[acb-hsp] Gender, Ethnicity . . . Article

[J.Rayl]

Gender, Ethnicity, and the Family Environment: Contributions to Assessment Efforts
within the Realm of Juvenile Justice*
by Stephen M. Gavazzi
The enormous rise in the rates of imprisonment in the adult population has generated
an increased appreciation of the criminal justice system's impact on the family,
especially regarding the deleterious effects of incarceration on family stability
and the ability to parent (Sandifer & Kurth, 2000; Western & McLanahan, 2000). In
tandem, adolescents themselves are being detained at greater rates, and increasingly
are being bound over to the adult criminal justice system, even though violent juvenile
crime has decreased (Schindler & Arditti, 2001). The underreliance on assessment
information contributes to an appreciable gap between the needs of courtinvolved
youth and the services they receive (Stiffman, Chen, Elze, Dore, & Cheng, 1997),
which may be one of the primary reasons behind the criminal justice system's overreliance
on detention.
Unfortunately, users of assessment instruments in the juvenile justice realm tend
to be more the exception than the norm. Although formal assessment procedures are
thought to provide greater validity, structure, and consistency to efforts that match
interventions to needs (Howell, 1995), more often than not anecdotal evidence suggests
that juvenile justice professionals rely on intuition and experience to guide and
direct their interactions with youth who have come to the attention of the court.
In turn, this lack of reliance on assessment information contributes to the "highly
subjective" nature of the juvenile justice system that can lead to "erroneous, inequitable,
and inconsistent decisions" (Stevenson et al., 1996, p. 8).
When assessment instruments are employed, typically they have been used as case management
tools that augment decision making about how stringent the level of supervision (i.e.,
the intensity of the probation or parole period) should be for a given jusvenile
offender. Risk assessment instruments that serve a case management function usually
focus on offense-related information (age at first arrest, severity of the current
offense, etc.) and typically are employed in order to reduce the likelihood of reoffending
behavior (recidivism) by the court-involved youth (Andrews & Bonta, 1994). Professionals
working with juvenile offenders also have employed instruments that measure certain
"extralegal" factors believed to be related to court involvement (Silver & Miller,
2002). For instance, juvenile court judges oftentimes will express interest in the
results of mental health and/or substance abuse screens as a means of informing their
court-ordered treatment plan (Herz, 2001).
Recently, more attention has been given to a crucial but often overlooked extralegal
factor: the families of court-involved youth. The literature to date suggests that
a variety of family factors are associated with delinquent behavior. Historically,
however, the criminology literature has paid most attention to the direct effects
of family structural characteristics on delinquency (Fox & Benson, 2000), such as
divorce, economic disadvantage, and residence in dangerous neighborhoods. It is only
more recently that researchers have begun to pay attention to the intermediary role
of family process variables and considered those strategies used by families to attain
goals related to member well-being (Day, Gavazzi, & Acock, 2001).
The growing body of family process research has linked numerous family factors to
adolescent delinquent behavior (LeBlanc, 1992; Patterson & Dishion, 1985; Rankin
& Kern, 1994; Simons, Wu, Conger, & Lorenz, 1994) and a corresponding set of studies
that document the effectiveness of family-based treatment for court-involved youth
(Barton, Alexander, Waldron, Turner, & Warburton, 1985; Dishion & Andrews, 1995;
Henggeler, Melton, & Smith, 1992). Most impressively, researchers have documented
how family processes such as monitoring/supervision, discipline strategies, and attachment
mediate the relationship between family structural characteristics and delinquency
(Patterson, Reid, & Dishion, 1992; Sampson & Laub, 1994; Simons, Chao, Conger, &
Elder, 2001).
This empirical knowledge base uniquely positions the family field to become an evergrowing
influence in how decisions are made about courtinvolved youth and their families,
especially if assessment methods are put into place that provide meaningful and user-friendly
information for juvenile justice professionals. The purpose of the present study
is to advance the field's understanding of how assessment information about family
and parenting characteristics, coupled with attention given to gender and ethnicity,
may provide assistance to professionals working with court-involved youth and their
families.
The Global Risk Assessment Device
The Global Risk Assessment Device (GRAD) is an Internet-based instrument that has
been developed as a reliable and valid tool for professionals working with court-involved
youth and their families. The GRAD originally was developed in order to enhance the
assessment requirements of family-based interventions in the juvenile justice realm
that followed a psychoeducational format (Gavazzi, Wasserman, Partridge, & Sheridan,
2000). These interventions were developed as a continuum of services that addressed
the global needs of court-involved youth and their families. Such services typically
include both diversion-oriented programs (Gavazzi, Yarcheck, Wasserman, & Partridge,
2000) and interventions for the families of youth released on parole from juvenile
facilities (Gavazzi, Yarcheck, Rhine, & Partridge, 2003; Partridge, Gavazzi, & Rhine,
2001). Hence, this instrument was built by and for family-based professionals who
were invested in the idea that broadband assessment could guide and direct appropriate
service delivery.
The GRAD includes several important features. First, cut-off scores are provided
to users that serve to instantaneously categorize youth as low, moderate, or high
risk in each risk domain. Although courts are given the opportunity to use cut-off
scores that are based on norms established on the sample of approximately 5,000 youth
and families assessed to date by the GRAD to date, so far, all courts have chosen
to have their cut-off scores based on the unique characteristics of their clientele.
This decision seems to balance the need for standardized questions across courts
with the desire to have respect paid to the "local culture" within a single court.
Other features of the GRAD information management system include a user registration
process that protects access to the court's database. Further, risk data are linked
to important demographic information that is collected on each youth who is assessed,
and all of that information is systematically aggregated and available to the user
for reporting purposes (i.e., weekly census figures, monthly case contact reports).
Also, there is a case management-tracking page that is designed to collect information
on referrals and subsequent service provision, and the database has been constructed
in order to facilitate the collection of longitudinal information from multiple perspectives
(youth, parent, professional) throughout the case management process. Finally, the
GRAD has the unique ability to insert a given county's specific matrix of programs
into the database, creating the opportunity to collect referral information that
is location specific to that county (Gavazzi, Novak, Yarcheck, & Distefano, 2004).
Recent efforts to document the psychometric properties of the GRAD have generated
support for its use in assessing potential threats to the wideranging developmental
needs of adolescents who come into contact with the juvenile justice system at various
points of entry. The GRAD typically is employed at the outset of a professional's
contact with a given youth and is used to make recommendations and referrals that
are based on reliable and valid information about risk factors and needs in a wide
variety of relevant domains, including peer relationships, substance abuse, mental
health issues, criminal activity, accountability for one's own actions, education
and vocational issues, exposure to traumatic events, health-related risks, and family
and parenting characteristics. The last domain was designed to measure many of the
same variables used in the family process research described earlier, including monitoring
and supervision, discipline, and parent-adolescent relationship attributes.
In terms of research-related support, the GRAD has been used in several studies that
have generated evidence of its reliability and validity. For instance, preliminary
empirical work with the GRAD reported excellent psychometric properties, including
a solid factor structure and high internal reliability coefficients (Gavazzi, Slade,
et al., 2003). Additional studies have generated concurrent validity evidence of
the GRAD with other well-established measures of risk/needs (Gavazzi & Lim, 2003),
especially with regard to its ability to tap into issues that concern substance use,
mental health, and family environment. Still, other empirical work has examined the
predictive validity evidence of the GRAD, generating support for this tool's use
in referring youth to the most appropriate level of care, including the need for
individual and family therapy (Gavazzi, Lim, Yarcheck, & Eyre, 2003). In essence,
the research evidence suggests an important bottom line: The GRAD items measure what
they are supposed to measure, they do so in consistent fashion, and the resulting
information can identify the need for direct services that are related to the global
risks and needs measured by this assessment tool.
More recently, the GRAD has been used in studies that have initiated an examination
of the impact that both gender and ethnicity have on the risks and needs of court-involved
youth and their families. Accumulating empirical evidence suggests that the role
of gender is quite profound (Chesney-Lind, 1997; Chesney-Lind & Okamoto, 2001), especially
when the pathways to delinquent behavior are considered (Belknap, 2001; Office of
Juvenile Justice & Delinquency Prevention [OJJDP], 1998). In turn, the impact of
ethnicity also is thought to be substantial (Leiber, 1994), especially in light of
mounting evidence regarding the disproportionate confinement of minorities in the
juvenile justice system (Dejong & Jackson, 1998; Steinhart, 1996). Hence, both the
consideration of gender and ethnicity are thought to be important variables necessary
in any empirical effort that targets status offender populations.
Gavazzi, Yarcheck, and Chesney-Lind (in press) used GRAD data from a sample of youth
assessed at a juvenile court detention facility in order to examine potential gender
differences regarding risks and needs. Some expected results were reported, such
that men scored significantly higher on prior offenses and women scored higher on
family/parenting, mental health, traumatic events, and health-related risks. However,
women also scored significantly higher than males in domains classically associated
with delinquent men: psychopathy, accountability, and delinquent peer relationships.
Further, gender differences also were found in this study in terms of the type of
offense that brought the male and female adolescents to the attention of the detention
facility. Here, women were more likely to be repeat status offenders (involving activities
that would not be illegal if done by an adult, such as curfew violation, runaway,
truancy, ungovernability, and underage smoking and drinking activity) or charged
with domestic violence, whereas men were more likely to be charged with property
crimes and person-related offenses. Also, whereas women in all offense categories
generally displayed higher levels of family/parenting risks than men, those men and
women charged with domestic violence consistently displayed the highest family/parenting
risk scores of each of their respective samples.
A second example involved a study that compared the risks and needs of status offenders
to an at-large juvenile court sample, with specific attention paid to factors related
to gender and ethnicity through comparison of risk scores for men and women and for
Caucasian and minority youth (Gavazzi, Yarcheck, & Lim, 2005). Status offender youth
displayed significantly higher risk scores than an atlarge court sample in domains
associated with both family/parenting concerns and educational concerns. Also, women
displayed higher risk levels than men, and White youth displayed significantly higher
risk levels than minority youth across a wide variety of risk domains.
Further analyses were conducted in this study regarding the participation of African
American men and women in a program designed as an alternative to detention. Interestingly,
although these two groups displayed similar rates of program success, the minority
male and female youth who completed the program displayed different needs at the
outset of their participation. More specifically, those African American women who
had successfully completed the alternative to detention program displayed significantly
greater problems in their peer relationships and more health-related risk behaviors
at the outset of their participation in the program. In contrast, African American
men who successfully completed the alternative to detention program displayed significantly
lower levels of problems in their peer relationships at the outset of their participation
in the program.
In combination, the studies reviewed above report findings that point to a complex
relationship between gender, ethnicity, and the risks and needs of court-involved
youth and their families. To date, however, no studies have considered the gendersensitive
and culturally specific nature of the most important extralegal factor contained
within the GRAD: family and parenting concerns. Therefore, the present empirical
effort examined potential variation in the family/parenting domain of the GRAD in
a sample of African American and Caucasian male and female adolescents who have come
to the attention of the juvenile court.
Based on several previous studies using the GRAD, it was hypothesized that women
would display higher risk scores than men on the family/ parenting domain of the
GRAD. Although no ethnicity differences have been reported on this domain in prior
studies, the only previous study to have examined the possibility of an interaction
between gender and ethnicity utilized a sample that was predominantly African American
(Gavazzi, Yarcheck, & Lim, in press). Hence, a sample with more balanced ethnic representation
was hypothesized to generate a significant Gender × Ethnicity interaction regarding
scores on the family/parenting domain.
Method
Participants
Data were collected from 1,609 youth who were assessed by juvenile justice professionals
in four United States Midwestern county juvenile courts. The sample was 37% women
(n = 590) and 63% men (n = 1019), with an average age of 14.9 years (SD = 1.6; range
= 11 - 17 years). Fifty-one percent of the sample was African American, and 49% was
Caucasian. The predominant household composition was single parent mother headed
(48%), followed by households with both biological parents (15%), stepfamily (14%),
foster family or other (11%), single parent father headed (6%), and grandparent headed
(6%). It should be noted that fully 54% of the youth in this sample were living with
a single parent, compared with 27.7% of all children in the United States under the
age of 18 years living with a single parent (Lugaila, 1998).
Measures
The data collection instrument used was the most recent youth report version of the
GRAD (Gavazzi, Slade, et al., 2003). The average time that it took to complete the
GRAD was about 25 min. Respondents were asked to respond to the items by indicating
on a scale of 0 - 2 (where 0 indicates no/never, 1 indicates yes/a couple of times,
and 2 indicates yes/a lot) how much each item applies to their life.
The family/parenting domain comprises 17 items that focus on family process factors
such as conflict (How often do you get into fights with adults who live in your home?),
parental monitoring (How much of the time do the adults who live with you not know
where you are?), and discipline strategies (When you are punished for your behavior,
is it harsh or inconsistent?). The family/parenting domain was significantly associated
with the Family Events Scale (Patterson et al., 1992) in one previous study (Gavazzi,
Lim, et al., 2003), and coefficient alpha was reported to average .86 across two
other studies (Gavazzi, Slade, et al., 2003). The total theoretical range of this
domain is 0 - 34; coefficient alpha for the GRAD family/parenting domain items in
the present study was .82 for the entire sample.
The prior offenses domain comprises five items that are designed to indicate the
amount of involvement in previous delinquent activity, including past involvement
in the juvenile justice system (How often have you appeared before a judge or magistrate?),
as well as delinquent activity that has gone undetected (How often have you done
something illegal that you did not get caught for?). Male adolescents have reported
significantly higher prior offenses than women in two studies (Gavazzi, Yarcheck,
& ChesneyLind, in press; Gavazzi, Yarcheck, & Lim, in press), and coefficient alpha
was reported to be .87 across two other studies (Gavazzi, Slade, et al., 2003). The
total theoretical range of this domain is 0 - 10; coefficient alpha for the GRAD
prior offenses domain items in the present study was .88 for the entire sample.
Procedure
Six hours of training were completed by juvenile justice professionals prior to their
participation in the data collection process. The youth included within the present
study were assessed by line staff working in one of the four county's detention,
probation, or intake/diversion departments. A total of 103 professionals took part
in the collection of data for this study. Each of these professionals had ongoing
telephone contact with their GRAD trainers and face-to-face consultation with their
administrative officers as part of a quality control effort designed to ensure competence
of administration. Court professionals are trained to read the GRAD items in verbatim
fashion but are encouraged to provide additional information as needed in order to
gain the most accurate responses from informants.
Results
To test for potential main effects and significant interaction terms among the groups,
a 2 (gender) × 2 (ethnicity) analysis of variance (ANOVA) was computed with the family/parenting
GRAD domain as the dependent variable. The results revealed significant main effects
for both gender, F(1, 1605) = 41.52, p < .001; and ethnicity, F(1, 1605) = 6.94,
p < .01; as well as a significant Gender × Ethnicity interaction term, F(1, 1605)
= 6.44, p < .01. The estimated marginal means for the family/parenting domain scores
as a function of gender and ethnicity are presented in Figure 1. Reflecting the interaction,
African American women displayed the highest risk scores on the family/parenting
domain-meaning that they reported the least functional family process experiences-followed
by Caucasian women, with Caucasian men and African American men displaying equally
lower scores than both of the female groups.
In order to ascertain the degree to which these family processes were associated
with adolescent delinquent behavior, the sample then was divided into two approximately
equal groups using a median split on the family/parenting domain scores, and a 2
(gender) × 2 (ethnicity) × 2 (high and low family/ parenting domain scores) ANOVA
was computed with the prior offense domain as the dependent variable. The results
revealed significant main effects for family, F(1, 1601) = 162.96, p < .001; gender,
F(1, 1601) = 20.87, p < .001; and ethnicity, F(1, 1601) = 19.73, p < .001; as well
as a significant Gender × Family interaction term, F(1, 1601) = 11.61, p < .001.
The estimated marginal means for the prior offense domain scores as a function of
gender and family for both African American and Caucasian youth are presented in
Figures 2 and 3. Reflecting the interaction, when family risks and needs were at
lower levels, men reported higher levels of prior offenses than women in the African
American and Caucasian samples. However, the presence of higher family risk and needs
coincided with African American and Caucasian women reporting relatively equal prior
offenses with their male counterparts.
Discussion
The present study contributes to the literature on assessment efforts in the juvenile
justice realm. Most importantly, this effort has focused attention on the families
of court-involved youth, yet another "hidden pocket" of families involved in the
criminal justice system (Arditti, Lambert-Shute, & Joest, 2003). Although family-based
professionals may tend to view as axiomatic the view that family is perhaps the most
important extralegal variable of them all, members of our field must keep in mind
that the juvenile justice field does not always share our deep appreciation for the
primacy of the family.
Clearly, the family's impact on the lives of courtinvolved youth is influenced by
both gender and ethnicity. As hypothesized, women did in fact display higher risk
scores than men on the family/ parenting domain of the GRAD. And a significant Gender
× Ethnicity interaction regarding scores on the family/parenting domain was found.
The fact that women, in general, scored significantly higher on the family/parenting
domain was consistent with the findings of previous studies. However, African American
women displayed higher scores than Caucasian women, something not previously reported,
although a similar ethnicity difference was not seen in the male scores. At least
one study using the GRAD has noted other differences between African American men
and women who have come to the attention of the juvenile court. Gavazzi, Yarcheck,
and Lim (in press) reported that African American women who had higher peer relationship
and health-related risk scores, and men who had lower peer relationship risk scores,
displayed better responses to treatment in a program designed as an alternative to
detention.
Clearly, the confluence of gender and ethnicity within the context of the family
is an area that is ripe for further empirical exploration. In turn, it is hoped that
this future research will make a significant contribution to the growing sophistication
of gender-specific and ethnically sensitive theories of delinquency.
In this regard, Hoyt and Scherer (1998) comprehensively reviewed various theoretical
explanations of female juvenile delinquency, including emphases on developmental
correlates (e.g., age of menarche), academic contexts (mixed-sex vs. all-girl schools),
sexual victimization (with women more often victimized), and differential delinquent
peer associations (whereby different meanings are associated with being in a female
vs. a male gang). These authors also highlighted theoretical work that rests on the
presumption that family factors play a particularly important role in the lives of
female adolescents who come into contact with the juvenile justice system. The findings
of the present study are thought to be consistent with the assertions made by these
authors regarding the need to pay less attention to structural family factors (such
as coming from a single parent-headed household), and more attention to family process
variables (such as monitoring and supervision) in order to advance the field's understanding
of the family's role in the larger context of a "gender-integrated" theory of delinquency
(Hoyt & Scherer).
A significant interaction also was found between gender and the family/parenting
risk scores when prior offense scores became the dependent measure. Men reported
higher levels of prior offenses than women in both the African American and Caucasian
samples when family/parenting risks were at lower levels, a finding that is axiomatic
in the literature on delinquent behavior (Gavazzi, Yarcheck, & Chesney-Lind, in press;
Poe-Yamagata & Butts, 1996). However, far from obvious was the additional finding
that African American and Caucasian women reported relatively equal levels of prior
offenses with the men when family/parenting risk levels were at higher levels. Although
purely speculative at the present time, it is possible that family factors may help
to explain the more recent rise in female delinquency rates. In contrast to the stable
rates of male delinquent behavior, women are entering the juvenile justice system
more frequently and at younger ages than ever before (Poe-Yamagata & Butts).
Hence, more research on family processes in diverse samples of court-involved youth
will make a solid contribution to our understanding of how women may be "catching
up" to their male counterparts in terms of illegal behaviors. The juvenile justice
community already has witnessed the call for a "different-needs" approach regarding
female offenders (Chesney-Lind & Okamoto, 2001) as the pathways that bring female
adolescents to delinquent behavior are thought to be considerably different than
the pathways for men (Belknap, 2001; Funk, 1999; OJJDP, 1998). The family field can
advance these efforts by generating new research and developing more refined therapeutic
interventions in work that targets the families of female adolescent offenders.
>From a research standpoint, it has been argued that there are significant gender
differences in terms of exposure to different types of negative life events, including
those occurring in the context of the family (Mazerolle, 1998). In turn, disruptions
in the family have been thought to differentially impact male and female offenders
(Dornfield & Kruttschnitt, 1992). A critical next step in the research process is
to further identify these family events and examine their association with the risks
and needs reported by court-involved youth.
Clinically, adolescent female offenders often seem to be "harder to work with" as
they typically present in therapy with more numerous and more serious problems (Baines
& Adler, 1996). This includes a range of mental health issues (Dembo, Schmeidler,
Sue, Borden, & Manning, 1995; Salekin, Rogers, Ustad, & Sewell, 1998; Travis, 1999)
that may become particularly pronounced when combined with exposure to negative life
events such as physical and/or sexual abuse (Chandy, Blum, & Resnick, 1996). Coupled
with the fact that female adolescents often are charged with offenses that seem rather
"gender specific" (especially runaway behavior and domestic violence charges), they
may in fact be displaying "survival reactions" to traumatic events in the home (Chesney-Lind,
1997; Chesney-Lind & Shelden, 1998). Hence, family-based professionals are in a position
to advance a therapeutic response to these needs, in contrast to the criminal justice
model that holds the individual entirely accountable and subject to punishment.
The present effort has examined the additional influence of ethnicity on both gender
and family factors. As with gender, the issue of ethnicity has gained more recent
attention in the juvenile justice realm, in large part due to the disproportionately
large numbers of minority youth involved in the court system (DeJong & Jackson, 1998;
Leiber, 1994). Whereas minority overrepresentation raises many questions about the
impact of economic disadvantage and residence in dangerous neighborhoods on the ethnic
composition of juvenile dockets (Peterson, Krivo, & Harris, 2000), meta-analyses
have yielded compelling evidence of racial bias at every point of court entry and
judicial processing point that cannot be explained away by poverty rates and related
community characteristics (Snyder, 1996). In addition, more recent studies of criminal
sentencing in adult populations have generated evidence of a complex relationship
between race, gender, and age (Steffensmeier, Ulmer, & Kramer, 1998). Hence, the
theoretical literature on ethnicity differences likewise may benefit from the present
study's advancement of attention paid to both gender and family process variables.
However, less attention has been paid to the combination of gender and ethnicity,
especially when it concerns the family life of court-involved youth. The findings
of the present study indicate that African American women, in particular, may be
most in need of family-based intervention. Researchers and clinicians both must apply
their efforts toward a better understanding of the family environment that these
female adolescents are facing, as well as what can be done to provide assistance
at a family system level.
This study represents the latest in a series of efforts to document the psychometric
properties of an instrument designed specifically for use with court-involved youth
and their families. Unfortunately, the current state of assessment in the juvenile
justice realm is fraught with several major limitations. Chief among these is how
little published literature is available regarding the psychometric properties of
many measurement tools employed by professionals working in juvenile justice settings.
In turn, the few instruments that do have published reliability and validity evidence
seem to have "trickled down" from the adult criminology literature (Andrews, Bonta,
& Hoge, 1990), instead of having been constructed with adolescent developmental issues
in mind (and attending to issues that concern the families of these adolescents).
Limitations
The GRAD was specifically designed for use with adolescents and their families who
are coming into contact with the juvenile court, and there have been a progressive
publication of articles regarding evidence of its reliability and validity. However,
a rather major limitation in the present study is the lack of parent reports to accompany
data gathered from adolescents. Multiple-family member perspectives would allow for
an examination of the family as the "unit of analysis" (Bartle-Haring & Gavazzi,
1996) in such measurement efforts.
Unfortunately, many court-based professionals utilizing the GRAD presently do not
have a process in place that would allow them to involve parents in the assessment
phase of their work. Several studies are underway that include efforts to obtain
both adolescent and parent GRAD data. However, the gathering of multiple-family member
perspectives runs counter to the environment of the typical juvenile court. In actuality,
the typical juvenile justice assessment instrument privileges the viewpoint of the
professional, not the youth, parents, or other relatives, stemming in large part
from the belief that offenders and their families will lie in order to protect themselves
from prosecution (Hamilton, Brantley, Tims, Angelovich, & McDougall, 2001).
In addition, family scholars must remain mindful of the rather widespread belief
in the juvenile justice realm that the utility of any assessment tool ultimately
is predicated on its ability to predict recidivism rates. In fact, because some criminology
studies have contended that extralegal factors are not necessary to predict recidivism
rates at all (Hoge, 2002), there is a tendency for family-based efforts to be pushed
to the periphery of the juvenile justice system. Hence, members of our field must
redouble their efforts to generate family-based evidence that professionals working
in the juvenile justice field will find useful. Whereas one study underway in our
own research program includes an effort to examine the multitude of ways in which
family factors assessed by the GRAD are associated with future criminal activity
(recidivistic behavior), having data on prior offenses only represents a significant
limitation on the present data, in that evidence of the association between family
processes as measured by the GRAD and delinquent behavior is based on past criminal
behavior.
Further, it was noted above that cut-off scores have been set according to the idiosyncratic
characteristics of each court employing the GRAD. Whereas this is a comfortable arrangement
to strike at the outset of this assessment tool's development, an important next
step will be to address issues related to "cross-jurisdictional validity" (Heilbrun
et al., 2000). In essence, the geographical diversity of new courts beginning to
utilize the GRAD will mandate an examination of this measure's generalizability across
samples of youth and families coming into contact with these courts.
More information also must be generated about the potential variability of scores
due to characteristics of the professional employing the GRAD. For instance, no studies
of interrater reliability have been published as yet. Also, more rigorous attention
must be paid to the ways in which information generated by the GRAD is utilized to
make appropriate referrals, building on the initial evidence that this assessment
tool enables its users to connect court-involved youth and their families to mental
health services (Gavazzi, Lim, et al., 2003).
One final limitation to be mentioned here concerns the general lack of information
about the degree to which information generated by the GRAD is useful to family members
themselves. It was mentioned at the beginning of this article that the GRAD was built
on previous family-based intervention work that followed a psychoeducational format
(Gavazzi, Wasserman, et al., 2000). A similar use of GRAD data to help family members
compare their own risks and needs with that of others facing similar circumstances
has not been examined to date.
Families involved with the juvenile courts are often both "wary" of helping professionals-because
they have been given the message that diey are to blame for their youth's misdeeds-and
"weary" of the demand for further involvement of services that to date may have been
ineffective in meeting their family's needs (Partridge et al., 2001). "Parent pages"
that explain GRAD scores in lay terms have been constructed (Gavazzi et al., 2004)
with this wariness and weariness in mind. In essence, work now must be undertaken
that ascertains the degree to which these pages are helpful in getting families to
the most appropriate services that they need and deserve.
Conclusions
The breakdown of traditional social welfare services continues to generate an emerging
"gap in services" that, in turn, increases the demand for juvenile courts to provide
or otherwise direct families to services that target these extralegal factors (Gavazzi,
Yarcheck, Edelblute, & Webb, 2005). As a result, more and more juvenile justice professionals
are referring to their courts as "emergency rooms" (Gavazzi, Lim, et al., 2003) that
are charged with the task of triaging youth and families to services that address
these various needs. Hence, the demand for assessment instruments that effectively
measure these global risks has never been greater.
In sum, the GRAD represents a psychometrically sound measure that can be of use to
researchers and clinicians who work with the families of courtinvolved youth. Increased
attention to the impact that gender and ethnicity have on the risks and needs of
these families clearly is warranted. Such efforts may contribute to the increased
"family acumen" of the juvenile justice realm, which in turn will generate greater
consideration of family-based research and intervention efforts that target courtinvolved
youth and their families.
-1-
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication Information:
Article Title: Gender, Ethnicity, and the Family Environment: Contributions to Assessment
Efforts within the Realm of Juvenile Justice. Contributors: Stephen M. Gavazzi -
author. Journal Title: Family Relations. Volume: 55. Issue: 2. Publication Year:
2006. Page Number: 190+. © 2006 National Council on Family Relations. Provided by
ProQuest LLC. All Rights Reserved.
Next Page

Jessie Rayl
EM: thedogmom63 at frontier.com
PH:304.671.9780
www.facebook.com/eaglewings10

"But they that wait upon the LORD shall renew their strength; they shall
mount up with wings as eagles. They shall run, and not be weary"--Isaiah 40.31
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.acb.org/pipermail/acb-hsp/attachments/20110624/7401eee4/attachment-0001.html>