Title: An Examination of Personality Profiles based on Psychological Assessments of Violent and Nonviolent Offenders
1An Examination of Personality Profiles based on
Psychological Assessments of Violent and
Nonviolent Offenders
- Erica Hoover, MA
- Doctoral Candidate
- Aldwin Domingo, PhD
Mark Hume, PhD - Clinical Research Project Committee Chair
Committee Member - American School of Professional Psychology at
- Argosy University, Southern California
2Offenders
3Violent Offenders
4What can be done?
- The need to better understand these individuals
and their behavior is great. - Psychologists often emphasize treatment of these
offenders, especially when they also have a
mental illness. - Gaining more information about their personality
characteristics and behavior can help clinicians
create beneficial intervention strategies and
inform appropriate management of these offenders
(Craig, Browne, Beech, Stringer, 2006).
5Introducing My Study
- Use psychological assessments to
- differentiate between violent and
- nonviolent offenders.
- Incorporating various types of assessment into a
complete personality structure of an individual
can yield invaluable information. - Understanding an individual from many
perspectives, such as how they think, feel, and
interact with the world, is essential in order to
create an integrative picture of their
personality.
6(No Transcript)
7Study Methods
- Archival data was collected from the California
Institution for Men (CIM) in Chino California. - CIM is a mens prison facility that opened in
1941 and currently holds over 5000 inmates. - The archival data consisted of previous testing
batteries that have been performed on inmates
from 2008-2012. - The subjects were all part of the mental health
system at CIM and their primary clinician
referred them to have a psychological assessment
completed on them by a psychodiagnositc practicum
student working at the facility.
8- The cognitive measure the Wechsler Adult
Intelligence Scale (WAIS-IV) - The objective measure the Personality Assessment
Inventory (PAI) - The projective measure the Rorschach Inkblot
test using the Rorschach Performance Assessment
System (R-PAS) to score and interpret the
results.
9Type of Offender
- Violent crimes, or crimes against persons
- involve force or the threat of force, and are
comprised of the following offenses - Murder
- Manslaughter
- Robbery
- Assault
- Sex offenses
- Kidnapping
- Non-violent crimes include property crimes such
as - Burglary
- Theft
- Or drug offenses
10Traits analyzed
11Traits Analyzed
12Results
- Descriptive Statistics
- sample size 36 subjects
- Ages of subjects ranged from 18 to 77 years old,
with an average age of 37. - 38.9 Caucasian, 30.1 African American,
27.8 Hispanic, 2.8 Biracial - Completed education levels ranged from 4th grade
to 4-year college degree, average completed
education level of 11th grade. - 23 Violent Offenders, 13 Nonviolent Offenders
13Results
- 37 personality factors across three assessment
measures were statistically analyzed using a
Binary Logistic Regression. - The alpha level of statistical significance was
set at 0.05. - The best resulting BLR model to categorize
between violent and nonviolent offenders involved
the combination of
14Results
- Variables in the BLR equation for this model
- The resulting BLR regression equation is
- (-0.123)(PRI)(-0.128)(DOM)(-0.06)(PHR)22.7
Variables in the Equation Variables in the Equation Variables in the Equation Variables in the Equation Variables in the Equation Variables in the Equation Variables in the Equation Variables in the Equation Variables in the Equation Variables in the Equation
B S.E. Wald df Sig. Exp(B) 95 C.I.for EXP(B) 95 C.I.for EXP(B)
B S.E. Wald df Sig. Exp(B) Lower Upper
Step 3 PRI -.123 .053 5.472 1 .019 .884 .797 .980
Step 3 Dominance -.128 .044 8.570 1 .003 .880 .808 .959
Step 3 PHR -.061 .031 3.869 1 .049 .941 .885 1.000
Step 3 Constant 22.700 8.263 7.546 1 .006 7219612421.431
15Results
- This model correctly classified 20 of the 23
violent offenders resulting in 87 correctly
classified. - 9 of the 13 nonviolent offenders were correctly
classified, corresponding to 69.2. - The model demonstrated an overall correct
classification of 80.6.
Classification Table Classification Table Classification Table Classification Table Classification Table Classification Table
Observed Observed Predicted Predicted Predicted
Observed Observed Offense Offense Percentage Correct
Observed Observed Violent Offense Non-Violent Offense Percentage Correct
Step 3 Offense Violent Offense 20 3 87.0
Step 3 Offense Non-Violent Offense 4 9 69.2
Step 3 Overall Percentage Overall Percentage 80.6
16Results
- Multivariate analysis of variance (MANOVA)
Trait Offense Mean Std. Deviation Sig.
FSIQ Violent 83.2174 14.89038 .090
FSIQ Nonviolent 75.2308 9.28398 .090
PRI Violent 90.7391 16.42674 .080
PRI Nonviolent 81.7692 9.31087 .080
Dominance Violent 55.2609 13.03219 .022
Dominance Nonviolent 44.8462 11.59631 .022
Diffuse Shading (Y) Violent 97.8696 14.53930 .068
Diffuse Shading (Y) Nonviolent 89.7692 6.80874 .068
17Summary of results
- A combination of PRI, DOM, and PHR is able to
correctly classify an offender as violent or
nonviolent 80 of the time, and can correctly
classify violent offenders 87 of the time. - The mean scores for PRI were marginally
significantly different, violent offenders
scores were higher. - The mean scores on the DOM scale were
significantly different, violent offenders
scores were higher. - The mean scores on the PHR scale were not
statistically significantly different, (violent
offenders M106.61, nonviolent offenders
M101.15). - Adding the PHR scale to the BLR model improved
the overall correct classification rate by 5. - The mean scores on the Diffuse Shading (Y) scale
were marginally statistically significant - But not included in the BLR equation
18PRI
-
- Factors that may be related to an individuals
score include - (Sattler Ryan, 2009)
- Violent offenders scored higher than nonviolent
offenders - M90.7
- Nonviolent offenders M81.8
- Larger PRI vs VCI split with violent offenders
- VCI was 4 points lower (on average) for violent
- 1 point lower for nonviolent
19DOM
20PHR
21Diffuse Shading (Y)
22Full Scale IQ (FSIQ)
- (Sattler Ryan, 2009, p. 134)
- Violent offenders M83.2
- gt1 SD below the mean (low average)
- Nonviolent offenders M75.2
- Almost 2 SD below the mean (borderline)
23Clinical Implications
- Better understanding of violent offenders
- Prominent features that differentiate this
population from nonviolent offenders are related
to interpersonal characteristics, and reasoning
and problem solving skills - Prone to using nonverbal problem solving and
reasoning - More dominant and controlling
- Based on self-report
- Problematic understanding of self and others
- This understanding can guide treatment planning
toward more interpersonal/social skills and
adaptively understanding and interacting with
others - especially during conflictual encounters and
those that require healthy problem solving
24Limitations
- Small sample size (especially nonviolent
offenders) - Generalizability
- Only inmates within the mental health system at
one prison - Not representative of general prison population
- Inter-rater reliability
- Tests were scored by various examiners
- Standard validity concerns for each assessment
measure - Effort, motivation, performing at optimal level,
self-reporting, defensiveness, impression
management, rapport with examiner,
administration, nature of setting/environment,
etc.
25Conclusion
- Comparative characteristics and styles of those
prone to harm others directly vs other crimes in
a small sample - This is preliminary research -there is much more
we can do! - Gather more assessment data
- Study how these results can affect treatment of
violent offenders to reduce future violent crime
and increase their likelihood of success while
incarcerated and when released into the community
26references
- California Department of Corrections and
Rehabilitation CDCR. (2012a). Adult Population
Projections. Retrieved from the CDCR website
http//www.cdcr.ca.gov/Reports_Research/Offender_
Information_Services_Branch/ProjeProjec/S1 2Pub.pd
f - California Department of Corrections and
Rehabilitation CDCR. (2012b). Prison Census
Data. Retrieved from the CDCR website
http//www.cdcr.ca.gov/Reports_Research/Offender_
Information_Services_Branch/Annual/Census/ CENSUSd
1206.pdf - Coram, G. J. (1995). A Rorschach Analysis of
Violent Murderers and Nonviolent Offenders.
European Journal of Psychological Assessment,
11(2), 81-88. - Craig, L. A., Browne, K. D., Beech, A.,
Stringer, I. (2004). Personality characteristics
associated with r econviction in sexual and
violent offenders. The Journal of Forensic
Psychiatry Psychology, 15(3), 532-551. - Craig, L. A., Browne, K. D., Beech, A.,
Stringer, I. (2006). Differences in personality
and risk characteristics in sex, violent and
general offenders. Criminal Behaviour Mental
Health, 16(3), 183-194. - Federal Bureau of Investigation FBI. (2010).
Crime in the United States Violent Crime.
Retrieved from the FBI website
http//www.fbi.gov/about-us/cjis/ucr/crime-in-the-
u.s/2010/crime-in-the-u.s.-2010/violent- crime
27References
- Glaze, L. E. (2011). Correctional Population in
the United States, 2010. Retrieved from the
Bureau of Justice Statistics website
http//bjs.ojp.usdoj.gov/index.cfm?typbdetailii
d2237. - Guerino, P., Harrison, P. M., Sabol, W. J.
(2010). Prisoners in 2010. Retrieved from the
Bureau of Justice Statistics website
http//bjs.ojp.usdoj.gov/index.cfm?typbdetailiid
2230. - Meyer, G. J., Viglione, D. J., Mihura, J. L.,
Erard, R. E., Erdberg, P. (2011). Rorschach
Performance Assessment System Administration,
Coding, Interpretation, and Technical Manual.
Toledo, OH Rorschach Performance Assessment
System, LLC. - Morey, L. C. (2003). Essentials of PAI
Assessment. Hoboken, NJ John Wiley Sons. - Polaschek, D. L. L., Reynolds, N. (2004).
Assessment and treatment Violent offenders. In
C. R. Hollin (Ed.), The essential handbook of
offender assessment and treatment (pp. 201-218).
Chichester Wiley. - Sattler, J. M., Ryan, J. J. (2009). Assessment
with the WAIS-IV. La Mesa, CA Jerome M. Sattler,
Publisher, Inc. - Walters, G. D. (2007). Predicting Institutional
Adjustment With the Lifestyle Criminality - Screening Form and the Antisocial Features and
Aggression Scales of the PAI. Journal of
Personality Assessment. 88(1), 99-105.