinement characteristics are the most commonly
collected by the reporting departments.  The data elements that
measure the criminal incident leading to the conviction offense,
which include data elements that describe victims of criminal
incidents, are the least commonly collected by the departments.



For the more commonly collected data elements, more than 45
departments maintain in high-availability form the data elements
on number and type of conviction offenses, county of sentencing
court, date and length of sentence, whether sentences are
imposed concurrently or consecutively, type and date of
commitment, expected dates of release, type of confinement
facility, and date of classification changes.


--------------------------------------------------
Chapter 3

Managing offenders in corrections facilities
---------------------------------------------------

Highlights

* All 52 departments maintain data electronically on offenders'
types and dates of releases from prison and dates and types of
transfers between facilities, and most (up to 39) report data on
the reasons for changes or adjustments to sentence and time to
be served, including good time and other credits.

* Forty-two departments maintain data on offenders'
participation in programs, and 28 of these do so at a
high-availability level.  In general, departments collect data
on programs or medical care in paper format.

* Thirty-three departments report that they maintain data on
drug tests since admissions, but only 18 maintain this
information electronically, and only 15 maintain data on the
results of the tests electronically. 

* Forty-seven departments maintain data about the most recent
occurrence of misconduct in prison and most do so
electronically.  More than 27 departments maintain detailed
information about these incidents -- such as who was involved,
whether drugs, alcohol, or weapons were involved, and injuries
-- but most of these maintain the data in paper form.

* Forty-two departments maintain data on victim notification
requirements.

* Data describing post-commitment movements, good time and other
adjustments to sentences, offender registry, and releases from
custody are more commonly collected than other areas of managing
offenders.

The third major stage of corrections processing relates to
managing offenders while they are in correctional facilities.
This stage includes data elements that describe movements of
prisoners, the procedures and actions that corrections officials
take to manage offenders in their custody, behaviors of
offenders leading to disciplinary actions, and official
responses to misconduct.  The Inventory includes 63 data
elements about managing offenders.  These elements are organized
into 3 broad categories that describe routine management and
program participation, the release of offenders from custody,
and internal security matters. 

To describe routine management activities, the 26 data elements
are organized into 4 dimensions:

* Post-commitment transfers between jurisdictions and movements
between and within facilities (7 data elements);

* Program participation by offenders (11 data elements);

* Drug testing since prison admission (2 data elements); and

* Medical care of offenders (6 data elements).


To describe how offenders are released from custody and the
processes leading to adjustments to their time served in prison,
the 18 data elements in this category are organized into 3
dimensions:

* Good time and other adjustments to sentences and length of
stay, as well as the reasons for the changes (10 data elements);

* Method of release from custody (5 data elements); and

* Offender registry requirements (3 data elements).


Finally, in this stage of managing offenders, the 19 data
elements related to internal order and security are organized
into 3 dimensions that describe behaviors of offenders and
official responses to misconduct:

* Misconduct and infractions -- describing events leading to
disciplinary actions (11 data elements);

* Responses to misconduct -- describing the immediate response
to misconduct taken by corrections officials (3 data elements);
and

* Proceedings against offenders -- describing the legal
proceedings and outcomes taken in response to misconduct (5 data
elements).

---------------------------------------------
Routine offender management
---------------------------------------------

Fifty-two departments maintain in high-availability form data
elements that track the movements of prisoners between
facilities and the transfer of offenders to other jurisdictions
(Footnote: High-availability format is defined as maintaining
data electronically for more than 75% of offenders table 3.1).
Forty-one departments maintain data elements that track internal
movements in high-availability form.  Slightly fewer departments
maintain high-availability data elements about the reason for a
transfer or internal movement (31) or the official who
authorized the movement or transfer (9 and 16 departments,
respectively).


In general, very few data elements are collected on programmatic
activities at a high-availability level.  Data elements on types
of programs are collected by 28 of the departments at a
high-availability level.  Twenty-eight departments collect data
at a high-availability level on the date the offender began the
program, and 26 departments do so on the date the offender ended
the program (table 3.2).  Data elements on programs tend to be
collected electronically for more than 75% of the corrections
population.  About a fifth collect this information in paper
format.  About half of the departments do not collect data on
reasons for program participation or on the authorization for
the program.

Of 52 departments reporting, 14 departments have a high
availability to produce data on drug tests of offenders since
admission, and 12 departments can do so on the date of the last
drug test (table 3.3).  About a third of the departments do not
collect either of these data elements, or collect these data in
paper records only.

In general, departments maintain data elements on medical care
of offenders in paper records only.  With the exception of the
current medical condition of offenders, for which 18 departments
have a high availability to produce data, less than one third of
the departments collect data in electronic format on the medical
condition of offenders for large segments of their populations
(table 3.4).  In addition, a few departments collect data in
electronic format for less than 75% of the offender population,
and more than 10 do not collect medical data on offenders at
all.  Twenty-four departments report that they do not maintain
data elements on the costs of medical treatment.

---------------------------------------------
Methods of release from prison
---------------------------------------------

Thirty or more departments maintain data elements in a
high-availability format on whether sentences are modified, by
what amount, and the dates and reasons for good time or other
adjustments.  More than half of the departments (28) have a high
availability on data relating to changes in available good time
credits and only 18 departments maintain data elements in
high-availability on special credits (e.g., housing credits).
Also, 29 departments do not collect data on these special
credits (table 3.5).

High-availability data elements on the type and date of release
from custody are maintained by all departments.  Thirty-four
departments can produce data at a high-availability level about
the time served in custody, and 36 departments can do so on the
type of facility that the offender is released to (e.g.,
community corrections facility, work release center, treatment
facility).  More than half of the departments have a high
availability to provide data on the agency gaining jurisdiction
of the offender on release.  About a quarter of the departments
do not collect data on time served, or on the jurisdiction or
facility to which the offender is released (table 3.6).

About two-thirds of the departments have a high availability to
provide data to comply with victim notification requirements
(table 3.7).  Less than one half of the departments can provide
data on whether an offender is required to register as a sex
offender under Megan's Law or some similar statute.  Only 14
departments can identify whether an offender actually registered
as a sex offender under such statutes.

---------------------------------------------
Internal order and security
---------------------------------------------

With the exception of the type of misconduct and the date of the
event, most departments do not collect data about misconduct and
infractions in a high-availability form.  Thirty-three
departments have a high availability to provide data on the type
of misconduct or infraction, and 34 can do so on the date of the
event (table 3.8).  Twenty-seven departments have a high
availability to produce data on the history of offenders'
behavior in custody.  About half of the departments collect data
in electronic form on the location of the event.

Overall, much of the data pertaining to internal security is
collected on paper records (table 3.8).  For example, more than
a third of the departments collect data in paper records about
who was involved in the event, who sustained an injury, the type
of injury sustained, whether drugs or weapons were involved, and
the amount of property damage.  Substantial numbers of
departments indicate that they do not collect data on certain
aspects of misconduct.  Twenty-two departments do not collect
data on whether weapons were involved, and 24 do not collect
data on whether drugs or alcohol were involved in the incident.

Only 12 departments maintain high-availability data elements on
the responses to infractions (table 3.9).  About one third of
the departments do not collect data at all on the type or date
of the immediate response, although another third do so in paper
form.  With the exception of the result of the immediate
response (on which 21 departments report a high availability to
produce data) relatively few departments produce these data
elements in electronic form.

Data elements about formal legal responses to violations of
internal order are also not generally collected at
high-availability levels.  For example, 26 departments have a
high availability to produce data about the disposition of the
proceeding, but fewer than half can do so on charges filed
against the offender (table 3.10).  Also, 20 departments can
produce data at a high-availability level on the date of legal
procedure.  Sixteen departments have a high availability on data
relating to who initiated the response, and 21 departments have
data at this level on the type of response.  About a third of
the departments report that they do not collect these data
elements at all.

-----------------------
Summary
-----------------------


Among the 63 data elements that describe management of
offenders, data elements that measure the form of release, good
time and other adjustments to sentencing, post-commitment
movements, and offender registry are commonly collected in
high-availability form by reporting departments.  Data elements
on program participation and medical care are commonly collected
but largely in paper form.  Data elements on drug testing,
offender misconduct and responses to misconduct are less
commonly collected, but many of them are also maintained in
paper form.  Overall, many of the data elements pertaining to
internal security and medical care are collected in paper
records.

------------------------------------------------------------------------------
Chapter 4

Supervising offenders on release and
maintaining public safety
------------------------------------------------------------------------------

Highlights

* Forty of the 52 departments maintain data about the behaviors
of offenders released into the community; 12 do not.  

* Thirty-eight of these departments record data on the reasons
for termination of supervision; 32 report on the type of new
crime committed by offenders under supervision; and 35 report
data on when offenders return to prison after having been
sentenced for a new crime.  Most departments maintain this
information electronically.  

* For crimes committed by offenders under supervision, 35
departments have data on the type of crime, but no more than 16
have data about victim-related elements of these crimes, and
fewer still maintain data on the characteristics of victims; for
those that do, most maintain victim information in paper form.

* While 31 of 40 departments maintain data on the address of
offenders released from prison, only 20 maintain data about
released offenders' living arrangements and 17 about their
employment; for many departments, this information is maintained
on paper.

* Data describing reasons for terminating supervision and the
criminal justice response to violations of conditions of
supervision are more commonly collected by these 40 departments
than other areas of supervising offenders.

The fourth stage of corrections processing in the Inventory
relates to the supervision of offenders released from custody
and the maintenance of public safety.  The Inventory includes 45
data elements related to this stage.  As with the second and
third stages, this fourth stage is divided into broad categories
which are then divided into dimensions.  The data elements in
this stage measure where offenders are in the community, what
they are doing there, and whether they have a record of criminal
activity after release.  The data elements also address the
behavior of offenders under supervision in the community, any
new crimes committed, and the response to these crimes. 
Additional data elements focus on information about victims of
crimes committed by offenders under supervision in the
community. 

These 45 data elements fall into three categories:  social
integration, offender behaviors after release from custody, and
new crimes and victims of crimes.  Social integration includes
one dimension of data elements about offenders' residence and
employment status during release.  The data elements about
offender behaviors on release are organized into 3 dimensions:

* Residence and employment during release (7 data elements);

* Behaviors on supervision leading to reasons for terminating
supervision (12 data elements); and

* Responses to new crimes and violations of conditions of
supervision (10 data elements).

The other major category of elements relates to new crimes and
victims of crimes.  This category is organized into 2 dimensions:

* Details of new crimes committed on supervision (9 data
elements); and

* Details about victims of crimes committed by offenders under
supervision (7 data elements).

In the supervising offenders stage, at most 40 departments of
correction use their adult sentenced prisoner information
systems to maintain data on offenders while they are under
supervision in the community:

* Forty departments in tables 4.1, 4.2, and 4.3 reported having
data about offenders under supervision; and

* Thirty-eight departments in tables 4.4 and 4.5 reported having
detailed data about criminal incidents committed by released
offenders.

------------------------------
Social integration
------------------------------

Relatively few of the 40 departments reporting that they
maintain data elements about released offenders maintain the
data elements on offenders employment experiences and residence
on release (table 4.1).  With the exception of the address of
the offender, which is collected in a high-availability form
(Footnote:  High-availability format is defined as maintaining
data electronically for more than 75% of offenders)  by 19
departments, most of the departments do not maintain the data
elements that describe personal data about offenders on release.

-----------------------------------
Offender behavior after release
------------------------------------

For the 40 departments that maintain data on offenders released
into the community, a large percentage are able to report data
in high-availability form. Most departments collect the key data
elements on completion of release supervision in a
high-availability form.  Of the 40 departments reporting, 32
each have a high availability to produce data on type of
supervision, on whether supervision was terminated, and on the
date supervision was completed (table 4.2).  Twenty-seven
departments maintain high-availability data elements about
whether an offender absconded while on release; 26 do so for the
type of new crime that was committed; and 25 departments
maintain data elements on the length of supervision in high
availability.  More than a third of the departments collect data
in a high-availability form on the type of technical violation,
and on the dates related to the new crime or violation.

For the data elements related to the criminal justice response
to new crimes and technical violations committed by offenders
under supervision in the community, departments vary in their
capacities to maintain information in electronic form.  For the
data elements that measure an offender's return to prison, such
as date of return to prison, and whether an offender was
sentenced, most of the departments (32 and 27, respectively)
have high-availability data elements (table 4.3).  Twenty-one of
the departments maintain high-availability data elements on
offenders arrested and subsequently adjudicated for crimes
committed while on release.  Most departments that report these
data elements obtain their data when offenders return to prison
(28 departments), but 18 obtain data on the new crimes committed
by offenders on release when the offenders are convicted, and 10
departments report that they obtain this information when
offenders are arrested (not shown in a table in this chapter).


------------------------------------
New crimes and victims of crimes
------------------------------------

Although a relatively high number of departments can identify
whether supervision terminated for reasons of a new crime (table
4.2), departments vary in their availability to maintain data
elements that describe the new crimes committed by released
offenders, the victims of the crimes, or the damages done by the
offender (table 4.4).  Of the 38 that report maintaining data
elements about criminal incidents involving offenders under
supervision, 31 departments report in a high-availability format
the supervision status of the offender, and 27 departments
report on the type of offense associated with the new crime. 
However, only 2 departments maintain high-availability data
elements about the victims of these crimes.  Six maintain the
data elements for the location of the event, and 11 maintain
data elements about victim restitution in electronic form for
more than 75% of the offender population.  Most of the
departments do not maintain these elements (table 4.4).

Few departments collect detailed data in any form about victims
in incidents committed by offenders on release.  Thirteen report
collecting data elements on the sex of the victim and whether
the victim was a child; 16 maintain data on the victim's
address.  In addition, the comparatively few data elements about
victims of crimes committed by offenders on release in the
community are maintained primarily in paper format.  For
example, only 4 departments report that they maintain data
elements on the sex of the victim in electronic form (table 4.5).

------------------------------
Summary
------------------------------

Twelve of the 52 departments in the survey report that they do
not maintain data elements on the behaviors of offenders under
supervision in the community. The other 40 departments report
that they maintain these data elements.  

For the 40 departments that maintain data about offenders under
supervision, the data elements that describe the behaviors of
offenders leading to terminations of supervision (table 4.2) and
the criminal justice response to these behaviors (table 4.3) are
more commonly collected in paper and in high-availability form
than the other areas of data. 

Most of the departments maintain data elements in
high-availability form on the type of supervision, whether
supervision was terminated, the length of supervision, and
reasons for termination of supervision. 

Most departments maintain few of the data elements about the
employment status and living arrangements of offenders on
release in the community.

----------------------------------------------------------
Chapter 5
Facility management information
----------------------------------------------------------

Highlights

* Forty-two departments maintain information about the type of
in-prison programs available to offenders; 37 report maintaining
data about assessments of these programs.

* At most, 25 departments maintain data about the quality and
availability of medical staff in prison facilities, and most
maintain these data in paper form.

* Thirty-four departments maintain some data about annual
operating costs of facilities; less than half maintain data on
whether facilities generate revenue and if so, how much.

The Inventory contains 15 questions about correctional
facilities, programs and their evaluation, medical care,
staffing, revenues, and costs.  In the area of program
management, the Inventory includes 3 questions about the types
of programs and their assessment.  On medical services, it asks
3 questions about the number of medical staff, their
qualifications, and their availability (Footnote:  Data elements
about medical services provided to offenders are included in
Chapter 3, "Managing offenders in corrections facilities.") 
Finally, the inventory includes 9 data elements about the number
of facilities, staff, costs, and revenues.  Departments were not
rated on a full-availability measure for these data elements. 
Rather, their capability to maintain these data elements
electronically is distinguished from their capability to
maintain them on paper.

-----------------------------
Program management
-----------------------------

Program management data are not widely collected.  About half of
the departments (25) collect data in electronic form on types of
programs offered by facilities under their jurisdiction.  A
third of the departments (17) collect information about programs
in paper form only, and about one half of the departments (24)
collect evaluation data on programs in electronic form.  Most
departments (36) do not collect data on dates of evaluation of
programs (table 5.1).

----------------------------------------------------------
Medical services offered within facilities
----------------------------------------------------------


Data about medical services staff in facilities are not widely
collected.  One half of the departments (25) do not maintain
data elements on medical services.  About one third of the
departments maintain data about medical services offered by
facilities in paper format only.  For example, 17 departments
report having data in paper format on the number of staff who
provide medical services; 21 departments report having data
elements on the qualifications of medical staff; and 18
departments maintain data on the availability of medical staff
(table 5.2).  Few of these departments maintain data elements
about medical services in electronic form.  For example, only 8
departments collect data in electronic from on the number of
medical staff that service their facilities.

-----------------------------
Managing facilities
-----------------------------

The Inventory included 9 data elements about the management of
facilities, including elements about staffing ratios, beds per
facility, revenues generated by facilities, and costs to operate
facilities.  Most of the departments (40) maintain data in
electronic form on the total number of facilities, and 35
departments maintain data electronically on the number of beds
per facility (table 5.3).  More than half of the departments
(29) maintain data electronically on the number of staff at
facilities, and 25 departments maintain data elements
electronically on the number of custodial staff at each
facility.  Up to a quarter of the departments report that they
do not collect these data elements (table 5.3).

About one half of the departments (25) do not maintain data
elements on whether facilities generate revenue.  In addition,
25 departments do not collect data on the amount of revenue
generated by facilities.  However, one quarter of the States
collect this type of data on paper records.  Almost half of the
departments (22) maintain records about costs to operate
facilities or on capital costs  in paper format only.  About one
third of them (16) do not collect these data elements.  Twelve
departments collect cost data in electronic form.

-----------------------------
Summary
-----------------------------

In general, departments maintain limited information about
programs, medical services, and facilities.  However, with the
exception of data elements about the number of facilities, beds
per facility, and staff per facility, few departments maintain
any of this information in readily accessible electronic form. 
Many departments maintain this information in paper form or do
not collect it.  This is especially the case for medical care
and some of the data about facilities for which more than half
of the departments do not maintain the data.



--------------------------------------
Chapter 6
Reporting capabilities
-----------------------------------

Highlights

* The extent to which departments maintain all 207
offender-based data elements electronically for a large majority
of offenders ranges from 85% to 16%.  Thirty-two departments
rate at or above 50% for all data elements on this rating of
data availability.

* Seven departments rate above 70% of full availability for the
data elements in the profiling offenders stage.  Twelve do so in
the committing offenders stage, as do 10 departments in the
managing offenders stage and 9 departments in the supervising
offenders stage. 

* Departments' ability to provide statistical information about
released offenders varies.  All departments maintain the records
of released offenders, and about half can electronically link
and retrieve archived records of these offenders when they
return to prison.

* Only 40 departments maintain data about the behaviors of
offenders under supervision in the community; and only 38
maintain data on the crimes they commit while under supervision. 

* Departments rate staffing and software problems as the most
severe problems they must overcome in providing statistical
information about offenders.  Thirty of the 46 that rate
staffing as a serious problem also rate software or data
availability as a serious problem.

Thus far, the report has focused on whether, and how,
corrections information systems maintain data elements.  This
chapter shifts the focus of the report to how departments use
data elements and to the obstacles and barriers they confront in
providing statistical information about offenders, and sharing
data electronically.

--------------------------------------------------------------
Forms of statistical information provided by departments
--------------------------------------------------------------

Statistical information describes outcomes, activities, or
events pertaining to groups of offenders or to a corrections
system as a whole.  Such information may be used for many
purposes--such as profiling the composition of offender
populations; developing management and budget plans; responding
to inquiries from the press, academics, or law makers;  and
developing corrections performance indicators.  Questions such
as, --How many offenders are in prison for robbery at
yearend?-are commonly requested pieces of statistical
information that profile offender populations.  Answers to
questions such as, --How many offenders who were released from
prison during 1995 returned to the prison from which they were
released within one year of their release?-- are often used for
evaluative purposes, either implicit or explicit.  Queries about
--the proportion of all offenders who remained drug-free during
the past year,-- or --the proportion of eligible offenders who
were involved in prison work or training programs during the
past year-- often are asked as indicators of the degree to which
a corrections system achieved a particular goal. 

Information officials report that departments receive many
different types of requests for statistical information.  The
most common are for summary statistics about specific groups of
offenders.  In addition to internal departmental requests for
information from corrections managers, departments also
regularly provide statistical information to governors,
legislators, and officials in other State agencies (e.g., State
auditors, departments of education, mental health, or labor).
Such summary information is used for a variety of purposes:  for
scheduling (courts), assessing suitability of offenders for
placement (halfway houses), sentencing and criminal
investigations (district attorney's offices), locating
--dead-beat dads-- (social service agencies), forecasting prison
population (State planning agencies), and verifying benefits
(Social Security Administration). 

Federal agencies request summary statistical information
regularly from corrections departments.  The Bureau of Justice
Statistics requests summary data on several surveys of
corrections populations: The National Prisoner Statistics
(summary data on prison admissions, releases, and stocks), the
Parole Data Survey (summary statistics on offenders on parole
and other forms of post-incarceration supervision), and the
Probation Data Survey (summary statistics on offenders on
probation).  For these particular surveys, departments are
required to provide statistical information that is based on
external standards or definitions.  Thirty-eight departments,
for example, provide data to the Bureau of Justice Statistics'
National Corrections Reporting Program, which requires them to
meet BJS definitional standards for counting offenders admitted
into or released from prison, their form of admission, sentences
imposed, and method of release.  Most departments that submit
data extracts to the BJS reporting program are able to meet
these definitional standards.  Departments that are unable to
meet the definitions provide reasons why they cannot do so.

Departments also respond to requests for data extracts that
requesters of the extracts intend to analyze for their own
purposes.  Such requesters include researchers, newspapers,
commercial banking systems and other private companies.  Data
extracts are provided on diskette, tape, or other medium, or via
File Transfer Protocol.  (For example, in Oregon, several
companies purchase data tape from corrections departments and
resell them to other entities looking into criminal histories of
potential employees.) 

-------------------------------
Availability of data
-------------------------------

Information systems cannot easily fulfill requests for
information if the data are not readily available for analysis
or for sharing with other jurisdictions. Maintaining data
elements electronically for all (or most) offenders allows for
greater data availability and facilitates responding to
statistical inquiries.  The Inventory rates data availability
using an index that measures the extent to which departments
maintain data elements electronically for a majority of
offenders (more than 75%).  The availability index ranges from
0% to 100%.  A rating of 100% means that a department maintains
all data elements in electronic form for the majority of
offenders (full availability), while a rating of 0% indicates
that a department does not collect any of the data elements
being rated.

To obtain a department's score on the availability index, each
data element in a set of elements is given a value of 3, 2, 1,
or 0, depending on how the department maintains the data
element.  High-availability data elements (maintained
electronically for more than 75% of offenders) are given a value
of 3.  Medium- availability data elements (maintained
electronically for less than 75% of offenders) are given a value
of 2.  Low-availability data elements (maintained in paper form
only) receive a value of 1.  Finally, no-availability data
elements (a department does not collect the element) are given a
zero (Footnote 1:  1This scoring system is described on pages
10-12 in the --Introduction-- to this report.)  After each
element is scored, the sum of the values for a group of elements
is computed.  This sum, also known as a department's
availability rating, is divided by the total number of points
that would be obtained if all data elements were maintained as
high-availability data elements and then multiplied by 100%.

For example, for the 207 data offender-based data elements,
Colorado receives an availability index of 83% of full
availability.  Colorado receives a total of 518 points as its
availability rating out of a possible 621 points, if it
maintained all data elements as high-availability data elements.
The rating of 518 is obtained from: 168 high-availability
elements (168 x 3 points = 504 points), 7 medium-availability
data elements (7 x 2 points = 14 points), and 32 no-availability
data elements (32 x 0 points = 0 points).  The sum of the
points, 504 + 14, equals Colorado's score of 518.  Finally, 518
divided by the 621 possible points yields the availability index
of 83% when multiplied by 100%.

Ten departments receive a full-availability rating above 70%
(table 6.1) for the entire set of 207 offender-based data
elements.  Nine of these departments are among the 40 that
maintain data elements for all 4 stages.  One of them is among
the 12 that maintain data on 3 of 4 stages.  Twenty departments
are rated at less than 50% of full availability.

Generally, a department's availability rating increases with the
number of data elements collected.  For example, Iowa collects
all but 5 data elements and  rates at 80% of full availability,
and Arizona, which rates at 85%, has all but 19 data elements
(table 6.1).  Many of the departments that rate below 50% of
full availability collect less than half of the data elements.
The two lowest rated departments, Alaska and the District of
Columbia, do not collect a substantial number of data elements
(157 and 156, respectively).

Within the four stages of corrections processing, the
availability of data among departments varies.  While no
department in any stage is rated at 100% of availability, some
stages have greater availability than others.  Within the
profiling offenders stage the availability index ranges from 80%
to 30% (table 6.2). Seven departments have full-availability
ratings above 70% and 22 departments are at less than 50% of
full availability.  The committing offenders stage ranges from
92% (Iowa) to 16% (Alaska) of full availability, with 12
departments rating above 70%.  One half of the departments are
at 60% or more of full availability.  Only 11 departments rate
less than 50% of availability.  In the managing offenders stage,
there are ten departments rated at above 70% of full
availability; while 20 operate at 50% of full availability.  The
full-availability ratings for managing offenders data range from
94% (Missouri) to 11% (Alaska).  Twelve departments in the
supervising offenders stage do not maintain in the information
systems data about released offenders, and 14 do not maintain
data about new crimes committed by offenders under supervision
(including the victims of these crimes).  For the 40 departments
that do collect data about either or both of these areas, the
full-availability ratings for this stage range from 93%
(Arizona) to 7% (District of Columbia).  Only two departments
receive a rating of 90% or more of full availability.  Less than
a third have full-availability ratings of more than 50%.

Not only does the availability of data among stages vary, but
the number of data elements maintained in high-availability form
also differs. In the profiling offenders stage, no department
has the capability to provide all 29 data elements in a
high-availability form (Appendix G).  Most departments maintain
some data in electronic form.  Thirteen States and the Federal
Bureau of Prisons (BOP) have the capacity to provide all 11 data
elements on demographic characteristics in a high- or
medium-availability form (not shown in a table).  Most
departments maintain very few of the elements on socio-economic
status of offenders in electronic form. Two departments (Georgia
and the BOP) maintain all five data elements about family
relationships in electronic form.

Most of the departments with relatively high-availability
ratings in the committing offenders stage maintain a large
number of data elements in high-availability form (Appendix G).
However, some of these departments also maintain many data
elements in a medium-availability form.  For example, Ohio and
Tennessee rate above 65% of full-availability, and each
maintains a relatively large number of data elements in medium
availability.

In general, departments maintain conviction, sentencing, and
commitment data with high availability.   In the area of
sentencing information, departments generally have much higher
capacities to produce all elements in electronic format.  Nine
departments (Alabama, Colorado, Florida, Georgia, Missouri,
North Carolina, Washington, Oregon, and South Carolina) maintain
all 13 data elements electronically on sentencing information,
and 26 departments have all three data elements on type of
commitment in high-availability form (not shown in a table).  No
department maintains in high-availability form all of the 14
data elements about the criminal incident. 

In the managing offenders stage, seventeen departments maintain
a third or more of the data elements in paper, or
low-availability form (Appendix G).  In general, these data
describe program participation and outcomes, drug testing,
medical treatment, and misconduct and infractions.  Data
elements that measure the form of release, good time and other
adjustments to sentencing, post-commitment movements and
offender registry are maintained in high-availability form. Data
about post-commitment transfers, and methods of release from
prison are maintained by all 52 departments, and a majority of
departments have a high availability for data about movements,
good-time adjustments, and victim notification requirements.

No department in the supervising offenders stage maintains all
the data elements in high-availability form (Appendix G).   Many
of the departments either do not collect sizable numbers of
these data elements, or maintain data in low-availability form.
For example, the District of Columbia  does not collect 42 out
of 45 data elements; Wyoming maintains 43 out of 45 data
elements in paper form.   The data with the highest availability
are those that describe offenders' behavior on release and the
response by corrections to violations of conditions of
supervision.  Few departments maintain high-availability data
about victims of crimes committed by released offenders.

Departments are not rated on a full-availability measure for
facility-based data elements such as program management, medical
services, staffing, and facility costs.   Rather, their ability
to maintain these data elements electronically is distinguished
from their ability to maintain them on paper.  Fourteen
departments maintain more than half of the 15 facility-based
data elements electronically (Appendix G).  But, 26 other
departments did not have at least 10 of the data elements.  Many
departments report that they do not maintain data electronically
on program management, medical services and staffing, and costs
of facilities.

-------------------------------------
Capacities to link and share data
-------------------------------------

To answer many statistical queries, departments need to link
data from several databases or files, or to databases maintained
by other sources.  For questions related to offenders'
histories, for example, departments need to link current records
with the records of past behaviors.  This may involve extracting
archived records from tapes or other media and linking them to
the existing case-management database.  For questions about
offenders' behaviors when they are outside the jurisdiction of
correctional institutions--such as when they are released into
the community--departments may have to link their records with
those in an information system outside corrections, such as that
maintained by a parole agency.  This requires corrections
departments to link parole records (which many do not record) to
those in their database, and process the combined information to
produce the desired statistic.

Respondents expressed mixed views about the need to link data
across jurisdictions.  Some maintain that information on inmate
movements after release (such as the police arrest data) is the
only area of interest for sharing information across
jurisdictions.  Others either see no need to share information
across jurisdictions, or think the task is virtually impossible
without a really thorough understanding of the definitions and
content of the information.  Still others are more expansive in
their views about the need to share information with departments
in other States and the need to conduct comparisons across
States.  -- We get tons and tons of questions from other States
about the number of offenders who have some characteristic, and
having data from those States would facilitate comparisons--  is
an example of this perspective.

The types of linkages most frequently cited were those to their
counterparts in other corrections departments.  While one
official noted that electronic linkages to share data would be
valuable, he stressed that direct connections with the human
resources of information systems were most crucial for him.  He
would like to have e-mail and telephone contacts with other
information systems officials so that he could ask simple but
very important questions about creating statistical information.
Having contacts in other departments to discuss questions such
as: --What do you do?-- or --How do you create that measure?-or
--What data are in that other data base?-- or --Who is the best
person to talk with in your system?-- are extremely valuable in
his view and the view of some other respondents.

Corrections departments link databases in a variety of ways. The
most advanced types of linkages occur when different agencies
share data systems.  Some departments are decentralized, but
have some form of communication system to link systems across
facilities.  These links are generally through advanced
communications systems such as LANs or WANs. But in some cases
the connections among facilities involve sharing the most recent
updates on diskettes or by fax.  In few departments, databases
are linked by giving users from other departments query-only
access.

The primary links in the departments' information systems are
for users at workstations in the system (correctional officers,
counselors, and personnel in the business offices).  These
officials are given routine access to the database tracking
offenders.  A considerable number of departments (23) have no
links to other agencies outside of the corrections system
(secondary links).  But a majority (28) have connections to at
least some parts of other agencies' database, typically on a
query–only basis (not shown in a table). For example, beginning
in 1993, the New Jersey Department of Law and Public Safety,
together with the Office of Telecommunications and Information
Systems (OTIS), the Administrative Office of Courts (AOC), and
the Department of Corrections (DOC) implemented a plan to
improve overall offender tracking whereby each could access the
other's independent systems and routinely update selected parts
of records1  (Footnote 2:  New Jersey Department of Corrections.
Computerized law enforcement systems in the Department of
Corrections.  (Trenton, NJ: Department of Corrections, Office of
Policy and Planning, 1994).  Most links to agencies, however,
are not through electronic means but through hard copy reports
or extracts of tapes.

The importance of working toward a goal of integrating data from
all criminal justice agencies--including corrections, probation
and parole offices, the courts, and the police--into one
comprehensive information system for users in all these agencies
was stressed by many corrections information officials.
Officials also made many other recommendations for improving
corrections information systems capacities to respond to
statistical queries.  These include: creating common
definitions, unique identifiers, and other standard formats for
linking records across agencies; converting systems currently on
mainframe (especially state-wide systems) into client-server,
stand-alone systems; transforming departmental systems into more
tightly-centralized operations; and integrating all in-facility
computer functions using one server or platform. 

Respondents stressed that existing systems have a good history
of service.  But they also think information systems need to be
much more flexible if they are to respond adequately and
efficiently to the volume of requests for data and information.
Many asserted that corrections information systems are in
overall need of improvements, and better linkages, to meet the
challenges of corrections change and to keep pace with
technological change. 

--------------------------------------------------------------
Internal capacities to extract and link archival records
--------------------------------------------------------------

The number of times that groups of offenders behaved in certain
ways is often an important focus of statistical questions.  For
example, questions about the number of infractions committed by
offenders having certain characteristics may involve extracting
historical records and linking them to the current
case-management records. "What is the average number of
disciplinary infractions committed during the first year of
imprisonment for offenders who entered during 1990 and stayed at
least one year, and how does that compare to the same statistic
for offenders who entered during 1995, after a new reform was
implemented?"  is a concrete example.  Information systems may
be structured so that individual records of disciplinary actions
are stored separately from the records related to current
information.  Before the averages can be computed to answer this
question, the information about past disciplinary infractions
for offenders who had more than one infraction needs to be
extracted from the historical record and linked to the current
records.

A large number of important corrections events have relevant
histories.  Several of them include offenders' commitments into
correction facilities, movements within a jurisdiction or
transfers between jurisdictions, behaviors constituting
misconduct or infractions, and behaviors on release in the
community.  Behaviors of offenders on release in the community
are particularly important for impacts of corrections policy on
public safety.

Departments vary in their capacities to store, retrieve, and
link data about these events.  Many keep all information about
these repeatable events on-line for offenders currently under
correctional authority.  Others also store these data and have
the capability to retrieve and link this information
electronically.  In general—for information about prison
commitments, behavior in prison, and prison releases—most
departments either store on-line histories of these repeatable
events or have the capacity to link archived records of these
events.

Forty-six departments maintain an on-line history of an
offender's commitments into prison.  Thirty-one archive
commitment histories, and of these, 28 departments have the
capability to retrieve and link electronically the archived
records (table 6.3).  With respect to information about an
offender's post-commitment movements, almost all departments
(49) maintain this information on-line, while about half of
these also archive this information, and 22 of this group of 26
have the capacity to retrieve and link the archived data
electronically.  In other categories of repeatable events, many
departments either store the information on-line or have the
capacity to link archived records.  All departments maintain
records of previously released offenders, with the majority (44)
keeping these data permanently available on-line.

On data about behavior on release, 40 departments maintain data
elements on the reasons for termination of supervision, and 38
departments obtain information about the new crimes committed by
offenders who were released in the community.  Of these 38, most
(28) obtain this information about new crimes only after the
offender returned to prison.

The capacity of corrections information systems to store,
retrieve, and link data on the supervising offenders stage may
be related to the organization of corrections in particular
states.  Of the 12 departments that do not record information
about crimes of offenders on release in the community, many are
"prison only" systems (Footnote 3:  Vital Statistics, American
Correctional Association, 1994.)  Other departments may be
integrated corrections systems that utilized an information
system other than the corrections information system to record
information about offenders on release in the community.  For
example,  in the State of Maryland, corrections is a division
within the Department of Public Safety and Correctional
Services.  The corrections division maintains data elements on
offenders in Maryland's prisons, while the Division of Parole
and Probation maintains data elements on offenders released into
the community.

--------------------------------------------------------------
Capacities to provide statistical information
--------------------------------------------------------------

In general, the overall ability of a corrections department to
provide statistical information depends upon the capabilities of
each of several components of its information system.  These
capabilities are organized into five categories:

* Legislative and institutional

* legal restrictions on access or use of data

* legislative reforms that affect operation of the information
system

* institutional requirements

* Hardware, meaning the computer system that maintains software
and data

* storage capacity

* capacity to process data

* ability to access historical data

* reliability (amount of downtime)

* Software, meaning programs that operate on the data (whether
these were developed from standard programming languages,
purchased off the shelf, or specific routines designed for
specific tasks)

* capability of existing software

* capability of existing query language

* ability to integrate data from separate files

* ability to integrate data from separate databases

* ability to structure data files

* Staffing, from data entry to management staff

* number of current programming staff

* lack of in-house programming staff

* experience level of programming staff

* ability to provide adequate training for staff
availability of funding to upgrade systems

* Data, including collection of data elements and the data
stored on each

* completeness of coverage for each data element

* accuracy of data for each element

* timeliness of data (Footnote 4: A copy of the questionnaire is
provided in Appendix B.)2

Problems that arise in any one of these areas can affect the
capabilities of information systems to provide statistical
information.  Conversely, strengths in one component of an
information system may be used to overcome deficiencies in
another. 

The Obstacles survey asked departments to rate each component on
a scale ranging from 1 (not at all) to 5 (critical problem). The
most severely rated obstacle to providing statistical
information is the number of analysis and programming staff
(table 6.4).  Across the 52 reporting departments, it receives a
mean score of 3.9 (on a scale of 1 to 5) and the least variation
around its mean.  Funding for systems upgrades, modifications,
or staffing are the second most severe barrier.  Departments
tending to experience this obstacle in a relatively severe
manner, as reflected in a mean ranking of 3.8 and relatively
little variation around the mean.  Other obstacles that rank as
relatively severe problems by the departments include: lack of
in-house programming staff, inability to provide adequate
training for staff, inability to integrate data from separate
databases, and the low experience level of programming and
analysis staffing.  Two additional obstacles  -- the accuracy of
the data and integrating data from separate data files  -present
somewhat of a barrier. 

Another eight obstacles present less severe barriers. These
items receive mean ratings of between 2.5 and 2.9 and include:
the data completeness, legislative reforms, the structure of
data files, capability of the query language, data timeliness,
statistical software capabilities, ability to access historical
data, and institutional system requirements, and there was
greater variability around the means.

Finally, four -- legal restrictions on data, capacity to process
data, storage capacity, and system downtime -- present
relatively minor barriers to departments, as reflected in their
average rankings of 2.4 or less.

The grouping of individual items into the five major obstacle
categories is shown in table 6.5 with the mean category score
and severity ranking of each.  The group averages range from 3.6
for staffing-related, the most serious group of obstacles, to
2.2 for hardware-related obstacles, the least serious set of
obstacles.  Software and data problems each average ratings of
2.9.  Institutional arrangements -- legislative reforms,
requirements to use specific hardware, and legal restrictions of
the use of data -- are rated at 2.6, on average.

Staffing related issues present severe obstacles, as the 3.6
average score for these obstacles indicates. The individual
items within the staffing group indicate that the number of
staff (too few), ability to provide for their adequate training,
and availability of funding for new staff all approach very
severe levels for the departments.  Software problems, such as
the capacity of query languages or of statistical software, and
data problems, such as the timeliness, completeness, and
accuracy of data elements, also present relatively severe levels
of barrier.

Staffing, software, and data are interrelated.  Having access to
sophisticated database software, query languages, and
statistical packages is not enough if a department lacks staff
trained in the use of these technologies.  Staff that are
knowledgeable in the use of these software tools but lacking
access to them cannot use their skills to produce statistical
information.  And having sophisticated software and data entry
procedures is not enough if staff are not adequately trained in
data collection, data entry, and other data preparation tasks.  

The reported deficiencies in the number of staff, lack of
funding for system upgrades, modifications (software problems),
and staffing skills combine to suggest that the primary obstacle
to overcome is lack of resources.  Additional resources will
allow departments to overcome these deficiencies with staffing
shortages, training deficiencies, and system inadequacies as
they see fit -- with maximum impact on their overall capacities
to provide statistical information.

----------------------------------------
Varying obstacles among departments
----------------------------------------

Departments vary in the severity of the obstacles they confront
(tables 6.6 and 6.7).  Individual staffing obstacles rate an
average of 5, the critical level, by 8 departments.  An
additional 20 departments rate staffing obstacles between 4 and
5 on average, indicating a very severe obstacle.  Only 6
departments rate staffing obstacles at 2 or less on average.

Software problems average slightly lower severity than staffing
obstacles.  Only two departments rate the 5 software obstacles
at the critical level, and 12 departments rate them as very
severe.  An additional 20 departments rate them as a moderate
obstacle.

Four departments rate the 3 data obstacles at a critical level,
on average, and an additional 8 departments rate them as very
severe.  Eighteen departments rate the severity of data
obstacles as being very little to none.

Legislative and institutional obstacles do not provide major
barriers but they can not be ignored either.  No department
rates these problems as critical, but  five departments rate
legislative and institutional obstacles as very severe and 25
rate them as a moderate obstacle. 

No department rates hardware problems at the critical level, and
one has very severe hardware problems.  Twenty-six of the
responding departments rate hardware obstacles as having very
little severity.

Staffing, software, and data problems tend to go together
(Footnote5:  5The clustering of obstacles within groups of
departments was addressed in more detail in a preliminary
analysis based on factor and cluster analyses.  The results from
these analyses confirm the observations here that staffing,
software, and data problems tend to cluster together, and that
different groups of departments experienced different degrees of
each cluster of obstacles.)  For the 46 departments that rate
staffing obstacles an average of 3 or higher, most also rated
software and data problems at average severity levels above 2
(table 6.6).  Of the same 46 departments, only 16 rate either
software or data problems at an average of 2 or lower. 

There are exceptions.  In Florida and North Carolina, for
example, staffing obstacles are reported to be more severe
(averaging 3.6 and 3, respectively) than software, data, or
hardware (2 or less on average).  North Carolina recently
completed a major redesign of its correctional information
system to improve its capabilities.  One of the models North
Carolina used to redesign its system was the information system
developed by the State of Florida.  Both departments indicate
that while staffing problems tend to be accompanied by software
and data problems, information systems with the best designed
software, good data, and advanced hardware can still confront
major barriers in providing statistical information.

-------------------------------
Summary
-------------------------------

Officials in corrections information departments report that
they routinely respond to requests for raw data and summary
information on offenders.  They also receive requests that
require analysis and processing of data elements into specified
formats to meet external definitions and standards.  Departments
use a variety of media to submit data, including hard copy,
tape, diskettes, or file transfer protocols.  In addition to
corrections staff users, requesters of corrections data include
Federal agencies, a wide range of State and local agencies,
researchers, and private companies. 

High-availability data varies widely among the departments and
all of the four stages of corrections processing.   No
department has all the data elements in high-availability form,
nor does any department have all of the elements that correspond
to each stage of corrections processing.  The stages related to
committing offenders and managing offenders have the most
departments with relatively high data availability index scores.
Twelve departments rate at 70% or above for all 72 data elements
in the committing offenders stage, and 10 departments in the
managing offenders stage rate above 70%.  The supervising
offenders stage has nine departments that collect data about
released offenders rated at above 70%, and seven departments in
the profiling offenders stage scored higher than 70% on the
index.

The information most available in high-availability form are
data that describe offenders' demographic characteristics,
conviction offenses, sentences imposed, current commitment,
expected time to be served, risk assessment, classification and
confinement decisions, post-commitment movements, good time and
other sentence adjustments, releases from custody, reasons for
terminating supervision, and the criminal justice response to
supervision violations.  In general, the information with the
lowest level of high-availability are data describing offenders'
socio-economic status, family characteristics, the criminal
incident, victim information, medical care, and employment and
residence information about released offenders.

To answer statistical queries, departments frequently need to
construct links among internal databases or between them and
databases of other agencies.  Commonly mentioned obstacles in
creating links to other data systems include: existence of
several platforms of different ages and data formats, which
makes interfaces between them complex; lack of common
definitions, unique identifiers, and other standard formats for
linking records across agencies; and outdated systems that do
not respond readily or flexibly to queries for information.
Corrections staff also frequently noted the importance of
working toward a goal of integrating all criminal justice
agencies into one comprehensive information system that would be
shared by all users.

A key issue for corrections information systems is the extent to
which departments can record data on event histories.  For
example, can an information system record the number of times an
offender commits an infraction, or the number of times an
offender enters and exits prison during a single prison term?
Data on such "repeatable events" may be required information for
measuring corrections system performance.  Departments generally
reported that their highest capacities for storing, retrieving,
and linking archived data on repeatable events for data elements
related to commitments into prison, post-commitment movements,
and releases from prison during a term.

Departments report that they do indeed confront several
obstacles in producing statistical information.  These range
from the need to reformat their data to comply with standards
and formats of the requester, through hardware limitations that
restrict the capabilities for executing queries, and software
limitations that require departments to create customized
programs to generate reports and data, to shortages of
experienced staff that prevent timely resolution of data
requests.

Responses to the Obstacles survey confirm that the most serious
obstacles encountered by corrections departments in producing
statistical information are staffing-related obstacles—including
the number of analysis and programming staff, their experience,
and the resources to further train them.  Staffing-related
obstacles are closely related in severity to software
applications and data constraints.  Forty-six departments rated
staffing obstacles as providing a serious barrier to their
providing statistical information.  All but 16 of these also
rated either software or data as serious constraints. Relatively
few departments rate either hardware or legislative and legal
factors as serious barriers to producing statistical information.


----------------------------------------
Chapter 7
An empirical common core
----------------------------------------

Highlights

* Departments maintain a common core of data about 14 dimensions
of corrections processing that contain 100 of the 207
offender-based data elements in this Inventory.

* For the 100 core elements, most departments (48) rate above
50% of full availability—the extent to which they maintain core
data elements electronically for more than 75% of offenders. 
Eight departments rate above 90% and 29 rate above 70%.

* Thirty-nine departments rate above 70% of full availability in
the profiling offenders stage, as do 35 in the committing
offenders stage, 22 in the managing offenders stage, and 18 out
of the 40 that maintain data on offenders released into the
community.

* Within non-core dimensions, there are 15 data elements that
more than 26 departments maintain in high-availability form.

--------------------------------------------------------------------
Commonly maintained, high-availability data elements
--------------------------------------------------------------------

Departments of corrections currently maintain a common core of
data about 14 dimensions of corrections processing.  These
dimensions describe several aspects of who offenders are, what
they have done, how they arrive in prison, how they are managed,
and what happens to them upon release.  The common core is based
on the dimensions of corrections processing for which most
departments maintain data in electronic form for most offenders.
 Each core dimension contains several data elements.  All of the
data elements in a core dimension are included in the common
core of data elements.  To say that a common core of data exists
for a given dimension does not mean that every data element in
that dimension is maintained in a high-availability form by
every department.  It means that, within a dimension, enough
data elements are maintained by a majority of departments in
high-availability form to constitute a meaningful core of
information measures for a given dimension.  Dimensions
determine the core because many corrections concepts are best
measured by several data elements.  The high-availability
standard (maintained electronically for more than 75% of
offenders) reflects the form of data that can most readily be
analyzed, shared electronically, and processed into the types of
statistical information that measure corrections performance.

Data elements in 14 of the 28 dimensions of corrections
processing are included in the common core (table 7.1).  Each
stage of corrections processing has at least one dimension in
the common core, and a total of 100 of the 207 offender-based
data elements fall within the dimensions that comprise the core.
By stage, the core dimensions include:

* Stage 1, profiling and identifying offenders:  demographic
characteristics (11 elements);

* Stage 2, committing offenders:  conviction offenses, sentences
imposed, current commitment, expected time to be served, risk
assessment, classification decisions, and confinement
characteristics (42 data elements); 

* Stage 3, managing offenders:  post-commitment movements,
good-time and other sentence adjustments, offender registry, and
releases from custody (25 data elements); and

* Stage 4, supervising offenders:  behavior on supervision and
responses to violations of conditions of supervision (22 data
elements).

The common core for stages 1, 2, and 3 is based on the responses
from all 52 departments.  The common core for stage 4 excludes
the 12 departments that do not use their adult, sentenced
prisoner information systems to collect data on offenders
released into the community; it is based on the responses of the
40 departments that do maintain data elements on released
offenders.

The common core describes offenders' demographic characteristics
such as age, sex, race, and country of origin.  It contains data
elements about commitments, convictions, sentences, and offenses
that describe how and why offenders arrived in prison.  The
common core also describes classification, confinement, and risk
assessment decisions, and it contains data elements that
describe how long offenders can expect to stay in prison.

In the management of offenders stage of processing, the core
describes reasons for changes in sentences and changes in
expected length of stay; it also describes offenders' movements
and releases from prison.  The core also includes limited
information about victim notification requirements.  Finally,
the core dimensions of the supervising offenders stage include
data elements that describe the form of supervision, the reasons
for termination of supervision, whether a new crime was
committed by an offender on supervised release, and if so,
whether an offender was arrested, convicted, sentenced, and
returned to prison.

Of the 100 core data elements, 8 are collected by all
departments, 60 are collected by more than 70% of departments,
and only 9 data elements are collected by fewer than 50% of
departments (not shown in a table).  The common core data
elements that are collected by all departments include transfer
to another facility, date of transfer, type and date of release,
sex and race of offender, and type and date of commitment.  The
60 data elements that most departments collect include nearly
all of the data elements that describe demographic
characteristics, sentencing, time to be served, classification
decisions, confinement characteristics, post-commitment
movements, good time and releases.  The nine data elements that
are less commonly collected are the offender's psychological
index, who authorized a prison movement, amount of change to
sentence, amount of change to expected release date, special
credits to sentences, if offender actually registered as a sex
offender, if and when a release violation was adjudicated, and
the date a released offender is rearrested (not shown in a
table).

-----------------------------------------------------
Full-availability ratings for common-core data
-----------------------------------------------------

Assessed on the extent to which they maintain all data elements
included in the common core in high-availability form, 29
departments rate at greater than 70% of full availability, and
only four departments rate at less than 50% (table 7.2). 
Colorado maintains all 100 core data elements in
high-availability form, and seven other departments maintain
more than 90% of their core data elements in high availability. 
Departments that rate less than 70% on the availability index
generally maintain many core data elements in paper form, or do
not collect a majority of this information.

For the 11 data elements in common core of Stage 1 (profiling
offenders)—the demographic characteristics of offenders—39
departments receive full-availability ratings of greater than
70%, and 9 have full-availability ratings of 100%.  Only 4
departments rate at less than 50% of full availability (table
7.3).

In Stage 2 (committing offenders), 35 departments receive
full-availability ratings above 70%; and 2 have
full-availability ratings of 100% for 42 core data elements. 
Only 4 departments receive full-availability ratings of less
than 50%.  

For the 25 core data elements in the third stage (managing
offenders), 22 departments receive full-availability ratings
greater than 70%.  Twelve received full-availability ratings of
less than 50%. 

Of the 40 departments that maintain 22 core data elements in the
Stage 4 (supervising offenders on release), 18 have
full-availability ratings greater than 70%.  Eleven departments
rate at less than 50% of full availability.

Departments maintain core data elements in high-availability form

Within each stage of processing, relatively few core data
elements are maintained in low-availability forms, and there are
few data elements that are not collected.   Only 12 departments
maintain fewer than 7 of the 11 demographic data elements in
high availability form, but 9 have all 11 data elements in this
form, and 20 collect all demographic core data elements
(Appendix H).  In the committing offenders stage, half of the
departments maintain high-availability core data for at least 30
out of the 42 data elements.  Colorado and North Carolina
maintain all core commitment data elements in high availability.
In the managing offenders stage, 2 departments—Colorado and
Ohio—maintain all 25 core data elements in high availability,
and an additional 12 departments maintain at least 20 high-availability
data elements.  Four departments collect all core management
data elements in some form.  In the supervising offenders stage,
8 departments out of 40 collect all 22 data elements in
high-availability form.  Seventeen departments collect all core
supervision data.

----------------------------------
Common definitions 
----------------------------------

Many corrections departments have data elements needed to
generate statistical information on a core set of information
issues.  Of the 40 departments that collect data on all 4 stages
of corrections processing, almost all are above 50% availability
for the 100 core data elements (table 7.3).  Colorado is at 100%
availability for all core data elements.  Departments maintain a
large number of data elements, and they do so in a
high-availability form that facilitates processing, analyzing,
and sharing the data maintained on these elements.  Not
surprisingly, the common-core data reflect information issues
that revolve around the day-to-day management concerns of
corrections.  These day-to-day management concerns include many
important pieces of information that are related to corrections
performance. 

For example, counts of offenders who enter prison, complete
their stay without incident, leave prison for supervision, and
complete supervision without incident are fundamental for
corrections performance indicators.  The ability to provide such
counts for subpopulations of offenders and according to criminal
justice processing variables (such as type of offense, or length
of stay) enhance the comparability of performance indicators. 
Further, the capability to measure duration of supervision and
returns to prison for new crimes or violations of conditions of
supervision lie at the basis of developing measures of
recidivism.  This is true regardless of the controversies
associated with interpreting a recidivism rate as a "good" or
"bad" indicator of performance.  In short, the capacities of the
departments to maintain a large volume of common data elements
about offenders is quite high.  This is particularly impressive
given the wide variation in corrections organization, legal
frameworks, and penal codes operating in the States, the Federal
government, and the District of Columbia.

Findings about common-core data elements do not necessarily
imply that all departments define the elements in exactly the
same way or collect exactly the same pieces of information in
the elements.  For example, several of the Inventory questions
asked departments to indicate the response categories (such as
type of program in which an offender participated) for data
elements they collect.  For some data elements, departments
collect different categories of information.  But for other data
elements, (such as type of commitment or method of release from
prison), there is greater agreement.  For the latter,
differences in response categories indicate a greater or lesser
degree of precision rather than differences in scope or
definition.

----------------------------------------------
Non-core high-availability data elements
----------------------------------------------

Within several core dimensions, there are some data elements
that are collected by fewer than 26 departments.  Conversely,
within several non-core dimensions there are data elements that
a majority of departments maintain in high-availability form. 
Out of the 100 core data elements, 20 are not maintained in
high-availability form by most departments; out of the 107
non-core data elements, 15 are maintained in high availability
by most departments. 

There is one data element in the core of the profiling offenders
stage that a majority of departments do not maintain in high
availability: offenders' citizenship status.  On the other hand,
there are two data elements in this stage that are not included
in the core but which a majority of departments maintain in high
availability form: Education level and marital status (table
7.4).

Among the common core in the committing offenders stage, there
are 8 data elements that are maintained in high availability
form by fewer than 26 departments (table 7.5).  These include:

* Charges on the indictment;

* The reason for a mandatory sentence;

* If monetary sanctions were imposed;

* Amount of sanctions imposed;

* If a weapon was used in the offense;

* Gang membership;

* Psychological index or score; and

* Medical index or score.

Also in this stage, there are three data elements that are not
included in the core but that a majority of departments maintain
in high-availability form: the criminal justice status of the
offender, the medical condition of the offender at admission,
and the types of need as determined by an assessment.

In the managing offenders stage, seven of the data elements
included in the core are not maintained by a majority of
departments, including: 

* Who authorized the transfer;

* Who authorized the move;

* Amount of change to sentence;

* Changes to release date;

* Special credits;

* If offender is required to register as sex offender; and

* If offender actually registered as sex offender (table 7.6).



On the other hand, there are six data elements outside of the
core for managing offenders that a majority of departments
maintain in high-availability form.  These include:

* Type of program participation;

* Date of program participation;

* Type of misconduct in prison;

* Date of misconduct in prison;

* History of misconduct in prison; and

* Disposition of the charges of a legal proceeding (table 7.7).



Within the supervising offenders stage, 4 of the data elements
in the core are collected by less than 26 departments.  Three
are in the responses to violations dimension: if offender was
adjudicated, date adjudicated, and date the offender was
arrested (table 7.7).  The type of technical violation is also
less commonly collected in high availability than the other
elements in the dimension describing violations committed after
release.  There are three non-core data elements in this stage
that are collected by a majority of departments: the address of
the released offender, if the offender was under supervision
when the crime was committed, and the type of the new offense
committed.

----------------------------------
Data elements about facilities
----------------------------------

Fifteen of the Inventory's survey questions are about
facilities, costs, and availability of medical care.  These data
are not collected by most departments in electronic form and are
not considered part of the core.  They generally are maintained
in paper form.  The number of facilities and the number of beds
per facility are the only data elements that are maintained
electronically by more than 70% of the departments.  Only 12
departments maintain data electronically about operational
costs, annual capital costs and facility-generated revenue. 
Fewer than half of the departments collect information in any
form about the number, qualifications, or availability of  their
medical staff.  

----------------------------------
Conclusions
----------------------------------

Of the 207 offender-based data elements in this inventory, 100
of them are included in the core 14 dimensions of corrections
processing.  Most of the common-core data are found in areas
related to committing offenders into prison (7 dimensions) and
managing offenders in corrections facilities (4 dimensions). 
For the departments that collect data on released offenders, 2
dimensions are included in the common core.  The profiling of
offenders has one dimension in the core.  

Within these core dimensions, 20 of the 100 data elements are
not maintained in a high-availability form by a majority of
departments.  Of the 107 non-core data elements, 15 are
maintained in high-availability form by most departments. 
However, these high-availability non-core data elements do not
measure an entire dimension of corrections processing, as do the
data elements currently in the core.  

Several implications for the objective of creating
commonly-defined indicators of corrections performance flow from
these findings.  First, the commonality with which departments
collect and maintain data in high-availability form does not
necessarily (and need not) imply that departments define these
commonly-collected data elements in exactly the same manner, or
that they apply to exactly the same groups of offenders among
the departments.  Nor should it be expected that the data
elements necessarily be defined in exactly the same manner among
all departments.  State penal codes and criminal procedures
vary.  And a department's adherence to its jurisdiction's laws
and regulations rightly takes precedence over defining a group
of data elements in the same way that other departments define
them, or applying them to the same groups of offenders.  

Second, there are important areas in which no common-core data
currently exists.  Under-covered areas in Stage 1 (profiling
offenders) include information about the social and economic
characteristics and family relationships of offenders.  These
variables can be used to indicate the degree to which offenders
maintain connections with mainstream institutions outside of
prison.  In Stage 2 (committing offenders), under-represented
areas include the criminal incident leading to the
conviction—especially victim-related information, criminal
history, and offender needs assessments.  Within Stage 3
(managing offenders), the under-covered areas include program
participation, drug testing, medical care,  misconduct and
infractions, and the corrections system responses.  The drug
testing information is especially important for developing
indicators of how well corrections institutions keep offenders
involved in maintaining positive behaviors.  Finally, Stage 4
(supervising offenders in the community), data elements about
offenders employment experiences, about new crimes they commit
on release and about the victims of the crimes is not widely
collected.

Although the methods to increase coverage of data elements may
vary from stage to stage and dimension to dimension, there are
two basic strategies to address the problem of under-covered
data elements: (1) departments may collect the data themselves;
or (2) departments may rely on other agencies to collect it and
then obtain the data or access to it in electronic format.  If
the strategy of relying on another agency to collect data
elements is used, then departments must solve both technical
problems related to transferring and linking data, and the
substantive problems associated with defining the data elements.
 Obtaining data from other agencies may also increase the
staffing requirements for corrections information systems
departments, perhaps by requiring staff with a higher level of
skill.

If departments rely on other agencies to collect data elements,
they may have to expand in both directions.  For example, to
obtain data on offenses and victims leading to a conviction,
corrections departments may have to develop links with
prosecutors or the courts.  Conversely, to develop additional
capacities in Stage 4 (supervising offenders on release),
departments may have to develop links with parole departments. 
In either case, the complications associated with using other
agencies' data exist and may be compounded.

Third, there may be important data elements beyond those in this
survey, and these may suggest areas to expand coverage of data
elements.

Fourth, the capacities of departments' information systems to
provide statistical information are constrained by staff,
software, and to a lesser degree, data.


----------------------------------
Chapter 8
Using the Inventory report
----------------------------------


The purpose of the Inventory project is to provide a basis for
improving the quality of corrections data and enhancing
electronic sharing of information.  This report identifies the
capacity of corrections departments to provide comparable data
for performance measures and for cross-jurisdictional research. 
It describes existing information systems, but does not
recommend a model system for all departments or develop a
strategy for future actions.  The report identifies a common
core of data elements that most or all departments collect;
describes and analyzes the obstacles departments face in
responding to statistical inquiries; and describes departments'
capacities for sharing and linking data internally and
externally.  Additionally, the report provides a list of
respondents (Appendix C) that may be used by departments or
researchers to obtain information or assistance.

This report may be used--

* by departments for expanding data collection.  Departments may
use information about the availability of the common-core data
elements to develop priorities for adding data elements and
improving the availability of existing data.

* by departments to assist in their ongoing information system
redesign activities.   Departments in the process of modifying
their information systems may use the report to identify
commonly collected data elements and to understand how
departments differ in their capacities to maintain data in
electronic form.

* by research directors and other corrections researchers to
determine availability of data elements in cross-jurisdictional
studies.  In designing comparative studies, researchers may use
the report to identify the reporting capabilities of
participating departments

* by ASCA members to develop strategies for establishing
performance measures.  ASCA members may use the report to
develop more specific priorities about measuring corrections
performance, to identify indicators based on commonly collected
data elements, and to decide what additional information is
needed for these performance measures.

----------------------------
Expanding data collection
----------------------------

Departments may use information about the availability of data
elements to develop priorities for expanding their data
collections.  Data collections may be expanded by adding data
elements and by improving the availability  --  their storage in
electronic medium  -- of data elements.  Departments using the
Inventory in developing priorities for expanding data collection
may wish to consider several related issues.

An advisory committee established priority information areas. 
The 207 offender-based data elements in the Inventory were
derived from the six priority information areas that the
project's advisory committee identified.  These six areas 
--offender profile, recidivism, program effectiveness, internal
order, public safety, and operational costs -- cover the scope
of corrections processing and reflect important corrections
management outcomes.  

The Inventory shows what departments have.  The Inventory shows
which of the 207 offender-based data elements departments
collect and how they maintain the data elements.  It shows which
data elements are more commonly collected and which are
collected by fewer departments. It permits departments to
compare their data collection with other departments.

The Inventory's common core is an experiential core.  The common
core of 100 data elements reflects what departments currently
collect and not necessarily what they should collect. 
Departments wishing to use this experiential core in
establishing priorities should recognize that expanding
collections up to the existing core will increase the
concentration of departments that collect core elements; but it
will not necessarily expand the scope of the common core.  To do
this, departments should consider the entire set of 207 data
elements and the six priority information areas.

High-availability formats facilitate sharing data.  Maintaining
data electronically can facilitate sharing information.  This
important objective can be met by expanding collections to
increase the number of data elements that are maintained in
electronic format.

Cross-agency linkages may be a way to obtain additional data
elements.  Departments may wish to consider developing
cross-agency linkages with other information systems as a method
for adding data elements.  In departments for which core data
elements are beyond the scope of the information system that
they use to manage adult sentenced prisoners, electronic
linkages with other agencies may provide a relatively
inexpensive method for gathering data or additional data
elements.

------------------------------------
Redesigning information systems
------------------------------------

The Inventory may help departments establish priorities for
upgrading their information systems.  It shows areas where
improvements are needed in data collection and maintenance, and
where problems are faced by many corrections departments in
reporting statistical information.  The results of the obstacles
survey do not suffice as or replace any internal audit of an
information system, but they can point to areas where
departments may want to concentrate efforts in MIS
re-engineering.

Establishing cross-agency linkages and offender tracking
systems.  As part of an MIS upgrade or independently of such
efforts, departments may wish to consider developing more
cross-agency linkages and better systems for tracking offenders.
 Such efforts may be undertaken in a variety of ways.  At a
simple level, data extracts can be shared on diskette, tape, or
physically transferable media.  At a higher level, the
capability to query another agency's database could be
established.  At a higher level still, agencies can participate
in an offender-based tracking system (OBTS).  An OBTS allows a
participating agency direct access to the data for which they
have collection responsibility, but only permits them to gain
access to obtain data in the system through a specific request
to information systems staff or staff from the agency with
collection responsibility.  At the highest level, agencies could
participate in an integrated information system (ITS) that
permits sharing of all automated data among all participating
agencies.  At the levels of sharing below those of the OBTS or
ITS, the major problem lies not in sharing information per se
but in linking it and ensuring that data elements are defined in
the same way among information systems.  Linking records is
greatly facilitated if all the agencies involved use a common
identifying number.  If that is not feasible, other methods
could  be developed to link records.

----------------------------------
Cross-jurisdictional research
----------------------------------

Corrections researchers may use the Inventory to help to
identify research topics for and potential barriers to
conducting cross-jurisdictional research.  In a survey of
research units in departments of corrections, researchers
identified several important topics for comparative research
(Footnote 1: Association of State Corrections Administrators
Subcommittee on Research.  Cross-jurisdictional survey of
correctional research offices, summary of findings: Final
report, Vol. 1. September 1995. (Prepared by the Pennsylvania
Department of Corrections, Office of Planning, Research, and
Statistics.)  These include: studies on recidivism, alternatives
to prison, sentencing structures, and evaluations of corrections
programs and policies.  Many of these topics1 are reflected in
the experiential core of data elements that currently are
commonly collected.  Researchers interested in topics that are
not reflected in the core can use the Inventory to design
research and plan data collection activities.  As the
availability of data elements in electronic form and the
resources to prepare extracts or research datasets pose
potential problems for conducting research, the Inventory can
show when and where these are likely to occur.  Researchers can
use this information to plan the scope of research and to learn
about data systems.

----------------------------
Performance measures
----------------------------

The Association of State Correctional Administrators has
expressed an interest in developing and using corrections
performance indicators to describe, measure, and compare the
management of corrections populations.  For the following
reasons, that goal is beyond the scope of this Inventory
project.  First, performance indicators in general are tied to
the mission, goals, and objectives of organizations. 
Comparative corrections performance indicators would have to
consider the varying missions, legal structures, and
organizational arrangements of corrections departments
throughout the country.  The standardization of measures that
take such factors into account is extremely complex.  Second,
standard or traditional approaches to measuring corrections
performance, such as those that use the crime-rate related
concepts of recidivism, deterrence, and incapacitation, are
difficult to measure and interpret.  More importantly, these
indicators establish a standard for corrections performance that
is based on what happens outside of prison or beyond the scope
of corrections supervision.  For example, an offender on release
in a community is subject to many factors that are beyond the
control of corrections.  Even if this offender commits a crime
while under supervision, the measure of the recidivism rate is
related to the performance of the police, prosecutors, and
judges in apprehending, convicting and sentencing offenders.

Third, alternative approaches to measuring corrections
performance, such as those proposed in several papers in the
Bureau of Justice Statistics Performance Measures for the
Criminal Justice System, provide a useful starting point for
developing corrections indicators that are tied to specific and
shared corrections goals.  These alternatives limit the mission
and goals of corrections to the activities and outcomes that are
within the scope of control of corrections.  For example, in his
article on "Criminal Justice Performance Indicators for
Prisons," Charles Logan (Footnote 2:  Logan, Charles H., Ph.D.,
"Criminal Justice Performance Indicators for Prisons," in
Performance Measures for the Criminal Justice System. 
Washington, DC: Bureau of Justice Statistics, NCJ-143505, 1993:
pp. 19-59.)  develops a series of measures for prisons that are
tied to a confinement model of prisons.  In this model, Logan
identifies the mission of prisons as "keeping prisoners,"
keeping them in, safe, in line, healthy, and busy and doing this
without undue suffering and as efficiently as possible (Footnote
3:  Logan, p. 25.)2  From this, he derives measures of
performance that are related to security, safety, order, care,
activities, justice (as fairness), conditions (without undue
suffering), and management.  Each indicator can be tied to the
effort of corrections officials.  

Similarly, in her article about community corrections in the
same volume, Joan Petersilia echoes many of Logan's sentiments. 
Petersilia argues that performance indicators for community
corrections should be based on: (1) an articulate mission
statement for community corrections; (2) a clear statement of
the goals contained within the mission statement; (3) specific
methods or activities that address each goal; and (4) measurable
indicators of performance for each goal.  (Footnote 4: 
Petersilia, Joan, "Measuring the Performance of Community
Corrections," in Performance Measures for the Criminal Justice
System.  Washington, DC: Bureau of Justice Statistics,
NCJ-143505, 1993: pp. 61-85.) She also stresses that the
performance and success of community corrections should "reflect
only activities that occur while the offender is formally on
community corrections status, not beyond" [emphasis original]
(Footnote 5:  Petersilia, p. 74.)

While much of the work related to developing corrections
performance indicators must be done by a deliberative body that
can address the complexities described above, the Inventory may
be useful in developing indicators in several ways.

The Inventory points to areas of commonality.  The Inventory
results show that for many important areas of corrections
processing, many departments collect roughly comparable data. 
This is reason for optimism.  If many or most departments have
the raw material needed to develop performance indicators, then
embarking on an effort to measure and compare performance could
be reasonably successful.

The Inventory shows that the common core reflects experience. 
The common core of data is based on what departments currently
collect, and performance indicators may be developed from these
experiential core data elements.   While the experiential core
may show what departments can measure more easily, indicators
that are measured by data elements that fall outside of the
common core can also provide departments with guidance in
expanding data collection.

The Inventory points to the need for precise definitions.  While
many departments collect roughly comparable data in many
important areas, departments may still define data elements
differently or they may use different categories to record data
about offenders.  Comparative performance indicators need to be
defined precisely and the differences in definition of data
elements assessed.

The Inventory points to the need to look at sources of
non-comparability.   While there is much commonality in what is
collected, there are sources of non- comparability in
corrections data.  These derive primarily from differences in
definition, scope of coverage, and methods for counting and
classifying offenders.  For example, definitions of a prisoner
may differ among departments that include offenders in halfway
houses or jails, and those that exclude them.  And differences
in defining sentences confound simple comparisons of time served
or the percent of sentence served.   Further, differences in
methods for classifying offenders—e.g., by offense category,
method of commitment, or other classes of offenders—need to be
considered when interpreting comparative indicators.   Any set
of comparative corrections performance measures that are
developed would have to be assessed empirically in relation to
these and other sources of non-comparability in measurement
among departments.

(See printed version or acrobat file for copy of Inventory of State 
and Federal Corrections Information Systems survey).

----------------------------------------
Appendix C
List of respondents

Table C.  List of respondents


Department      Respondent's name                Address                                   Contact information
                Programmer/Analyst

Alabama         Jake Jacobs                          Department of Corrections                 (334) 242-9187
                                                     P.O. Box 30150
                                                     Montgomery, AL  36130

Alaska          Annette Smith                        Department of Corrections                 (907) 465-3313
                  DP Manager                         P.O. Box 112000
                                                     Juneau, AK  99811

Arizona         Chet Homan                           Department of Corrections                 (602) 542-4527
                EDP Project Manager                  1601 W. Jefferson Mail Code 310 
                Rich Camine                          Phoenix, AZ  85007
                 EDP Programmer Analyst  

Arkansas        Bob G. King                          Department of Corrections                 (501) 247-6341
                 Senior Project Leader               P.O. Box 8707 
                                                     Pine Bluff, AR  71611

California      Judy Metz                            Department of Corrections                 (916) 323-4062
                 Chief, Correctional Case            1515 S Street
                   Records Services Peter Lai        Sacramento, CA  94283

Colorado        Gary Saddler                         Department of Corrections                 (719) 540-4775
                Director of Information Systems      2862 S. Circle Drive Suite 400
                 Jerry Hunter                        Colorado Springs, CO  80906
                Data Base Administrator    

Connecticut     Edmund Hayes                         Department of Correction                  (860) 692-7667
                 Andrew P. Shook                     24 Wolcott Hill Road 
                Research Analyst                     Wethersfield, CT  06106

Delaware        Rodney Gibbons                      Department of Corrections                 (302) 739-5601
                 Ed Babowski                        Central Admin. Building
                                                    80 Monrovia Ave
                                                    Smyrna, DE  19977

District of      Thomas Hoey                        Department of Corrections                 (202) 673-2300
  Columbia        Steve Fezuk                       1923 Vermont St. NW
                                                    Washington, DC  20001

Federal Bureau  Meredith Barosso                    Federal Bureau of Prisons                 (202) 307-3065
 of Prison      Supervisory, Computer Specialist    320 First St. 
                                                    Washington, DC  20534

Florida         Paul Mauer                          Department of Corrections                 (904) 488-5963
                                                    2601 Blair Stone Road 
                                                    Tallahassee, FL  32399

Georgia         Fred Radford                        Department of Corrections                 (404) 656-4609
                Douglas Engle                       Two M.L. King Jr., Drive, SE
                Systems Manager                     Atlanta, GA  30334

Hawaii          Ken Hashi                           Department of Public Safety               (808) 587-1237
                Research Statistician               919 Ala Moana Blvd
                Mike Mamitsuka                      Honolulu, HI  96814
                Judy Yamada  
                Computer Programmer  

Idaho           Craig Potcher                       Department of Corrections                 (208) 332-8298
                 John R. Hofland                    500 South 10th Statehouse Mail
                 John Hoffman                       Boise, ID  83720

Illinois        Mike Noga, Supervisor               Department of Corrections                 (217) 522-2666
                 Offender Systems Report            1301 Concordia Ct.
                   Management                       P.O. Box 19277
                                                    Springfield, IL  62794

Indiana         Robert W. Hughes                    Department of Corrections                 (317) 232-6930 (317) 233-5400
                Jeanne McFarland                    Government Center S. 
                Systems Analyst  Supervisor         302 W. Washington St
                                                    Indianapolis, IN  46204

Iowa            John Baldwin                        Department of Corrections                 (515) 281-4807
                                                    Capitol Annex 523 E. 12th St
                                                    Des Moines, IA  50319

Kansas          Patricia Biggs                      Department of Corrections                 (913) 296-5515
                 Director, Research & Planning      900 S.W. Jackson St. 4th Floor
                 Jeff Lewis, Carlos Usera,          Topeka, KS  66612
				 Cathy Clayton
                 Programming & Analysis Supervisor  

Kentucky        Louis Smith                         Department of Corrections                 (502) 564-4360
                                                    State Office Building 5th Floor 
                                                    Frankfort, KY  40601

Louisiana       Walt Worley                         Department of Public Safety & Corrections (504) 342-8770  (504) 342-8782
                Project Leader                      P.O. Box 94304       
                Terry Clair                         Capitol Station
                                                    Baton Rouge, LA  70804

Maine           Lita Cunningham                     Department of Corrections                 (207) 287-4343
                Jerry Steeves                       State House Station 111 
                Michael Hughes                      Augusta, ME  04333

Maryland        Edmen Tausendschoen,
                  Program Analyst                   Division of Correction                    (410) 764-4107
                Lawrence Zamarski,                  6776 Reisterstown Road Suite 311 
                  Program Analyst                   Baltimore, MD 21215

Massachusetts   Lisa Sampson                        Department of Corrections                 (617) 727-8857  617) 727-2106
                Systems Analyst                     Technical Services Route 1A 
                Curt Wood,  Robert Hughes           Norfolk, MA  02056

Michigan        Terrence Murphy                     Department of Corrections                 (517) 335-1383
                 Steve Paddock                      P.O. Box 30003
                Manager of MIS                      Lansing, MI  48909
                Jeff Anderson
                Projection Specialist    

Minnesota       Dan Storkamp                        Department of Corrections                 (612) 603-0194   612) 642-0301
                Director, Office of Planning        1450 Energy Park Dr. Suite 200
                and Research                        St. Paul, MN  55108
				Mark Evenson, 
				Database Administrator

Mississippi     Audrey McAfee                      Department of Corrections                   (601) 359-5608
                 Applications Analyst Manager      723 N President St.
                                                   Jackson, MS  39202

Missouri        David Schulte Director,            Department of Corrections                   (573) 526-6452
                 Information Systems               PO. Box 236 
                 Deborah Stegman Manager,          Jefferson City, MO  65102
                 Applications Development  

Montana         Mike Cronin                       Department of Corrections                    (406) 444-4907
                Dewey Hall                        1539 11th Avenue
                Research Specialist               Helena, MT  59620

Nebraska        Steve King                        Department of Correctional Services          (402) 479-5767
                Judy Egger                        P.O. Box 94661 
                Applications Analyst              Lincoln, NB  68509

Nevada          Glen Whorton                     Department of Prisons                         (702) 887-3277
                Chief, Classification            P.O. Box 7011
                  & Planning                     Carson City, NV  89702

New Hampshire   Mary Keniston                    Department of Corrections                     (603) 271-5609
                Supervisor, BIS                  105 Pleasant St.
                                                 P.O. Box 1806 
                                                 Concord, NH  03302

New Jersey      Stan Repko                       Department of Corrections                     (609) 984-4587
                Hank Pierre Chief,               Whittlesey Road
                Bureau of CICS (OBCIS)           Trenton, NJ  08625
	            Jim Atkins                       
                Karen Hughes 
                Administrative Analyst (CMIS)
				 
New Mexico      Pashella Reynolds-Forte          Corrections Department                        (505) 827-8631
                Robert Sego Management Analyst   P.O. Box 27116 
                Ralph Casados                    Santa Fe, NM  87502
                IS System Supervisor
				
New York        G. Ronald Courington             Department of Correctional Services           (518) 457-2540
                Bob Meidenbauer Manager,         State Office Building, #2 
                DP Services                      1220 Washington Ave.
                                                 Albany, NY  12226

North Carolina  Bob Brinson                      Department of Corrections                     (919) 733-5711
                                                 2020 Yonkers Road
                                                 Raleigh, NC  27604

North Dakota    Patrick Foley                    Department of Corrections & Rehabilitation    (701) 328-6607
                 Program Coordinator             3303 E. Main 
                David Huhncke                    Bismarck, ND  58502
                Cathy Jensen
                 Inmate Records  

Ohio            Peggy Ritchie-Matsumoto          Department of Rehabilitation & Corrections    (614) 752-1271
                Jerry Holloway Systems Analyst   970 Freeway Drive North
                Umang Nanda,                     Columbus, OH  43229
				Information System North 
                James Jeyarag , Ed White  

Oklahoma        Jim West Administrator, IS       Department of Corrections                     (405) 425-2546
                Bill Chown Administrator,        3400 M.L. King Jr. Ave
                Research Manager                 Oklahoma City, OK  73136

Oregon          Jean Hill                        Department of Corrections                     (503) 945-0920
                Vickie Ross                      2575 Center St.
                Steve McDowell                   Salem, OR  97310
                Randall Ireson, Research Manager    

Pennsylvania    Andy Keyser                      Department of Corrections                     (717) 730-2732
                Ron Peters                       P.O. Box 598 
                Heather Yates                    Camp Hill, PA  17001
                Special Projects Assistant    

Rhode Island    Steven Chianesi,                Department of Corrections                     (401) 464-3901
                Associate Director              40 Howard Ave.
			    Kevin Major, MIS Manager        Cranston, RI  02920
                                                 

South Carolina  Dr. Lorraine Fowler             Department of Corrections                     (803) 896-1748
                Division Director, Resource &   P.O. Box 21787 
                Information Management          Columbia, SC  29221                                                 

South Dakota    Laurie Feiler                    Department of Corrections                    (605) 773-3478
                Perry Delzer                     115 East Dakota Ave.
                                                 Pierre, SD  57501

Tennessee       Tim Beck                         Department of Corrections                    (615) 741-0900
                Gary A. Lukowski, PhD Director,  320 Sixth Ave., North 4th Floor
                Planning, Research               Nashville, TN  37243
                and Management
				
Texas           Linda Burney                     Department of Criminal Justice               (409) 294-6391
                Ann Christian                    P.O. Box 99
                Application Project Manager      Huntsville, TX  77342

Utah            Gae Lynn DeLand                  Department of Corrections                    (801) 265-5508 (801) 265-5597
                Christine Mitchell,              6100 South Fashion Blvd.
				Valerie Stagg                    Murray, UT  84107
                 Information Analyst              

Vermont         John Perry                       Department of Corrections                    (802) 241-2307 (802) 241-2293
                R. Barre Davis                   State Complex 
                                                 103 South Main St. 
                                                 Waterbury, VT  05676

Virginia        Frank Zera                       Department of Corrections                    (804) 674-3497
                 MIS Director                    P.O. Box 26963 
                                                 Richmond, VA  23261

Washington      Dale Putnam , Steve Collins      Department of Corrections                    (360) 586-6396
                 Application Service             Capitol Center Bldg
                R. Peggy Smith, PhD Planning     P.O. Box 41101
                & Research Manager               Olympia, WA  98504

West Virginia   Henry Lowery Records Supervisor  Division of Corrections                      (304) 558-2036
                William K. Davis Commissioner    112 California Ave. Building 4 
                Carl Graves Computer             Charleston, WV  25305
                Project Team
				  
Wisconsin       Mark Loder                       Department of Corrections                   (608) 266-8718
                 System & Application            P.O. Box 7925 
                 Development Chief               Madison, WI  53707
                    
Wyoming         John Lighty,                     Department of Corrections                   (307) 777-7405
                Jerry Pieper,                    Herschler Building, 
                Suzanne Rauth                    1st Floor East 122 W. 25th St. 
                IT Specialist                    St. Cheyenne, WY  82002



---------------------------------------
Appendix F
Data notes


Chapter 1. Profiling and describing offenders

Michigan

Age at commitment can be calculated. Employment status prior to
arrest can be inferred from occupation data element.

New Mexico

For education level prior to admission, high school dropout and
high school graduate can be calculated.

North Dakota

As of January 1998, the country of birth is not recorded, but
will be added soon.

Oregon

Education level prior to admission is recorded on a separate,
non-integrated database.

Pennsylvania

Age at commitment can be calculated. Employment status prior to
arrest is recorded only for the last six months before arrest. 
Most specific employment status information is recorded on paper.

South Dakota

Length of employment prior to commitment is recorded for the
last employment, which may be just prior to commitment or long
before commitment.

Vermont

Age at commitment can be calculated.

West Virginia

Age at commitment, citizenship, and illegal alien status can be
derived indirectly.

Chapter 2. Committing offenders into correctional authority

Alaska

Alaska records up to five conviction offenses. Only previous
incarcerations are recorded as criminal history. Data relating
to sentences imposed is incomplete.  The only monetary sanctions
recorded are fines and restitution. Classification is recorded,
but initial classification is not identified.



California

California does not record the agency with the authority to
release the offender from custody because of their sentencing
practices.  The amount of restitution due to victims of crimes
is recorded only if the victim contacts the Department of
Corrections.   Habitual offender information contains
convictions only.

Delaware

The State of Delaware does not record parole release data
because they do not have parole.

Indiana

Information about the criminal incident is found in the police
report in the offender's packet. Criminal history information is
stored electronically for priors resulting in DOC custody;
manual data storage is used for less serious, very old, or out
of state arrests.

Kansas

Criminal justice status of offenders at arrest is recorded only
for Kansas offenders.

Massachusetts

Criminal justice status is recorded for offenders who were on
probation for split releases only. Type of commitment for
probation violators is recorded only for those who were serving
a split sentence and returned. Risk assessment is only performed
on offenders within 1 to 6 years of their release date, and the
score is calculated upon commitment to prison. Needs assessment
is not done upon admission, but rather upon admission into a
program. Medical condition assessment is done upon commitment to
prison.

Michigan

Address of victim is recorded only if the victim requests to be
notified under Crime Victims Rights Act. The number of
conviction offenses can be calculated. Offense severity level is
not applicable because prison houses only felony offenders.
Criminal history includes only felony convictions, and priors
can only be calculated for offenders with Michigan prison
sentences. Offenders currently in prison at time of arrest
(criminal justice status) can be calculated. The number of
sentences imposed, the total sentence length, and whether a
sentence was a mandatory minimum can be calculated. Returned
from bond or appeal, transferred from another jurisdiction,
returned escapee, and returned AWOL/absconder are not commitment
types. The agency having the authority to release the offender
from custody is not applicable. The date of expiration of
sentence includes credits. The date of sentence expiration
without credits applied can be calculated.  For medical
conditions at admission, those conditions requiring chronic care
or psychiatric follow-up are identified. Drug testing at
admission is recorded, but not all offenders are tested. Initial
classification information, including classification index and
risk assessment index, is retained only until the offender's
classification changes. 

Mississippi

The offense severity level is not recorded because they only
record felony convictions. They are in the process of expanding
automated sentence length computation.

New Hampshire

The medical condition of offenders at admission is recorded, but
not for all offenders, especially out-of-state inmates. The
special unit housing the offender is recorded on paper for
offenders in special treatment programs, (e.g., drug or alcohol
programs).

Nevada

Type of commitment does not include returned escapees,
absconders, or detainees.  This is considered criminal justice
status, not a commitment. Criminal history includes only felony
offenses.   

New Mexico

Only firearms are recorded for weapon involved in criminal
incident. Up to seven offenses are recorded for number of
conviction offenses and number of sentences imposed. All values
of an offender's criminal justice status at  time of arrest can
be calculated. Offender criminal history contains up to six
previous offenses. If the offender is a habitual offender is
recorded based on the six most recent offenses. A history of
escape/AWOL can be calculated for risk assessment.

Oregon

The total length of sentence imposed can be calculated. Criminal
history includes Oregon offenses only. The agency responsible
for classification of offenders is not recorded because it is
done by the Department of Corrections itself. 

Pennsylvania

The number of conviction offenses is not currently recorded
electronically, but the DOC is developing electronic multiple
sentencing. A written description of the offense and the offense
severity level is only recorded for Pennsylvania offense code
(for current and prior offenses). Location of incident only
contains the committing county. The number of victims and
address of victims is recorded only when victims are registered
with the Office of Victim Services. Victim registration is
voluntary. The length of community supervision of the sentence
is not recorded because Pennsylvania has indeterminate
sentencing. information about expected time to be served is
recorded for predominant sentences only. For classification,
security level at admission is the custody level, and the
classification index is based on custody/housing level. A
psychological index is the offender's stability level. A risk
assessment index is used for housing placement; community risk
is not part of the index. A Bed Management System is being
developed which will record the type of housing unit in which
offender is placed.

South Dakota

Location of criminal incident includes only the county. The age
of the victim is recorded only for victims of sex offenders.
Criminal justice status at time of arrest is recorded only if
offender is parole violator. For criminal history, the number of
felony convictions is recorded electronically, but the specific
crime is stored on paper. Specific medical conditions at
admission are not recorded, but grouped for needs assessment.

Tennessee

For whether sentence was a mandatory minimum sentence, Tennessee
uses a sentencing grid. For example,  LWOP is 85% - 100% of
sentence; life is a minimum of 25 years.

Vermont

Changes in classification information can be calculated.

West Virginia

The type of conviction offense can be derived from other data
elements. For type of facility housing the offender, those in
county jails are recorded in a different system. 

Chapter 3. Managing offenders in corrections facilities	

Alabama

The medical staff has the information about conditions acquired
before and after commitment, but it is not in the offender-based
system.

Alaska

Good time credit data are incomplete. For type of facility
released to, only transfers or furloughs are recorded. For
behavior in custody, only offenders' infractions are recorded;
misconduct is not.

Idaho

Offender registry will be in their system by July 1998.  

Indiana

Program participation includes work assignments. However, since
almost everyone is eligible for work assignments, offender
eligibility evaluations are not done. Evaluations are done for
voluntary programs, such as substance abuse, anger management,
etc. Drug test tracking (a 5% monthly sample of offenders) is
recorded electronically at the aggregate level by the Central
Office, and manually at each facility, by offender. Not all
programs in which an offender participated can be identified.
Special good-time credit is given for education. Death is
recorded as a release type,  but not the specific cause of
death.	

Massachusetts

Program participation is based on risk assessment and is
voluntary. Only a substance abuse residential program is
recorded as ongoing in-prison program.  A new medical
information system is operating that collects data
electronically. Prior to this system, data was in paper form. 
To identify HIV/AIDS cases developed in custody, offenders must
consent to be tested. Special good time credits are not
applicable.	

Michigan

The total amount of good time credits available is not recorded.
It can be calculated, but requires extensive calculations. The
change in good time credits can be calculated. The reason for
change in good time credits is recorded only for infractions and
misconduct, not for new crimes. The type of medical treatment
recorded is for mental health or chronic physical problems
needing treatment. The date medical treatment started is
recorded only if offender is hospitalized or placed in treatment
facility.  Medical conditions developed in custody can be
calculated. Death is recorded as a release type, but not the
specific cause of death. Time served in custody can be
calculated, but with difficulty. If offender is required to
register as a sex offender can be calculated. The result of
official response to misconduct in custody is recorded as the
offender's status pending a hearing. Only hearings and appeals
are recorded for the type of legal procedure against offenders
as a result of misconduct.

Montana

The entity who authorized an internal movement is monitored by
each facility. Montana does not have good time.

Nebraska

Drug testing since admission is random drug testing only.

Nevada

Escape is not considered a type of release. Victim notification
is recorded when the data are provided by the victim. Recording
the medical care of offenders started recently.

New Hampshire

The reason for transfer/internal movement is recorded
electronically for a change in security level, protective
segregation, a medical condition other than HIV/AIDS, or a
psychiatric referral; but a transfer/internal movement for a
parole board hearing, HIV/AIDS cases, or a request by an
offender is recorded on paper. Releases from custody due to
death of offender are recorded on paper.

New Jersey

Releases from custody do not include work release and offenders
who abscond or go AWOL. Time served in custody can be calculated.

New Mexico

Death is recorded as a release type, but not the specific cause
of death. The type of facility released to is recorded by
general category. Victim notification upon the offender's
release is recorded for some offenses.

North Dakota

None of the program participation information is recorded
electronically.

Oregon

The type of program in which the offender participated is
recorded only since June 1996. Information about the medical
care of offenders is maintained in a separate, confidential
database. Death is recorded as a release type, but not the
specific cause of death. History of misconduct/infractions is
recorded electronically only for the last year; previous years'
information is stored on paper. 

Pennsylvania

Pennsylvania is developing a Transportation System that will
record who authorized transfers to another facility. They are
also developing a Bed Management System to record movements
within a facility and the date of the movements. These systems
will eventually record information for 100% of offenders.  The
reason for a transfer is currently recorded, but the reason for
an internal movement is not. Medical care of offenders is
recorded only for tuberculosis.  For conditions developed in
custody, hepatitis and chronic medical conditions are recorded
in paper format. Good time credit information is not recorded
because Pennsylvania does not have good time. Natural death is
recorded as a type of release, but illness is not recorded.

South Carolina

There is limited access to the data about conditions developed
in custody.  Death is recorded as a release type, but not the
specific cause of death.

South Dakota

For change in sentence length due to modifications, the
offender's record is adjusted to reflect sentence modifications.
Death is recorded as a release type, but not the specific cause
of death. The State Police are responsible for an offender
actually registering as a sex offender. The date of a
misconduct/infraction event is recorded as the date of the
disciplinary hearing. 

Utah

Good time credit information is not recorded because Utah does
not have good time.

West Virginia

Changes to sentence length and expected release date can be
calculated. Time served in custody can be calculated. 

Chapter 4. Supervising offenders on release and maintaining
public safety

Alaska

Data elements about released offenders and crimes committed by
offenders in the community are maintained only if they are
returned to prison. They do not distinguish between probation
and parole as types of supervision. The date an offender
absconded is the date the record was entered into the system. 
If an offender on release commits a new crime, the date of
criminal incident can be a series of years - if the event
occurred over a period of years.  They do have a 'date of
occurrence'.

Georgia

The Parole Board records residence information about offenders
on release. 

Idaho

Information about the victims of new crimes is confidential, and
is available only to the Parole Commission. Only felonies are
recorded for adjudication of a violation/new crime.

Indiana

Employment on release information is tracked by an agent.
Responses to release violations is recorded electronically for
felonies and manually for misdemeanors. Information about the
new crimes and the victims of new crimes is found in the police
report in the offender's packet.

Kansas

Information about violations/new crimes committed by offenders
on release is recorded only for Kansas felonies.

Massachusetts

Information about violations of release conditions and the
responses to release violations are collected for a study on
recidivism.  The date the offender is returned to prison is
collected for all offenders.

Michigan

Most information recorded on released offenders is for parolees
only. The actual time in supervision can be calculated.  The
type and date of a new crime can be calculated. Only if offender
was sentenced to a new crime is recorded, not violations. The
same information is recorded about the new criminal incident as
was for the original prison sentence. 

South Carolina

The address of a victim of a crime is recorded if the victim
chooses to register in a notification program. 

West Virginia

Parole is the only type of supervision recorded. The actual
length of time in supervision can be calculated.

Chapter 5. Facility management information

California

The medical care of offenders is recorded only if the medical
treatment is provided by a contract agency outside the prison.

Indiana

Not all program assessments are recorded electronically. 

Massachusetts

Program assessment is not done for all programs.

Michigan

The number of facilities can be calculated.

Mississippi

The number of offenders in a program is recorded electronically.
The number of program staff, program accreditation, program
cost, and source of funds is recorded on paper.

Pennsylvania

For types of programs offered to offenders, most information is
recorded in paper format; only the number of offenders in a
program is recorded electronically. The availability and storage
of program assessment information depends upon the individual
program.

Wisconsin

For types of programs offered to offenders, only the number of
offenders in program is recorded electronically.

Chapter 6. Reporting capabilities

Alaska

Only offenders' infractions are maintained on-line. 

California

Records of previously released offenders since 1977 are
permanently available on-line.  Only paper records of these
offenders are archived.

Kansas

Kansas is currently implementing a pilot program/software
application for collecting and maintaining electronically
information on released offenders.

Massachusetts

The Department of Correction collects data on offenders released
by parole or certificate of discharge to the street. The
follow-up period is one year from release date.  Data are
collected for probationers with split sentences.

Michigan

Michigan only collects data on crimes committed by offenders
under supervision if a new prison sentence is imposed.

Mississippi

The system maintains records of previously released offenders if
they were in the custody of Mississippi DOC.  It is uncertain if
the system maintains data on crimes committed by offenders under
supervision in the community.

Nebraska

Data are collected on crimes committed by offenders under
supervision only if the offender is convicted and sentenced.

New Jersey

The State Department of Probation and Parole tracks released
offenders, and shares some information with the Department of
Corrections, but the information is not considered part of the
DOC's information system.

Nevada

If an offender is incarcerated, his entire disciplinary history
is on-line. Prison records are maintained in the system for all
paroled offenders.  Some information is maintained for every
discharged offender, e.g., name, ID number, demographics. Only
paper records of released offenders are archived. Information
about offenders after they are released from prison is not
collected.

North Dakota

The system maintains an on-line history of offender's
misconduct/infractions only for loss of good time.

Rhode Island

Paper records are archived, but not electronic records.

South Carolina

All misconduct/Infraction records are kept on-line for the
current commitment. Previous commitments' infraction history can
be linked to current commitment. The system maintains data
elements on released offenders and collects data on crimes
committed by offenders only for supervised furlough releases.

South Dakota

The system maintains records on offenders released to the
community and offenders that are returned to custody.

Virginia

The system maintains an on-line history of external transfers,
but not internal movements.

West Virginia

The system maintains an on-line history of transfers, but not
internal movements.

General information

Michigan

Michigan did not include in their questionnaire any data
elements that are stored in paper format.  Their responses only
reflect data elements in their automated systems.  Their paper
files contain data elements that apply to the State of Michigan.

New Jersey

New Jersey has two offender-based information systems, CMIS and
OBCIS. They completed a questionnaire for each system.  We
combined the data from both questionnaires to represent the
highest degree of capacity.  For example, if a data element was
not collected by the CMIS system, but was collected by the OBCIS
system, we entered 'yes', the data element was collected. If
both systems collected a data element but for a different
percent of offenders, or stored in different formats, we entered
the response with the greatest percent of offenders or the one
stored in electronic format.

Wyoming

Wyoming maintains three separate information systems: one for
male offenders; one for female offenders; and one for
probationers and parolees.

-------------------------------------
Appendix I

Glossary



Availability index — a measure of the extent to which
departments maintain high-availability data elements
(electronically for more than 75% of offenders).  The index
ranges from 0% to 100%.  A rating of 100% means that a
department maintains all of the data elements in a group in
high-availability form (is at full availability); while a rating
of 0% means a department does not collect any of the data
elements in the group.

Availability indicator — a measure of the extent to which
information systems maintain data elements electronically for
most offenders.

High availability — a data element is in electronic form for
more than 75% of offenders.  This high percentage indicates
extensive coverage on an element. The electronic form indicates
the data potentially can be extracted, linked, and easily shared
electronically.

Medium availability — a data element is in electronic form but
for less 	than 75% of offenders.  It indicates a medium level of
availability because the scope of coverage is less.  It also
indicates that information about a comparatively large
percentage of offenders is more likely to be missing than in the
high-availability indicator.

Low availability — a data element is available only in paper
form.  Data 	elements available in low-availability form cannot
be extracted, linked, 	and shared electronically.  For the
purposes of using offender-based data elements to generate
statistical information, low-availability data elements present
large obstacles for departments' capacities.

Unknown availability — a data element is maintained by the
information system, but the form in which the data element is
maintained or its scope of coverage was not indicated on the
survey.

No availability —  a data element is not maintained in the
information system in any form.

Availability rating — the percentage of full availability on the
availability index that a department achieves for a group of
data elements.  The rating is derived by dividing a department's
availability scores for these elements by the data's
full-availability index score. 

Availability score — a value assigned to a group of data
elements based on their level of availability. 
High-availability elements receive a score of 3 points,
medium-availability elements receive a score of 2 points, low-
and unknown-availability elements each receive a score of 1
point.  No-availability elements receive a score of zero.  The
values of individual elements in a group are summed to create a
department's availability score.

Common core of data  — dimensions of data elements that are
maintained electronically for more than 75% of offenders by a
majority of departments.  The common core consists of 14
dimensions of corrections processing containing 100 data
elements.  Each of the four stages of processing has at least
one dimension in the common core.

Dimensions of corrections processing — 28 subsets of information
contained in the four stages of corrections processing.  They
are relatively homogenous groupings of data elements that
measure events in these four major phases of the corrections
process.  Stage 1, profiling offenders has 3 dimensions.  Stages
2 and 3, committing offenders and managing offenders, have 10
dimensions, and Stage 4, supervising offenders, has 5
dimensions. 

Full availability —  a rating of 100% on the availability index
or the score that a department would receive if it maintained
all of the data elements in a group in a high-availability form
(electronically for more than 75% of offenders).

Priority information areas  — six important information areas
identified by the Inventory's advisory committee: offender
profile, recidivism, program effectiveness, internal order,
public safety, and operational costs.  These area were used to
guide the development of the Inventory.  The first five areas
were organized into four stages of offender processing through
corrections systems.  The sixth area, operational costs,  was
used to develop data elements related to facility management.

Offender profile — a priority information area that covers a
wide range of Information covering offenders demographic
characteristics and risk potential, as well as their offenses,
criminal histories, sentences, types of admissions, and releases
from prison.  For clarity, this priority area was divided into
two stages of corrections processing—one that described
demographic and social characteristics of offenders, and another
that described both the behaviors and decisions leading to
commitment to prison and the assessment and placement decisions
made upon entry into prison.  

Recidivism — a priority information area that includes data
describing the rearrest, reconviction, and return to prison of
released offenders.  This area was combined with public safety
into the fourth stage of corrections processing—supervising
offenders.

Program effectiveness — a priority information area that
includes program participation by offenders, treatment, medical
problems, and medical care.  This area was combined with program
effectiveness into the third stage of corrections
processing—managing offenders.

Internal order — a priority information area that includes
information related to offender misconduct, violations of rules,
safety considerations, use of restraint, and drug and alcohol
use.  This area was combined with program effectiveness into the
third stage of corrections processing— managing offenders.

Public safety — a priority information area that includes data
describing the harm to the public by released offenders, the
crimes they commit, and information about victims of these new
crimes.  This area was combined with recidivism into the fourth
stage of corrections processing— supervising offenders.

Operational costs — a priority area that includes
non-offender-based data, such as those that measure staffing
ratios, program effectiveness, and costs of operating facilities.

Stages of corrections processing — method of organizing the 207
offender-based data elements, contained in the six priority
information areas identified by the advisory committee, into
substantive categories that describe major stages of corrections
processing, beginning with intake of offenders into the system
through supervision of released offenders in the community.

Stage 1. Profiling offenders — the first stage of corrections
processing.  It contains 29 data elements that describe
offenders' demographic characteristics, socio-economic status,
family characteristics, and living arrangements.

Stage 2. Committing offenders —  the second stage of corrections
processing.  It includes 70 data elements that describe the
offenses and sentencing decisions leading up to commitment into
prison and elements describing the assessment and placement of
offenders upon commitment.  The data are organized into three
broad categories that provide information about the offenses
leading to the conviction and sentences, about the sentences
imposed by the court, and about the assessment and confinement
decisions made by corrections officials upon receipt of an
offender from the court or other authorities.

Stage 3. Managing offenders — the third stage in corrections
processing.  It contains 63 data elements that describe the
movement of prisoners, the procedures and actions that
corrections officials take to manage offenders in their custody,
behaviors of offenders leading to disciplinary actions, and
official responses to misconduct.  The data are organized into
three broad categories that describe routine management and
program participation, the release of offenders from custody,
and internal order.

Stage 4. Supervising offenders — the fourth stage of corrections
processing.  It contains 45 data elements that describe where
offenders are in the community, what they are doing there, and
whether they have a record of criminal activity after release.
The data are organized into broad categories that describe the
supervision of offenders released from custody and details about
new crimes committed by released offenders.


End of file
pm 10/98