Sexual Abuse: A Journal of Research and Treatment, Vol. 17, No. 3, July 2005 (
Freeman, K. A., Mazza-Dexter, E. T., & Hoffman, K. C. (2005). Comparing
personality characteristics of juvenile sexual offenders and non-sex offending
delinquent peers: A preliminary investigation. Sexual Abuse: A Journal of
Research and Treatment, 17, 3–17.
In the reference cited above, we reported on similarities and differences be-
tween juvenile sexual offenders (JSOs) and nonsexual offending delinquent peers
(non-JSOs) on the Minnesota Multiphasic Personality Inventory—Adolescent
Version (MMPI-A). Our initial analyses indicated no statistically signiﬁcant dif-
ference in scores between experimental groups on any of the validity or clinical
scales of the MMPI-A. However, inspection of data revealed that the mean score
on Scale 4 (Psychopathic Deviant) fell above the cut off for clinical elevation for
the non-JSO group but in the nonclinical range for the JSO group. Therefore, we
created a dichotomous variable of “elevated” versus “not-elevated” on Scale 4. In
our original article, we reported that chi-square analysis indicated that a signiﬁ-
cantly greater number of participants in the non-JSO group scored in the clinically
signiﬁcant range than those in the JSO.
Unfortunately, we have discovered an error that occurred when we created
the dichotomous variable, such that two participants in the JSO group were mis-
classiﬁed. After correcting for this error, we ﬁnd no statistically signiﬁcant group
differences in the number of participants with scores in the clinical range on Scale
(1,N = 36) = 0.18.
Given re-analysis of our data after correcting the identiﬁed error, our in-
terpretation of the ﬁndings as being consistent with those of Oliver and Hall
(1993) is incorrect. Oliver and Hall found that nonsex offending delinquent peers
showed more signs of psychopathology than did juvenile sex offenders. Instead,
our ﬁndings are more consistent with those reported by Truscott (1993) and
Valliant and Bergeron (1997). Both studies showed that JSOs and nonsex of-
fending delinquent peers included in their samples did not differ on versions
of the MMPI used in those investigations (i.e., MMPI and MMPI Form 168,
2005 Springer Science+Business Media, Inc.