In the late 1970s, the Rand Corporation pioneered a method
of collecting crime rate statistics. They obtained reports of
offending behavior--types and frequencies of crimes
committed--directly from offenders serving prison sentences. The
current study extends this research by exploring the extent to which
variation in the methodological approach affects prisoners'
self-reports of criminal activity. If the crime rates reported in this
survey remained constant across methods, perhaps one of the new
techniques developed would be easier and/or less expensive to
administer. Also, the self-reported offending rate data for female
offenders in this collection represents the first time such data has
been collected for females. Male and female prisoners recently
admitted to the Diagnostic Unit of the Colorado Department of
Corrections were selected for participation in the study. Prisoners
were given one of two different survey instruments, referred to as the
long form and short form. Both questionnaires dealt with the number of
times respondents committed each of eight types of crimes during a
12-month measurement period. The crimes of interest were burglary,
robbery, assault, theft, auto theft, forgery/credit card and
check-writing crimes, fraud, and drug dealing. The long form of the
instrument focused on juvenile and adult criminal activity and covered
the offender's childhood and family. It also contained questions about
the offender's rap sheet as one of the bases for validating the
self-reported data. The crime count sections of the long form
contained questions about motivation, initiative, whether the offender
usually acted alone or with others, and if the crimes recorded
included crimes against people he or she knew. Long-form data are
given in Part 1. The short form of the survey had fewer or no
questions compared with the long form on areas such as the
respondent's rap sheet, the number of crimes committed as a juvenile,
the number of times the respondent was on probation or parole, the
respondent's childhood experiences, and the respondent's perception of
his criminal career. These data are contained in Part 2. In addition,
the surveys were administered under different conditions of
confidentiality. Prisoners given what were called "confidential"
interviews had their names identified with the survey. Those
interviewed under conditions of anonymity did not have their names
associated with the survey. The short forms were all administered
anonymously, while the long forms were either anonymous or
confidential. In addition to the surveys, data were collected from
official records, which are presented in Part 3. The official record
data collection form was designed to collect detailed criminal history
information, particularly during the measurement period identified in
the questionnaires, plus a number of demographic and drug-use items.
This information, when compared with the self-reported offense data
from the measurement period in both the short and long forms, allows a
validity analysis to be performed.