Criminal History Analysis with Suspects Arrested at Portland State University Kris R. Henning, Ph.D. Christian Peterson Portland State University Greg Stewart, Sgt. Portland Police Bureau February 22, 2012 1
OBJECTIVE The present study sought to characterize the criminal suspects arrested at Portland State University (PSU) by officers from the Campus Public Safety Office (CPSO). This includes an analysis of the suspects demographic characteristics, criminal histories, and their risk for subsequent offending. Three distinct data sources were used in this research and the findings specific to each dataset are documented below. CPSO CUSTODY DATA CPSO maintains a database on all criminal incidents and arrests occurring on campus grounds, campus buildings, and some adjacent locations (e.g., South Park Blocks). For the present study we extracted data on all custody reports between 12/16/2006 and 1/15/2012. Custodies include cases where the suspect was arrested by CPSO officers for a campus crime, cases where the suspect was arrested for an outstanding warrant, and it also includes citations in lieu of a physical arrest. The resulting dataset was analyzed and the findings are presented below. 1,146 custodies were reported between 12/16/2006 and 1/15/2012. o On average CPSO made 18.5 arrests per month or one arrest per 1.6 days. o Arrests were slightly higher in September, October, April, May, and June than other months. January and February had the lowest numbers. o Arrests were 49% higher on weekends as compared to weekdays. The average age of the suspects was 35.0 years old. o 1.6% were under age 18 o 33.0% age 18 to 25 o 18.2% age 26 to 35 o 47.3% age 36 or older 91.3% of the suspects were male. 78.0% of the suspects were White. o 12.7% were Black o 2.1% were Asian o 7.2% were classified as Other or unknown 2
4.9% of the suspects were Hispanic. 52.7% of the suspects were reported as being homeless or transient at the time of their arrest. 81.2% of the suspects had no official role at PSU when they were taken into custody. o 16.5% were students o.1% were faculty or staff o 2.3% the role at PSU was unknown PPB CRIMINAL HISTORY DATA The broader list of 1,146 suspects was then reduced through the following steps: First, we removed anyone identified as a current student, staff person, of faculty member, leaving only people with no official ties to the University (cases where the role at PSU was unknown were also excluded). We then selected only those cases where the suspect was arrested and transported to the county jail run by the Multnomah County Sheriff s Office (MCSO). Finally, we reduced the list so that each named person was only included one time. People with multiple custodies were identified using their name and date of birth and only the most recent custody was retained. From the resulting list of cases we then randomly selected 200 names. The list of 200 names was transferred to Sgt. Greg Stewart at the Portland Police Bureau (PPB). Under Sgt. Stewart s supervision a PSU intern, Christian Peterson who had undergone a background investigation by the Bureau, looked up each person in the Portland Police Data System or PPDS. This database contains criminal incident reports for Portland and some adjoining jurisdictions dating back to the mid-1970s. Unlike CPSO s current database, PPDS employs an authentication protocol to ensure that records are correctly linked to a person regardless of the spelling used in his/her name or the use of aliases. This system allowed us to narrow our sample further from 200 to 188 distinct people. We then extracted the entire arrest history from PPDS for each of the 188 individuals in the final sample. In addition to the dates of the arrest we had access to the types of crime involved (e.g., violent, property, violation of public order). These data were used to quantify each suspect s local criminal record at the time of his/her arrest by CPSO. Provided next are the findings from these data. 3
On average, the suspects had 19.7 prior criminal incident reports involving an arrest at the time they were taken into custody at PSU. The median number of prior incident reports was 9.0 (i.e., one half of the suspects had at least 9 prior incidents in PPDS). o The discrepancy between the median and average results from outliers with extensive criminal histories. o Seven of the 188 individuals for example, had 100 or more prior incident reports in PPDS. The majority (87.2%) of suspects had been arrested at least once prior to their current arrest at PSU. None 24 12.8% 1 to 4 39 20.7% 5 to 9 32 17.0% 10 to 14 15 8.0% 15 to 19 17 9.0% 20+ 61 32.4% 41.0% of the suspects had a recorded history of VIOLENT offending (e.g., assault, robbery, threats, stalking, harassment, rape). None 111 59.0% 1 to 4 65 34.6% 5+ 12 6.4% 56.4% of the suspects had a history of PROPERTY offending (e.g., burglary, auto theft, larceny, vandalism). None 82 43.6% 1 to 4 63 33.5% 5+ 43 22.9% 4
70.2% of the suspects had a history of criminal activity involving SUBSTANCES (e.g., driving under the influence of alcohol or drugs, illegal alcohol possession or use, manufacturing, possession, sale, or use of illicit substances). None 56 29.8% 1 to 4 63 33.5% 5+ 69 36.7% 73.4% of the suspects had a history of criminal activity involving a violation of PUBLIC ORDER (e.g., trespassing, disturbances, noise, littering, gambling, disorderly conduct, animal ordinances). None 50 26.6% 1 to 4 76 40.4% 5+ 62 33.0% 66.0% of the suspects had a history of VIOLATING A COURT ORDER (e.g., fugitive, escape, park violation, violation of exclusion orders, driving with a suspended license, felon in possession of a weapon, violation of stalking orders). None 64 34.0% 1 to 4 66 35.1% 5+ 58 30.9% 14.4% of the suspects had a history of criminal activity involving the use of a FIREARM OR OTHER WEAPON. None 161 85.6% 1 to 4 25 13.3% 5+ 2 1.1% 5
OREGON DEPARTMENT OF CORRECTIONS DATA 1 State correctional records were also examined for the CPSO arrestees that were randomly sampled. These records were only available for people convicted of a crime in Oregon and remanded to custody with the Department of Corrections (DOC). Only 99 of the 188 arrestees (52.7%) met these criteria 2. For this subset of arrestees we obtained information on their state-level arrest history and their risk for violent and property recidivism. The latter was based on a relatively new risk assessment scale created by the DOC called the Public Safety Checklist or PSC for short. The PSC considers an offender s demographics (e.g., current age, gender), criminal history (e.g., age at first arrest, arrests for violent crime, property offending), the current offense, and his/her history of supervision by state authorities (e.g., number of prison cycles, revocation of probation/parole). Various combinations of these variables were found to reliably predict new arrests in the validation sample of 59,089 DOC offenders. This included 37,937 people sentenced to probation and 21,152 released from prison between 2000 and 2005 (Bellatty, Prins, Shu, & O Connor, 2012). Two distinct scores are produced using the PSC: a violence risk total and a property risk total. These scores can then be used to identify the expected recidivism rates for violent and property offending. The expected recidivism rates are based on the actual re-arrest rates found with the PSC validation sample. A given offender s violence risk score can also be compared with the scores found among all DOC controlled offenders over the prior five years. This allows for a relative risk score which compares a given offender to all of the other offenders in the DOC system. 3 Provided next are the key findings from these data. 1 The authors would like to thank Margaret Braun and Paul Bellatty from the Oregon Department of Corrections for their assistance with this project. 2 It is important to understand that the subset of cases with a documented history in the DOC no longer qualifies as a random sample. As such, we cannot generalize the findings from this subset to all of the cases seen by CPSO. 3 One additional caveat with these findings is that the criminal histories and risk ratings were based on what was known as of January 2012 rather than the suspect s actual arrest date on campus. Some or all of the offenses reported herein for example, might have happened after the person s arrest by CPSO. 6
70.7% of the CPSO arrestees in this subset of cases had an arrest in Oregon for a VIOLENT CRIME (as of January 2012). Over one-half (54.5%) had an arrest for violence in just the past five years alone. ALL YEARS f % None 29 29.3% 45 45.5% 1 to 4 55 55.6% 51 51.5% 5+ 15 15.2% 3 3.0% Total 99 99 PAST 5 YEARS 68.7% of the CPSO arrestees in the DOC sample had an arrest in Oregon for a PROPERTY CRIME (as of January 2012). ALL YEARS f % None 31 31.3% 52 52.5% 1 to 4 54 54.5% 45 45.5% 5+ 14 14.1% 2 2.0% Total 99 99 PAST 5 YEARS The table below provides the projected recidivism rates for the arrestees in this sample. More specifically, nearly one-half (40.4%) of the individuals have a 40% or higher chance of being arrested for a new violent crime over the next 10 years. More than one quarter (29.3%) of these individuals have a 40% or higher chance of recidivating with a new property crime. VIOLENT RECIDIVISM Risk Level f % f % LOW (<20%) 13 13.1% 38 38.4% MOD (20-39%) 46 46.5% 32 32.3% HIGH (40+%) 40 40.4% 29 29.3% Total 99 99 PROPERTY RECIDIVISM 7
The final data table below allows for a comparison of the CPSO arrestees to Oregon offenders as a whole over the past five years. We found that 34.3% of the CPSO offenders scored at the 75 th %ile or higher on the PSC Violence risk scale and 42.2% scored at the 75%ile or higher for on the PSC Property risk scale. Another way of thinking about this is that one-third (34.3%) of the people in the CPSO sub-sample were among the state s worst offenders with regards to violence potential (i.e., top quartile). Remember, however, that these cases do not represent a representative sample of people arrested by CPSO because roughly one-half of the campus arrestees did not have a DOC record. Nevertheless, these findings suggest that a sizable portion of offenders arrested at PSU are at high risk for continued violent and property offending. VIOLENT PROPERTY Relative Risk f % f % 0 to 24th %ile 14 14.1% 22 22.2% 25th to 49th%ile 24 24.2% 11 11.1% 50th to 74th %ile 27 27.3% 24 24.2% 75th+ %ile 34 34.3% 42 42.4% Total 99 99 SUMMARY & CONCLUSIONS The findings presented in this report indicate that the majority of people arrested on campus at PSU have no official role with the institution. Many of these individuals are homeless, the majority of arrestees are male, White, and as a group they are older on average than our student population. We also found that most of these individuals have had prior involvement with the local and state criminal justice system. In some cases the extent of their prior criminal history was significant, including arrests for serious violent crime, property offending, and substance offenses. A smaller subgroup of individuals ranks among the state s highest risk offenders. These findings in themselves cannot speak to whether PSU s campus is particularly dangerous in comparison to other universities or even other areas of Portland. They do, however, highlight the unique challenge faced by urban institutions like PSU as compared to colleges and universities situated in rural areas or smaller communities. Crime prevention and interdiction efforts with the latter, particularly residential-based institutions, usually focus on students, because students account for the majority of victims and offenders. PSU and CPSO deals with these same issues among its student body (e.g., alcohol and drug use, vandalism, theft, assault, harassment), but also must contend with criminal activity by outsiders. Limited access control to campus buildings and 8
grounds, our proximity to downtown Portland, accessible public transport, and plentiful targets (e.g., bicycles, laptops, secluded locations for drug use) all increase our attractiveness to potential offenders. 9