Self-Confidence

1

Predicting Level of Denial Among Sex
Offenders Caitlyn E. McNeil University of
Nebraska-Lincoln

Results
Introduction
Table 2 Means (Standard Deviations) and ANOVAS
for the Discriminating Variables
Discriminant analysis was used to determine if
sex offenders who fully admitted, qualified
admitted, or denied their crimes differed in
terms of if they were impulsive, confused, had
relapses of violent non-sex charge, diagnosed
with substance abuse/depression, their education
level, age diagnosed with paraphilia rape and
psychosis. Table 2 shows the Means and standard
deviations for each treatment group and related
F-tests. The LSD minimum mean differences were
used to determine the pattern of mean differences
between each level of denial for each of the
various predictors. For impulse, those who did a
full admission of their crimes were found to have
significantly lower scores than both those who
did a qualified admission and denied their
crimes, while there was no mean difference
between those who denied and qualified their
admission. For confused and education level,
those who had a qualified admission were found to
have significantly higher scores than those who
denied their crimes, while there was no mean
difference between those who fully admitted their
crimes w/ those who denied and qualified their
admission. For relapse of violent non-sex
charges, diagnosis of substance abuse, diagnosis
of paraphilia rape, diagnosis of psychosis, and
age, no mean differences were found between any
of the levels of denial. Multivariate analysis
revealed that both discriminant functions
reliably differentiated between the groups. The
first function produced ? .661, X2(16)
38.729, p .001, with an R2-canonical .470.
The second function produced ? .848, X2(7)
15.369, p .032, with an R2-canonical .389.
Table 3 shows the standardized canonical
coefficients and structure weights for these two
models. For the first function, impulse, relapse
of non-sex charge, diagnosis of substance abuse,
age, and diagnosis of paraphilia rape effectively
discriminated between the groups. The second
function was found to have confusion, education
level, and age of the sex offenders contributing
to the discrimination among the groups. ANOVA
analyses w/ LSD minimum mean difference
comparisons were preformed to determine which
groups were significantly discriminated in the
first discriminant function. Figure 1 (LDF plot)
gives a graphical depiction of the multivariate
results and table 4 shows the re-classification
results from the LDF model. For the first
function, those who fully admitted their crimes
were found to be discriminated from both those
who did a qualified admission and who denied
their crimes F (2, 97) 13.766, p lt .001. For
the second function, those who denied their
crimes were found to be discriminated from both
those who qualified and fully admitted their
crimes F (2, 97) 8.665, p lt .001.
The purpose of this study is to examine sex
offenders relationships of level of
responsibility to eight factors impulse,
confusion, relapse of violent non-sexual charges,
diagnosis with substance abuse/depression,
education level, age, diagnosis with paraphilia
rape and psychosis. Though there has been scant
research on the content, structure, and reasons
for excuses among sex offenders, the intent of
this study was to explore various ideas to better
predict sex offenders level of denial and to see
their effect combined. Findings from Previous
Research There is no difference between
deniers and admitters in age, education, and IQ
(Langevin, Lang, Curnoe (1998). Deniers
had a lower IQ and were younger than admitters
(Baldwin Roys, 1998 as cited in Cooper,
2005). Excuses made by child molesters for
their sexual misconduct may offer important
indicators of remorse, rehabilitation potential,
and the probability of recidivism. They may also
show underlying cognitive disorders (Pollock
Hashmall, 1991). Cognitive deconstruction
enables individuals to distance themselves from
stress affiliated with self- awareness that puts
attention on the concrete rather than the
abstract level. It would explain how many sex
offenders convinced themselves that their
victim was enjoying their abuse (Ward, Hudson,
Marshal, 1995 as cited in Mann, 2004). Sex
offenders and non-sex offenders both experience
deviant fantasies (Crunoe Langevin, 2002).
25 of sex offenders reported having sexual
fantisies (Langevin, Lang, Crunoe, 1998 as cited
in Curnoe Langevin, 2002). Psychosis is
rare among sex offenders and deviant fantasizers
(Crunoe Langevin, 2002). Recidivists
demonstrated more divers victims, stranger
victims, more juvenile offenses and paraphilias
(Hanson Harris, 1998 as cited in Craig,
Browne, Stringer, 2003). Research
Hypotheses 1.) Diagnosis of psychosis would
not play any part in predicting a sex offenders
level of denial. 2.) If a relapse of violent
non-sexual charges was involved in predicting the
offenders level of denial, then this would also
be coupled with being diagnosed with paraphilia
rape.
Table 3 Standardized Canonical Coefficients and
Structure Weights for the Discriminant Function.
Discussion
The purpose of this study was to examine sex
offenders relationships of level of
responsibility to eight factors impulse,
confusion, relapse of violent non-sex charges,
diagnosis with substance abuse/depression,
education level, age, diagnosis with paraphilia
rape and psychosis. The hypothesis was that
diagnosis of psychosis would not play any part in
predicting a sex offenders level of denial and
that if a relapse of violent non-sex charges was
involved in predicting the offenders level of
denial, then this would also be coupled with
being diagnosed with paraphilia rape. As
hypothesized, psychosis did not play a part in
predicting a sex offenders level of denial which
confirms the results from Curnoe and Langevin
(2002) which stated that psychosis is rare among
sex offenders which led to correctly determine
that it psychosis would not be a main
contributing factor to predicting sex offender
level of denial. Also as hypothesized,
recidivism of violent non-sexual charges was
coupled with sex offenders being diagnosed with
paraphilia rape which confirms the results from
Hanson and Harris (1998) found that recidivists
in general demonstrated more diverse victims,
stranger victims, more juvenile offenses and
paraphilias (as cited in Craig et al.,
2003). Overall, there are several improvements
which might be made while considering future
research. Finding an accurate sex offender
representation of the population may be
difficult. Most sexual assaults are not detected
or reported or if they are reported are not
recorded. Some offenders may under or over
report their offensive behavior depending on the
social desirability of responding, which may then
throw off the researchers data (Tierney McCabe,
2001). The sex offenders present in the justice
system may not be representative of the whole sex
offender population. If the population is not
representable then the information gathered from
previous and future research may be only be
applicable to a portion of all sex offenders
(Wood, Grossman Fichtner, 2000). Until there
is a complete true justice system, all that is
known about sex offenders is biased toward a
section of sex offenders. The only way to get to
this complete justice system, if at all
attainable, would be to keep researching and
discovering new data to help us understand and
further our knowledge. Future research should
further look into whether or not IQ and education
have an impact on sex offenders level of
responsibility for their actions. It may also be
advantageous to examine other characteristics
such as age, sexual orientation, other
paraphilias, substance abuse, or even empathy for
their victims to see how those variables impact
the offenders acceptance of responsibility for
their crimes. Categorizing the sex offender
population into groups from different types of
sex offenses to age groups to grouping by mental
disorders, may be beneficial to determine if all
the variables relate equally across the different
populations. It may also be interesting to
discover if the level of education relates to the
severity of the crime. It would also be
beneficial to examine which level of
responsibility, experienced by certain types of
sex offenders, would yield better treatment
results.

Method
Table 4 Results from reclassifying Sex
Offenders into Groups

• Participants
• 387 male sex offenders ranging in age from 16
  to 83 with a mean age of 35.8 (S.D. 12.00).
• Convicts were from a Midwestern state and
  released from either an inpatient treatment for
  correctional facility between 1991 and 1997.
• 178 (46) were in the correctional facility and
  209 (54) were in the inpatient treatment
  facility.
• 280 (72.4) were European American, 36 (9.3)
  African American, 21 (5.4) Hispanic American, 21
  (5.4) Native American, and 5 (1.3) categorized
  themselves as other.
• Measures
• Data was archival and was collected from
  correctional, treatment, and legal records.
• Demographics (education level and age) were
  based on offenders condition at the time of
  arrest.
• Offenders deviant sexual activities, mental
  deficiencies, and emotions were taken from
  pre-sentencing investigations, mental health
  evaluations, and treatment reports.
• Offenders level of responsibility for their
  crimes were based from any of the available
  sources.
• Procedures
• ? Treated offenders either volunteered or were
  civilly committed by the courts while the
  untreated offenders were those who refused
  treatment and were then randomly selected based
  on the time of discharge as the treated group.
• ? Treated offenders received cognitive-behavioral
  programming group and individual therapy in a
  secure forensic hospital.

62 correctly re-classified
Figure 1
Hi confused Hi education level Younger
1
Hi impulse Low relapse of violent non-sex
charge Older NO - dx substance abuse NO - dx
paraphilia rape
Low impulse Hi relapse of violent non-sex
charge Younger YES - dx substance abuse YES - dx
paraphilia rape
1
-1
Qualified
Full
Denied
-1
Low confused Low education level Older