Table A2: Details of Scales Used in Developing Risk
Control Models
| Variable Label |
Variable Name |
Item Labels |
Sources |
Comments |
| Black-NonHispanic |
blknhisp |
Black-NonHispanic (dichotomy) |
OCFS |
Most of the remaining cadets were NonBlack-Hispanic.
There were negligible numbers of cadets who were NonBlack-NonHispanic. |
| Black-Hispanic |
blkhisp |
Black-Hispanic (dichotomy) |
OCFS |
| Age at First Release |
tranage |
Age at First Release (in years) |
OCFS |
Calculated as of the date of first release from the YLA
to community supervision. |
| Problems in Prior Placement |
failooh |
Failure in Prior Out-of-Home Placement, Recoded(dichotomy:
0 or 1) |
YLA |
Simple sum of three items. Missing values replaced
by "0" (zero) for Failure in Prior Out-of-Home Placement, Recoded
and Prior Voluntary Agency Placement, Recoded. |
| Prior Voluntary Agency Placement, Recoded(dichotomy:
0 or 1) |
YLA |
| Prior Escape Attempts (dichotomy: 1 or 2) |
YLA |
| Chronic Involvement |
chronic |
Court History (a 5-value rating ranging from "no
evidence of prior troubles with police or courts" to "chronic
criminality," collapsed to 4 values for analysis.) |
YLA |
Weighted sum of 9 items. The dichotomy Age
at First Out-of-Home Placement and the dichotomy Prior State Placement
were each multiplied by 2 prior to summation. |
| Age at First Out-of-Home Placement (a dichotomy identifying
youth who were 13 years old or younger at the time of first out-of-home
placement) |
YLA |
| Prior Movement Pattern (a dichotomy identifying youth
transferred to the YLA from another OCFS facility at the same security level) |
OCFS |
| Prior State Placement (a dichotomy identifying youth
with a prior placement to DFY/OCFS) |
OCFS |
| Number of Prior Offenses (the number of prior offenses
recorded in the OCFS data base, collapsed to 3 values for analysis) |
OCFS |
| Prior Drug Offenses (a dichotomy indicating one or
more prior drug offenses recorded in the OCFS data base) |
OCFS |
| Number of Prior Findings (the number of prior family
court adjudications recorded in the OCFS data base, collapsed to 3 values
for analysis) |
OCFS |
| Prior Out-of-Home Placements (the number of prior
out-of-home placements) |
YLA |
| Prior Record Seriousness (the sum of the statutory-class
ranks for prior offenses recorded in the OCFS data base, collapsed to 4
values for analysis.) |
OCFS |
| History of Low-Level Violent Crime |
minorv |
Number of Misdemeanors Against Persons (recoded into
3 categories) |
OCFS |
Simple sum of 5 items. Raw counts for constituent
items were collapsed into categories to combine low frequency cells and
reduce positive skew prior to summation. |
| Number of Misdemeanors (recoded into 3 categories) |
OCFS |
| Number of Current Assault Offenses (number of assault
offenses associated with the current placement) |
OCFS |
| Number of Historical Assault Offenses |
OCFS |
| Number of Historical Weapons Offenses (recoded into
3 categories) |
OCFS |
| Violent Criminal History |
viocrim |
Felonies Against Persons (a dichotomy identifying
youth with 1 or more current felonies against persons) |
OCFS |
Weighted sum of 4 items. The dichotomy Felonies
Against Persons and the dichotomy Historical Violent Non-Sex Crimes
were each multiplied by 2 prior to summation. |
| Number of Current Robberies |
OCFS |
| Number of Historical Robberies |
OCFS |
| Historical Violent Non-Sex Crimes (a dichotomy) |
OCFS |
| Crime with firearm |
armedoff |
Crime with Firearm (a dichotomy indicating whether
youth was ever alleged to have committed a crime while armed with a firearm) |
YLA |
|
| Optimum Length of Stay |
t2mu |
Optimum Length of Stay (time from YLA entry to first
release, coded in 3 categories: 150-179 days ÷ 0 120-149 days OR 180-209
days ÷ 1 < 120 days OR > 209 days ÷ 2) |
OCFS |
A value of zero reflects an optimum length of stay and higher
values reflects non optimum lengths of stay. |
| Disruptive Substance Abuse |
disrup_i |
Disruption Due to Alcohol (a dichotomy identifying
youth whose use of alcohol was judged to be disrupting normal functioning) |
YLA |
Simple sum of 2 items. Missing values for individual
items were replaced by the respective mean values. |
| Disruption Due to Drug Use (a dichotomy identifying
youth whose drug use was judged to be disrupting normal functioning) |
YLA |
| Problems in School |
school_r |
Academic Performance (a 5-point rating ranging from
"honor student" to "failing most classes") |
YLA |
Sum of 6 items. Raw sum ranged from 4 to 20
but was collapsed to values 1 through 5 for analysis. Missing values for
individual items were replaced by the respective mean values. |
| School Behavior (a 5-point rating ranging from "no
behavioral problems in school" to "incorrigible") |
YLA |
| School Attendance (a 3-point scale ranging from "no
unexcused absences" to "truancy petition filed") |
YLA |
| Truancy (a 5-point scale ranging from "committed
to school" to "totally out of school") |
YLA |
| Disobedient in School (dichotomy) |
YLA |
| Disruptive in School (dichotomy) |
YLA |
| Antisocial Attitudes |
behav_i |
Accepts Responsibility (a 3-point rating ranging
from "accepts responsibility for anti-social behavior" to "proud
of anti-social behavior") |
YLA |
Sum of 5 items. Missing values for individual
items were replaced by the respective mean values. |
| Empathy (a 3-point rating ranging from "has
empathy" to "no empathy") |
YLA |
| Motivational Focus (a 5-point rating ranging from
"driving interest" to "no faith in future") |
YLA |
| Prosocial Values (a 4-point rating ranging from "primarily
positive" to "resents prosocial values") |
YLA |
| Respect for Authority (a 4-point rating ranging from
"respects most authority figures" to "hostile to authority") |
YLA |
| Violent Behavioral History |
violence |
Threatened Students (dichotomy) |
YLA |
Simple sum of 6 items. Missing values for individual
items were replaced by 0. |
| Assaulted Students (dichotomy) |
YLA |
| Threatened School Staff (dichotomy) |
YLA |
| Assaulted School Staff (dichotomy) |
YLA |
| Threatened Shooting (dichotomy) |
YLA |
| Threatened Stabbing (dichotomy) |
YLA |
| Family Attachment |
famatt_r |
Parental Control (a 3-point rating ranging from "usually
obeys" to "consistently disobeys or is hostile") |
YLA |
Sum of 3 items. Missing values for Parental
Control were replaced by the mean value. The raw sum ranged from 3 to
13 but was collapsed to values 1 through 4 for analysis.Reverse coded so
that a lower value reflects more family attachment and a higher value indicates
less family attachment. |
| Family Support (a 5-point rating ranging from "structured"
to "rejecting") |
YLA |
| Street History (a 5-point rating ranging from "family
oriented" to "lives on the street") |
YLA |
| Criminogenic Family |
crimfam |
Family Drug Problems (dichotomy) |
YLA |
Simple sum of 3 items. |
| Family Involved in Crime (dichotomy) |
YLA |
| Victim of Neglect (dichotomy; same as below) |
YLA |
| Victim of Neglect |
neglect |
Victim of Neglect (a dichotomy indicating that youth
was alleged to be the victim of neglect) |
YLA |
Analyzed both as a distinct item and as a component of the
Criminogenic Family scale. |
| Hostile Home |
hoshom_r |
Negative Relationship with Mother (dichotomy) |
YLA |
Simple sum of 15 items. Raw sum ranged from
1 through 5 but was collapsed to values 1 through 3 for analysis. |
| Negative Relationship with Stepmother (dichotomy) |
YLA |
| Negative Relationship with Foster Mother (dichotomy) |
YLA |
| Negative Relationship with Grandmother (dichotomy) |
YLA |
| Negative Relationship with Female Sibling (dichotomy) |
YLA |
| Negative Relationship with Other Female (dichotomy) |
YLA |
| Negative Relationship with Father (dichotomy) |
YLA |
| Negative Relationship with Stepfather (dichotomy) |
YLA |
| Negative Relationship with Foster Father (dichotomy) |
YLA |
| Negative Relationship with Grandfather (dichotomy) |
YLA |
| Negative Relationship with Male Sibling (dichotomy) |
YLA |
| Negative Relationship with Other Male (dichotomy) |
YLA |
| Runaway (dichotomy) |
YLA |
| Victim of Physical or Sexual Abuse (dichotomy) |
YLA |
| Parents are Hostile (dichotomy indicating most extreme
rating on a 4-point scale) |
YLA |
| Years Behind Grade Level in Reading |
rlagimp |
Years Behind Grade Level in Reading |
OCFS |
Calculated as approximate age-appropriate grade level (age
- 5) minus Woodcock-Johnson reading grade equivalent score (truncated to
a whole number) |
| Years Behind Grade Level in Math |
mlagimp |
Years Behind Grade Level in Math |
OCFS |
Calculated as approximate age-appropriate grade level (age
- 5) minus Woodcock-Johnson math grade equivalent score (truncated to a
whole number) |
| Math Grade Equivalent |
mgeimp |
Woodcock-Johnson Math Grade Equivalent Score |
OCFS |
|
| Reading Grade Equivalent |
rgeimp |
Woodcock-Johnson Reading Grade Equivalent Score |
OCFS |
|
| Precinct-level Nonfelony Arrest
Rates |
ythnf_az |
Violation Arrests per 100K Under 25 (z-score) |
NYPD |
Sum of 8 z-scores. Z-scores calculated among
the 532 observations for 76 precincts by each of 7 years. Z-scores summed
to total scale value for each of the 532 precinct by year observations prior
to attaching a scale value to each person according to the year and precinct
to which youth was released |
| Violation Arrests per 100K Ages 21-24 (z-score) |
NYPD |
| Violation Arrests per 100K Ages 16-20 (z-score) |
NYPD |
| Misdemeanor Arrests per 100K Under 25 (z-score) |
NYPD |
| Misdemeanor Arrests per 100K Ages 21-24 (z-score) |
NYPD |
| Misdemeanor Arrests per 100K Ages 16-20 (z-score) |
NYPD |
| Violation Complaint Arrests per 100K All Ages (z-score) |
NYPD |
| Violation Summons Arrests per 100K All Ages (z-score) |
NYPD |
| Precinct-level Total Felony Arrest Ratio |
thftrtiz |
Precinct-level Total Felony Arrest Ratio (z-score) |
NYPD |
Ratio of number of felony arrests to number of reported
felony crimes, by precinct-year. Z-scores calculated among the 532 observations
for 76 precincts by each of 7 years (see above). |
| Precinct-level Theft Arrest Ratio |
totfrtiz |
Robbery Ratio (z-score for the ratio of the number
of robbery arrests divided by number of reported robberies) |
NYPD |
Sum of 4 precinct-by-year z-scores (see above) |
| Burglary Ratio (z-score for the ratio of the number
of burglary arrests to the number of reported burglaries) |
NYPD |
| Grand Larceny Ratio (z-score for the ratio of the
number of larceny arrests to the number of reported grand larcenies) |
NYPD |
| Other Felony Ratio (z-score for the ratio of the
number of "other" felony arrests to the number of reported "other"
felonies) |
NYPD |
| Precinct-level Felony Arrest Rates |
arrest_z |
Felony Assault Arrests per 100K population (z-score) |
NYPD |
Sum of 6 precinct-by-year z-scores (see above) |
| Murder/Non Negligent Manslaughter Arrests per 100K Population
(z-score) |
NYPD |
| Robbery Arrests per 100K Population (z-score) |
NYPD |
| Felony Arrests per 100K Under 16 (z-score) |
NYPD |
| Felony Arrests per 100K Ages 16-20 (z-score) |
NYPD |
| Felony Arrests per 100K Ages 21-24 (z-score) |
NYPD |
| Precinct-level Serious Crime Rates |
offens_z |
Felony Assault Offenses per 100K population (z-score) |
NYPD |
Sum of 3 precinct-by-year z-scores (see above) |
| Burglary Offenses per 100K population (z-score) |
NYPD |
| Robbery Offenses per 100K population (z-score) |
NYPD |
| Tract-level Singles and Service
Workers |
stable_z |
Percent Service Workers (z-score) |
RSITE |
Sum of 4 z-scores. Z-scores calculated among
the 286 census tracts for the year 1990. Z-scores summed to total scale
value for each tract prior to attaching a scale value to each person according
to the tract to which the youth was released. |
| Percent Population Black (z-score) |
RSITE |
| Percent of Males Never Married (z-score) |
RSITE |
| Percent of Females Never Married (z-score) |
RSITE |
| Tract-level Economic Status 1 |
income_z |
Percent of Household Income <15K (z-score) |
RSITE |
Sum of 5 tract-level z-scores (see above). |
| Percent of Household Income 15-25k (z-score) |
RSITE |
| Median Gross Rent (z-score reversed) |
RSITE |
| Percent of Occupied Units with no Vehicles (z-score) |
RSITE |
| Median Value Owner Occupied (z-score reversed) |
RSITE |
| Tract-level Economic Status 2 |
malinc_z |
Percent of Household Income <25K (z-score) |
RSITE |
Sum of 5 tract-level z-scores (see above). The
variable `Percent of Males Ages 10-19 (z-score) ' utilized data for
the year 1991, as it was not available in the Right Site data base for this
specific age group in 1990. |
| Percent of Population 25+ with Education < hs (z-score) |
RSITE |
| Median Household Income (z-score reversed) |
RSITE |
| Percent of Males Employed (z-score reversed) |
RSITE |
| Percent of Males Ages 10-19 (z-score) |
RSITE |
| Tract-level Culture Index |
cultur_z |
Percent of Occupied Structures with 2 Units (z-score
reversed) |
RSITE |
Sum of 4 tract-level z-scores (see above). The
variable `Culture Index' was collected at the county level and attributed
to the tracts by the developers of the Right Site data base. |
| Culture Index (z-score) |
RSITE |
| Percent of Blue Collar Workers (z-score reversed) |
RSITE |
| Percent of Units Occupied by Renters (z-score) |
RSITE |
| Tract-level Population Influx (1985-1990) |
move_z |
Percent Moved Between 1989-1990 (z-score) |
RSITE |
Sum of 2 tract-level z-scores (see above). |
| Percent Moved Between 1985-1988 (z-score) |
RSITE |
| Tract-level Population Influx (1989-1990) |
pcmv89z |
Percent Moved Between 1989-1990 (z-score) |
RSITE |
Analyzed both as a distinct item and as a component of the
Tract-level Population Influx (1985-1990) scale. |
