CELF4 PERFORMANCE AND SEVERITY OF CRIMINAL OFFENSE IN ADJUDICATED YOUTH

1
CELF-4 PERFORMANCE AND SEVERITY OF CRIMINAL
OFFENSE IN ADJUDICATED YOUTH Kimberly J. Brock,
M.S. Donald R. Fuller, Ph.D. Eastern Washington
University Spokane, Washington
Table 1. Juvenile offense levels and
descriptions for the state of Washington.
INTRODUCTION In juvenile offenders there is a
combination of overlapping develop-mental
disorders and influences such as chaotic home
environment, conduct disorders, learning
disabilities, truancy, drug abuse, cognitive
disturbances, and poor social skills (Foley,
2001 Snow Powell, 2002). Many of these
offenders enter the correctional setting with a
variety of problems that often present challenges
in rehabilitative program-ming. Most juvenile
programs focus on specific developmental
disorders such as learning and attention
problems. Rehabilitative services are typically
limited to academics (e.g., obtaining the GED),
social skills, and pre-employment training, but
speech-language services are typically not
provided (Snow Powell, 2002). Studies have
shown that delinquent youth tend to have language
difficulties (Davis, et al., 1991 Sanger et al.,
1997). Further, there appears to be a higher
prevalence of language disorders in adjudicated
youth than the general population. The prevalence
of language disorders in the general population
has been estimated between 4 and 6 percent
(Falconer Cochran, 1989). By comparison, Davis
et al. (1991) found that 37.5 of 24 delinquent
offenders in their study qualified for language
services. Similarly, in three studies with female
juvenile delinquents, Sanger et al. (1997 2000
2001) found a preva-lence rate for language
problems between 14 and 22. Although these
rates vary from study to study, there appears to
be a higher prevalence of language problems in
troubled youth by comparison to the general
population. Although the prevalence of language
problems in the juvenile delinquent population is
higher than the general population, to date there
hasnt been a study to determine if there is a
relationship between severity of criminal offense
and language ability. This was the purpose of the
present study.
METHODS Participants Sixty (60) youths between
13 and 18 years of age (M 15.9 years SD 0.5)
participated in this study. All 60 had been
adjudicated by a Washington state juvenile court
for a criminal offense and were incarcerated in a
juvenile facility. To be included, youths had to
be English-speaking with essentially normal
hearing acuity. Gender, ethnic background, and
socioeconomic status were not exclusionary
factors. In the state of Washington, youth are
adjudicated according to the severity of criminal
offense as well as the number of prior
convictions. Offenses are categorized on a scale
from A to E, where A refers to criminal
offenses of greater magnitude and E represents
criminal offenses of lesser magnitude (see Table
1 at right). To facilitate statistical analysis
in this study, this scale was converted to a
number scale ranging from 1 (most severe) to
5 (least severe). Materials and
Procedures Hearing screenings were performed
using a Beltone AudioScout portable audiometer
calibrated to ANSI (2004) standards. All
participants passed a hearing screening at the
frequencies of 500, 1000, 2000, and 4000 Hz
presented at 25 dB. The Clinical Evaluation of
Language Fundamentals4th Edition (Semel et al.,
2003) was administered to all participants
according to the protocol for administration. For
all 60 participants, a raw score was obtained for
the Composite Language Score (CORE) and the
Expressive Language Index (ELI). The raw scores
for these two components of the CELF-4 were
converted to standard scores with
percentiles. Analysis Standard scores for the
CORE and ELI were ranked-ordered, as well as the
level of criminal offense. Data were analyzed
using a Spearman-rho correla-tion coefficient
procedure.
RESULTS AND DISCUSSION Relationship Between
Ranked Criminal Offense and Language
Performance Figure 1 provides a scattergram of
the relationship between the ranked severity of
criminal offense and ranked performance on the
CELF-4 CORE composite. A Spearman-rho correlation
coefficient of 0.06 indicated virtually no
relationship between ranked severity of criminal
offense and performance on the CELF-4 as a
whole. Figure 2 provides a scattergram of the
relationship between the ranked severity of
criminal offense and ranked performance on the
CELF-4 ELI subtest. The ELI is a more specific
measure of expressive language ability. A
Spearman-rho correlation coefficient of 0.00
indicated virtually no relationship between
ranked severity of criminal offense and
performance on the expressive portion of the
CELF-4. It appears that the relationship between
the severity of criminal offense a juvenile
commits and his or her language abilities is
random. In other words, poor language performance
does not appear to be a foregone conclusion for
juveniles who commit more serious crimes. By the
same token, a juvenile who commits lesser crimes
is just as likely to have a language disorder as
not. Based on these results, there does not
appear to be a link between the severity of
criminal offense a juvenile commits and
concomitant language difficulties. To further
test this hypothesis, the 60 participants in this
study were divided into two groupsgroup 1
included individuals adjudicated with more
serious crimes (defined arbitrarily as any
offense between A and B-, and included 27
participants) group 2 included individuals
adjudicated with lesser offenses (offenses
classified from C through E, and included 33
participants). Two independent t-tests were
conducted on the CORE and ELI scores for the two
groups.
The mean CORE score for group 1 was 91.41 (SD
20.31) and for group 2 it was 94.85 (SD 19.41).
A t-test t(58) 0.671, p gt .05 revealed no
significant difference between the two groups on
the CELF-4 composite language score. The mean
ELI score for group 1 was 86.33 (SD 17.83) and
for group 2 it was 88.18 (SD 17.00). A t-test
t(58) 0.411, p gt .05 analysis revealed no
significant difference in expressive language
abilities between the two groups. Results of
these t-tests supported the Spearman-rho
results. The Prevalence of Language Disorders
in Adjudicated Youth Although there doesnt
appear to be a link between severity of criminal
offense and language ability (or lack thereof),
it is interesting to note that this study
supports findings from previous studies that
there is a higher prevalence of language problems
in juvenile delinquents than in the general
population of children (Davis el al., 1991
Sanger et al., 1997 2000 2001). Most public
schools in the state of Washington follow the
state associations guideline that a standard
score at least 1.5 standard deviations below the
mean is indicative of a language disorder (on the
CELF-4, this would equate to a standard score of
approximately 78). Using this guideline, 10
participants in this study (16.7) would be
classified as having a language disorder based on
the CORE standard score. More interestingly, 17
participants (28.3) would be classified as
having an expressive language disorder based on
ELI performance. Such a high prevalence of
language problems in incarcerated juveniles
clearly indicates that language services are
sorely needed. The problem is that for most
youths, incarceration is only temporary. This
fact should lead states to providing itinerant
speech-language pathology services for such
at-risk youth. This appears to be a population of
young people that is not being served.
Figure 1. Scattergram of ranked offense levels
and ranked performance on the CELF core
composite.
Figure 2. Scattergram of ranked offense levels
and ranked performance on the CELF ELI subtest.