Language: critical components in readers with criminal referral history.

Introduction

Reading is the "single most important skill" that helps youth succeed in life academically, socially, and vocationally (Leone, Krezmien, Mason, & Meisel, 2005, p.239). When students do not become proficient readers in elementary school, they are at risk for problem behaviors as well as future delinquency (Daal, Verhoeven, & Balkom, 2007, Harris, Baltodano, Artiles, & Rutherford, 2006, Rivera, Al-Otaiba, & Koorland, 2006, Leone, et al. 2005). VanderStaay (2006) reviewed longitudinal research on relationships between reading and criminal, delinquent, and antisocial behaviors and concluded that reading and the beginning, ongoing, and escalating of antisocial behaviors are a "key public health concern" (p. 336). In response, Federal legislation and public policies are in place that mandate schools more effectively teach youth at risk with reading problems. In both Individuals with Disabilities Education Act (IDEA) and No Child Left Behind (NCLB), practitioners are required to implement research-based practices [(614) (d) (1) (A) (iv)]. This emphasis is critical since it means that practitioners must locate the best and most effective practices for teaching reading, rather than rely on their own styles or knowledge that may or may not encompass practices that work for students with behavior problems.

In particular, the newly authorized special education law (IDEA 2004), moves the field of special education from a deficit oriented perspective in which children have to fail over a significant period either behaviorally or academically before they were identified for and placed in special education programs. The more traditional deficit model has transitioned into a new prevention-oriented, early-intervening approach to learning called Response to Intervention (RTI). RTI has the promise of offering earlier supports to at risk youth, and potentially, preventing some at risk students from moving into offender status [(614) (b) (6) (A-B)]. This is particularly true since there is evidence that behavior and academic performance as interrelated and that at risk behavior emerges from poorly developed academic and learning skills and vice versa (Morgan, Farkas, Tufis, & Sperling, 2008).

The National Reading Panel (2000) identified core skills for reading achievement and concluded that the critical components are phonemic awareness, phonics, fluency, vocabulary, and text comprehension. The components of teaching reading are simple to break down at the early stages of education as students are taught specific reading skills; but as they grow older, reading proficiencies and deficits become more complex; thus, reading problems are more difficult to remediate with time. Decades of research show that behavior problems often co-occur with reading problems (Daal, Verhoeven, & Balkom, 2007; Harris, Baltodano, Artiles, & Rutherford, 2006; Rivera, Al-Otaiba, & Koorland, 2006; Leone, et al. 2005). One recent study by Daal et al. (2007) found that phonology problems were associated with a higher incidence of external behavior problems, specifically, aggressive and delinquent behaviors in five year olds. If appropriate reading and language assessment and intervention are not implemented early, behavior problems may overshadow academic deficits.

School is an academic and social setting and reading encompasses the student's social and cultural understandings of the world, particularly when addressing the semantics or meanings of words. Frustrated readers are more likely to give up on academic success and seek alternative definitions of the purpose of schools. When this happens, school focus can shift from academic performance to over-emphasis on developing social, athletic, or anti-social networks that have more meaning and value to students who fail or struggle academically. For instance, if academic performance causes students to feel inadequate, one reaction is to overcompensate socially. This can be accomplished through acting out or withdrawing. Delinquency becomes a different means of maintaining social status and successful relationships outside the expected norms. Students develop their identities through socialization with others. Some students who develop poor reader identities early on may choose to develop their social status by disassociating with academic success. Since their reading performance brings social stigma, they disengage from academic learning because they have difficulty reading, a skill which should have been learned at an early stage of school. Students lack the foresight and understanding that failing to read will become a means of social isolation as they become adolescents and have opportunities to enter the work force or continue their education.

Students struggling academically get educational help if reading problems are recognized in addition to their behavior problems. In some systems, the learning disability (LD) category brings a set of services designed to improve students' access to the general education content. In contrast, students who are identified as having emotional and/or behavior disorders (EBD) likely receive services designed to improve their behavioral outcomes, often with little attention and effort placed on building academic skills (Sheila, Ellen, Lara, & Christa, 2007). From students' perspectives, these labels identify them as being different, less capable, and less valued than other students. Students often drop out to avoid the stigma of these labels, thus isolating them even more. By opposing schooling and the stigma of school labels, students who drop out place themselves in direct opposition to societies expected progression towards adulthood. Such resistance and opposition heightens their risk for internalizing and externalizing behaviors such as withdrawal, anxiety, depression, aggression, and delinquency (Daal, et al., 2007), which can then lead to a host of self-defeating life choices.

Extensive research with students who have major behavior problems and who stay in school shows that EBD, language deficits, and reading disabilities often co-exist. Recently, Benner and his colleagues (2002) systematically reviewed the research specific to EBD and language skill deficits. In their review of 18 studies, they found almost three quarters of youth formally identified with EBD had significant language problems. They also found four elements that were prevalent across all the studies: (a) students with language problems are at high risk for antisocial behavior; (b) students with language problems have high rates of behavior problems that often go undetected; (c) students' antisocial behaviors may increase over time; and (d) these characteristics affect interpersonal relationships with peers, authority figures, family, as well as society in general (Benner, Nelson, & Epstein, 2002). It becomes more difficult to detect reading skill deficits unless youth with EBD or who act out are continuously assessed academically and evidence-based interventions designed for this population are available.

Further, researchers found that youth with limited skills in language functioning used more physical actions to solve interpersonal problems (Benner et al. 2002). Youth in the justice system have higher prevalence of EBD (47.7%) compared to this population in public schools (Quinn, Rutherford, Leone, Osher, & Poirier, 2005), early detection and prevention of reading and language difficulties may help some students avoid associated behaviors that could later place them in the justice system. A consistent finding is that incarcerated youth are several years behind non-incarcerated youth in reading achievement (Jerse, 1978; Brunner, 1993; Johnson, 1999; Foley, 2001). Close to 15 years of research in juvenile justice continue to show averages that incarcerated 15 and 16 year old youth read at the fourth or fifth grade level (Brunner, 1993, Harris et al., 2006). In a longitudinal study, VanderStaay (2006) found that behavior and reading problems constitute major risk factors for continued involvement in delinquency. Further, Brunner (1993) reported on an experimental study of a Juvenile Justice Literacy Program in Orange County, California that focused on reading instruction with incarcerated youth. He concluded if youth participated in 50 or more hours of direct reading instruction significant reading gains could be accomplished, recidivism could be decreased by 20%, and costs were reduced (for every dollar invested in reading saved the community $1.75 in reduced recidivism). The overrepresentation of youth in the juvenile justice system with reading problems needs more in-depth exploration and a deconstructing of the relationship between delinquency and reading for better understand in designing interventions.

This study examines youth that are past the early reading intervention stages (grades K-3) and may or may not have been identified for special education services. These youth were at a critical stage when they were first incarcerated in detention centers and had behaviors that escalated from school or family rule breaking to a law breaking cycle which included physically acting out or violating other's property. With early detection and intervention at this stage, researchers suggest that further involvement in the justice system can be diverted and increased engagement with a transition back to school or the workforce can be attained (Brunner, 1993; Johnson, 1999; Foley, 2001, Leone et al., 2005, VanderStaay, 2006). To build such interventions, we must know more about the specific reading and language skills that this set of youth bring to detention centers.

Method

Setting and Sample

This study took place in a secure urban detention facility in the Southwestern United States. Within the juvenile justice system, there were two types of incarceration facilities: corrections and detention. In corrections or long-term facilities youth are incarcerated for extended periods of time for treatment and services. In detention or short term facilities youth are incarcerated for their initial involvement with the justice system and wait for a judge to impose short term treatment or consequences for their criminal offenses. McGlynn (2003) reported that, on average, youth were detained nine times at the detention level before being sentenced to corrections. Youth are detained after they have broken the law and the police have presented the referral or warrant. Youth are then screened to assess their risk to the community. If they are deemed to be a threat, based on past history and the current referral, they stay detained. Forty percent of these youth were brought to detention with a new police referral for breaking the law. The other sixty percent had a past referral that needs to be taken care of such as a warrant for not appearing for their court date. It could also be a violation of probation, for example, not complying with court ordered sanctions such as missing school, drug use, not completing work hours (Maricopa County Data Book, 2002). According to state statute, the youth are seen by a judge within 24 hours after being arrested for a determination of required detention. If it is decided they should remain detained, they will typically stay 30 to 60 days, depending on their court processes.

This was a convenience sample of 47 juveniles among the 184 who were detained. Research in correctional facilities is difficult to conduct because the institutional focus is typically on safety and controlling behavior, therefore reading assessments are limited with no universal academic standards of assessments. Limited access to incarcerated youth makes research with this population problematic (Quinn, et al., 2005). When access is attained, the minimal educational focus further adds to the challenge of this type of research. The descriptive data were collected during intake by detention staff and include both demographics and court documentation (of current and past behaviors). Variables collected were age, ethnicity, number of times detained, number of days detained, and number of prior referrals. These variables were studied with the student performance on the administered reading assessment.

Measurement

The academic reading assessment used was New Century Education (2006), a computer generated assessment that measured students' reading levels, specifically, literal comprehension, inference, main idea, and vocabulary/language. The validity of the New Century assessment was compared to the Gates-MacGinitie Test of general reading achievement by the test manufacturer. The test manual reports that a mean grade equivalents were run on both measures and the difference was significant with a modest effect size [t (184) = 3.35, p= .001, d=.48] using a one-way repeated measure analysis of variance on 186 third and fourth graders. New Century yielded a slightly higher mean average 3.6 to 3.24 and 4.47 to 4.12. The relationship was examined with a substantial linear association using Pearson Product r= .76 (p=<.001) concluding, that New Century was likely a valid instrument is measuring general reading levels slightly inflated. F. J. Boster (personal communication to Jim Griffin, April 19, 2004). New Century is an integrated instructional system that has been used throughout the country for over forty years. The reading curriculum is based on state standards and uses a balanced approach between phonics and whole language philosophies. For more than twelve years, the New Century assessment was the initial academic evaluation used to place youth in appropriate reading instruction in the detention school.

The assessment measures four elements of reading: literal comprehension, inference, main idea, and vocabulary/language. Items range in difficulty from first to tenth grade levels. At the initial stages of the assessment, the student sees and hears questions that measure his or her overall reading abilities. The test used both visually and auditory modes to assess reading abilities. If the student correctly answered a question the difficulty level increased and if they answered incorrectly the difficulty level decreased. At the completion of the assessment, a reading level was reported and then a series of activities and assignments were provided. The results of the initial reading assessments were used in this study.

Results

All of the subjects in the study were male and the ethnic representation included: 21 (45%) Anglo; 19 (39%) Hispanic; 5 (11%) African American; and 2 (4%) Native American. This distribution of ethnicity is typical for this detention facility. The youth varied widely in their experience with the juvenile justice system from early delinquency to extensive criminal histories and from low reading levels to very high (see Tables 1 & 2).The mean age was 16.2. The number of times youth in the sample were detained ranged from their first time to ten times (M = 3.51). The number of referrals ranged from one to eighteen (M = 6.6).

Table 1. Descriptive Statistics of Detained Youth and Reading Levels
(n=47)

                  Minimum  Maximum   Mean    Std. Deviation

Age                12.00    18.01   16.1840      1.4993
Times Detained       1       10       3.51        2.40
Days Detained        6       276     102.94       69.23
Referral History     1       18       6.60        4.41

Table 2. Reading Grade Equivalent Levels on the New Century Assessment
for the sample of 47 detained youth

                       Minimum  Maximum  Mean  Std. Deviation

Literal Comprehension     0        9     5.85       2.00
Inference                 0       10     5.19       2.22
Main Idea                 0       10     5.22       2.17
Vocabulary/ Language      0        8     4.94       1.32

The mean reading levels were similar to national findings that, on average, youth in corrections read at the 5th grade level (Leone et al., 2005; Harris et al., 2006). The average reading grade level of the study sample was 5.3 (5.85 literal comprehension, 5.19 inference, 5.22 main idea, and 4.94 language). However, as the descriptive statistics in Table 2 show, there was great variance in the reading levels of the 47 youth in the sample from 0 to 10th grade. A partial correlation was performed with the variables controlling for the others. A bivariate correlation examined the strength of the relationship between each variable. However, no significant relationships were found between age, ethnicity, the number of times youth were detained, the length of their detainment, or any of the reading components.

A multiple regression analysis was conducted to evaluate how well the reading components predicted referral history. The predictors were the four reading assessment scores, while the decisive factor (or dependent variable) was referral history. In Table 3, the linear combination of the four reading components was significant as related to referral history, F (4, 42) = 2.64, p < .05. In a linear model, reading scores accounted for 12% of the unique variance in the number of referrals to the juvenile justice system. The correlation coefficient in Table 4 and the slopes for each reading component in Figure 1, demonstrate a significant relationship between language skills and referrals; that is, youth who scored lower on language were likely to have more referrals. There were no significant relationships among the other reading components (literal comprehension, inference, or main idea) with referrals.

Table 3. The Bivariate and Partial Correlation of the Predictors with
Referrals

Reading Predictors     Correlation between   Correlation between
                            predictors          each predictor
                       and referral history  and referral history
                                             controlling for all
                                               other predictors

Literal Comprehension          -.10                  .21

Inference                      -.11                  .07

Main Idea                      -.10                  .04

Vocabulary/ Language           -.31 *               -.44 *

* p<.01

Table 4. Correlation among the four reading components and referrals

                                     Literal
                       Referrals  Comprehension  Inference  Main Idea

Literal Comprehension    -.102

Inference                -.116       .864 **

Main Idea                -.109       .840 **      .860 **

Vocabulary/ Language     -.314 *     .847 **      .814 **    .783 **

* p< .05 ** p< .001

[FIGURE 1 OMITTED]

Limitations and Future Research

As with any research the limitations of this study should be considered. First, this research had a small sample of all male youth detained at one time point. These findings would be strengthened by analyzing reading and detention referrals over multiple time points. Future research on the language and reading needs of this population should consider both male and female samples to examine if gender differences occur. Second, this study used only one reading assessment in the analysis. Other standardized measures of reading ability would enrich the current findings. Future studies should consider using multiple measures of reading abilities and, in particular, language abilities. Finally, referral history was not separated by violations of probation (breaking court mandates) and police referrals (breaking the law). This separation could provide further information in defining noncompliance behavior as compared to continued law breaking.

Discussion

This study found a significant relationship between youths' prior history of referral and language skills but no significance was found for skills in literal comprehension, inference, or main idea. All of the reading components had a downward slope; however, language was more prominent and significantly correlated with referral history (see Figure 1). This provides additional evidence for the relationship between language skills and behavior problems in adolescents and, particularly for youth in the early stages of criminal behavior. This research furthers the work of Daal et al.'s (2007) where they studied and related phonology to early external behavior problems with 5-year-olds. Over time and with escalating behavior problems, in this case criminal referrals, there is a constant relationship with language deficits.

There is limited research in the field of incarcerated youth and reading interventions to date. There have only been five published research articles of reading interventions in correctional settings (Hodges, Giuliotti, & Porpotage, 1994; Campbell, Marsh, & Stickel, 1993, Malmgren & Leone, 2000; Drakeford, 2002; Harris, et al., 2006). Four of the five research interventions were conducted in juvenile corrections/long term facilities. There need to be more at the early stages of juvenile justice involvement such as in detention. Malmgren and Leone (2000) conducted a reading intervention in a short term detention and concluded that reading levels did significantly improve with a short, intense intervention. They could not pinpoint the particular intervention component that was responsible for these gains but found significant improvement did occur with an intensive reading program in a short term facility.

With so few studies focusing on youth in juvenile justice facilities and even fewer examining evidence-based practices, this area would benefit from more studies analyzing programs aimed to specifically strengthen reading skills (Harris et al., 2006) including language skills as an important component. There is a need to continue to explore intense immersion in language instruction in detention and correctional facilities to possibly reduce future referrals. Recommendations of needed research in correctional facilities emphasized the need of court involved youth to improve language skills as an on-going effort to improve their reading ability (Leone et al., 2005). There is a significant relationship between youth incarcerated with below grade reading and recidivism and Brunner (1993) asserts that there is ample evidence to show a link between recidivism and reading problems, this is a valuable line of inquiry. Teaching language skills will not correct behavior problems as there are additional ecological aspects to youth behavior. However, additional studies need to be done in exploring the relation, especially at early stages, between reading/language deficits and behavior problems/increased crime. This relation can facilitate the understanding of the specific language components that have an impact on behavior.

References

Brunner, M. S. (1993). Reduced recidivism and increased employment opportunity through research-based reading instruction. (NCJ Publication No. I41324). Washington, DC: Office of Juvenile Justice and Delinquency Prevention.

Benner, G. J., Nelson, J. R., & Epstein, M. H. (2002). The language skills of students with emotional and behavioral disorders: A literature review. Journal of Emotional and Behavioral Disorders, 10, 43-59.

Campbell, D. M., Marsh, D. T., & Stickel, K. (1993). Intervention with adjudicated adolescents: Academic and psychological effects. Journal of Offender Rehabilitation, 19(1-2), 101-111.

Daal, J., Verhoeven, L., Balkom, H. (2007). Behavior problems in children with language impairment. Journal of Child Psychology and Psychiatry, 48, 1139-1147.

Drakeford, W. (2002). The impact of an intensive program to increase the literacy skills of youth confined to juvenile corrections. Journal of Correctional Education, 53, 139-144.

Foley, R. M. (2001). Academic characteristics of incarcerated youth and correctional education programs: A review. Journal of Emotional and Behavioral Disorders, 9, 248-260.

Harris, P. J., Baltodano, H. M., Artiles, A. J., & Rutherford, R. B. (2006). Integration of culture in reading studies for youth in corrections: A literature review. Education and Treatment of Children, 29, 749-778.

Hodges, J., Giuliotti, N., & Porpotage II, F.M. (1994). Improving literacy skills of juvenile detainees. Juvenile Justice Bulletin. (NCJ Publication No. 150707). Washington, D.C.: Office of Juvenile Justice and Delinquency Prevention.

Individuals with Disabilities Education Improvement Act of 2004. H.R. 1350, 108th Cong. (2005).

Jerse, F. W. (1978, Winter). Juvenile delinquency and academic deficiency. Contemporary Education, 49, 106-109.

Johnson, R. (1999). Destiny's child: Recognizing the correlation between urban education and juvenile delinquency. Journal of Law and Education, 28, 313-317.

Leone, P. E., Krezmien, M., Mason, L., & Meisel, S. M. (2005). Organizing and delivering empirically-based literacy instruction to incarcerated youth. Exceptionality, 13, 89-102.

McGlynn, M. (2003). Short-term transition of youth leaving secure care. Unpublished Doctoral dissertation, Arizona State University, Tempe, Arizona.

Malmgren, K.W., & Leone, P. E. (2000). Effects of a short-term auxiliary reading program on the reading skills of incarcerated youth. Education and Treatment of Children, 23, 239-248.

Maricopa County Data Book (MCDB). (2002). Maricopa County Juvenile Probation Department, Superior Court of Arizona: Comparison Data 1998 to 2002.

Morgan, P. L., Farkas, G., Tufis, P. A., & Sperling, R. A. (2008). Are reading and behavior problems risk factors for each other? Journal of Learning Disabilities, 41, 417-436.

National Reading Panel. (2000). Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction (NIH Pub. No. 00-4769). Washington, D.C.: National Institutes of Health.

New Century Education Corporation (2006). New Century Education. Retrieved November 04, 2006, from http://www.ncecorp.com/.

Linares-Orama, N. (2005). Language-learning disorders and youth Incarceration. Journal of Communication Disorders, 38, 311-319.

Quinn, M.M., Rutherford, R.B., Leone, P.E., Osher, D.M., & Poirier, J.M. (2005). Students with disabilities in detention and correctional settings. Exceptional Children, 71(3), 339-345.

Rivera, M. O., Al-Otaiba, S., Koorland, M.A. (2006). Reading instruction for students with emotional and behavioral disorders and at risk of antisocial behaviors in primary grades: review of literature. Journal of the Council of Children with Behavioral Disorders, 31, 323-339.

Sheila, R. A.-M., Ellen Matheson, R., Lara, L. A., & Christa, M. M. (2007). Effects of Repeated Readings, Error Correction, and Performance Feedback on the Fluency and Comprehension of Middle School Students With Behavior Problems. The Journal of Special Education, 41(1), 17-30.

VanderStaay, S. L. (2006, July/August/September). Learning from longitudinal research in criminology and the health sciences. Reading Research Quarterly, 41, 328-350.

Biographical Sketches ____________________________________________________

DERRICK PLATT PH.D., Juvenile Probation Officer for 18 years, received his interdisciplinary Special Education Ph.D. at Arizona State University with a focus on Juvenile Justice and Education. He was also a Research Assistant with Education Disabilities and Juvenile Justice (EDJJ).