A comparative analysis of language, suspension, and academic performance of students with emotional disturbance and students with learning disabilities.
Article Type:
Disabled students (Behavior)
Disabled students (Education)
Academic achievement (Psychological aspects)
Students (Rating of)
Goran, Lisa G.
Gage, Nicholas A.
Pub Date:
Name: Education & Treatment of Children Publisher: West Virginia University Press, University of West Virginia Audience: Professional Format: Magazine/Journal Subject: Education; Family and marriage; Social sciences Copyright: COPYRIGHT 2011 West Virginia University Press, University of West Virginia ISSN: 0748-8491
Date: Nov, 2011 Source Volume: 34 Source Issue: 4
Geographic Scope: United States Geographic Code: 1USA United States

Accession Number:
Full Text:

This study focused on the relationship among language, behavior, cognitive ability, and academic performance constructs for school-aged students identified with educational disabilities. The authors provide a review of research findings in regard to the language and academic deficits of students with emotional disturbance (ED) and specific learning disabilities (LD). Using an extant school database and structural equation modeling, the interrelationship among the constructs were examined, finding language to be a significant predictor of cognitive ability and academic performance, but not behavior, and that no differences were evident between the disability groups except on the behavioral measure.

Key Words; emotional disturbance, learning disabilities, language, academic, behavior, high-incidence disabilities

Students identified with emotional disturbance (ED) and specific learning disabilities (LD) share a common core characteristic: the absence of cognitive impairment (IDEIA, 2004). Aside from cognitive similarities, the two categories are typically associated with two different performance related manifestations of disability: academic skill deficits for students with LD and social and behavioral deficits for students with ED. However, recent research has highlighted that students with ED exhibit low academic achievement across core content areas (Lane & Menzies, 2010; Kauffman & Brigham, 2009) and increased rates of language disorders (Benner, Nelson, & Epstein, 2002), both characteristics often associated with LD (Fletcher, Morris, & Lyon, 2003; Mann, 2003; Pullen, Lane, Ashworth, & Lovelace, 2011).

With such high rates of co-morbidity, the interrelationship among language deficits, academic performance, and behavior problems warrants further investigation. Language is an important part of academic and behavioral performance, suggesting that poor language skills may predict poor academic and behavioral performance (Tomblin, Zhang, Buckwalter, & Catts, 2000). Based on this assumption, language deficits may be present in both students with ED and students with LD, as both exhibit academic performance deficits, yet no research has specifically examined similarities and differences in academic, behavioral, cognitive, and language constructs between the two disabilities. To address this gap, we examined the interrelationship among the constructs between the ED and LD disability categories. First, we review the research on academic performance and language deficits for students with ED, followed by a brief overview of the research for students with LD, focused on reading-related disabilities as they are most often associated with potential language deficits. Last, based on the research, the hypotheses tested in this study are discussed.

Children and youth with ED present many challenges, both internalizing (e.g. social withdrawal and depression) and externalizing (e.g. aggression and noncompliance) behavioral manifestations, as well as academic and social difficulties in schools (Kauffman & Landrum, 2008). Long-term outcomes for youth with ED are troubling, including academic failure, increased dropout rates, high unemployment, low participation in postsecondary education, increased social isolation, and increased levels of juvenile and adult crime (Bradley, Doolittle, & Bartolotta, 2008; Maag & Katsiyannis, 1998). ED, as defined by the IDEIA (2004), focuses on the interconnection of behavioral, academic and social deficits of students, and defines the disability as an inability to learn, an inability to build or maintain satisfactory relationships, and inappropriate behaviors under normal circumstances that adversely impact educational performance. Students with ED exhibit academic deficits in addition to behavioral concerns, yet etiological bases for the academic deficits are still developing (Trout, Nordness, Pierce, & Epstein, 2003). In addition, research suggests that between 50-70% of students with ED have co-morbid speech, language, and/or communication disorders (Benner et al., 2002; McCabe & Meller, 2004). Are academic difficulties a result of the manifestation of the behavior problems, either internalized or externalized, or are the difficulties related to language deficits that are concomitant with either type of behavior (Kaiser, Hancock, Cai, Foster, & Hester, 2000)? In other words, for students with ED, are academic and behavioral deficits related to language deficits, such as poor receptive or expressive language skills?

Academic and Language Deficits for Students with ED

Academic deficits for students with ED have been well documented in the literature (Reid, Gonzalez, Nordness, Trout, & Epstein, 2004), and this has contributed to an increase in research addressing academic performance for this population over the past decade (Gage, Lewis, & Adamson, 2010). In a meta-analysis of academics and students with ED, Reid and colleagues (2004) found that most students with ED performed at or below the [25.sup.th] percentile in general academic functioning and found a moderate to large negative effect difference (-.69) between students with ED and nondisabled students. Recent research corroborates these findings, including Lane, Barton-Arwood, Nelson, and Wehby's (2008) study of 42 students with ED educated in self-contained or segregated classroom settings, finding that the sample performed below the [25.sup.th] percentile on all math, reading and written expression measures. In addition, Nelson, Benner, Lane, and Smith's (2004) study of 155 K-12 grade students with ED found that 83% of the participants scored below the mean on the Total, Broad Reading, Broad Math, and Broad Written Language composites of the Woodcock Johnson III Test of Achievement (Woodcock, McGrew, & Mather, 2001).

Benner, Nelson, and Epstein (2002) examined 26 studies (n = 2796) investigating students' with ED language skills, finding that 71% of students with ED had co-occurring language deficits; and that 57% of students with diagnosed language disorders also were identified as having ED. More specifically, the review found that 71% of students with ED had pragmatic language deficits, 64% had expressive deficits, and 56% had receptive deficits. These findings, along with the findings of other studies (Gallagher, 1999; Hummel & Prizant, 1993; Nelson, Benner, & Cheney, 2005; Rogers-Adkinson, 2003; Ruhl, Hughes, & Camarata, 1992) suggest that students with ED exhibit language skill deficits. However, questions still remain as to the nature of the interaction among language, behavior, and academic performance.

Studies have found mixed results in regard to behavioral manifestations (i.e. internalizing and externalizing) and language and academic outcomes. Gresham, Lane, MacMillan, and Bocian (1999) found that students with internalizing behaviors were less academically competent and scored lower on reading and math assessments than students with externalizing behavior. However, students exhibiting both internalizing and externalizing behaviors performed significantly worse than nondisabled students across all academic assessments. Nelson, Benner, and Cheney (2005) examined the relationship between internalizing and externalizing behaviors and language abilities in students with ED, finding that although 90% of students with ED exhibit language deficits (Benner, et al., 2002), those with externalizing behaviors were more likely to have language deficits than students with internalizing behaviors. As the research associated with students with ED and academic and language deficits increases (Hodge, Riccomini, Buford, & Herbst, 2006; Kostewicz & Kubina, 2008; Mattison, 2008), interventions to target etiological bases, such as language deficits, will also increase (Nelson, Benner, Neill, & Stage, 2006a).

Academic and Language Deficits for Students with LD

A learning disability is defined in IDEIA (2004) as a disorder in one or more of the basic psychological processes involved in understanding or using language, spoken or written, which may be manifested as an inability to listen, think, speak, read, write, spell, or do math. Since the understanding or use of language is a primary component of definition of LD, it is logical that language skill deficits often are noted in conjunction with reading and reasoning skills for students with LD (Beck, McKeown, & Kucan, 2002; Gersten, Fuchs, Williams, & Baker, 2001; McArthur, Hogben, Edwards, Heath, & Mengler, 2000; McKeown & Beck, 2004; Scarbourough, 2001). McArthur and colleagues, in an examination of the similarities between students identified with LD and students identified with language impairments, determined that over half of each group could have just as easily been classified in the other category as they met the eligibility requirements (McArthur, et al., 2000). This overlap in groups can be attributed to the language skills that are embedded within learning tasks such as reading and reasoning. As further support for the overlap in identification, the researchers found that 55% of children with reading difficulties have impairments in oral language and 51% of children with language deficits have a concomitant reading disability.

Scarborough (2001) highlighted the fact that reading comprehension skill deficits of students beyond [2.sup.nd] grade are essentially oral language limitations. These limitations include the child not knowing the word in spoken form, showing an inability to parse semantic and/or syntactic relationships among words, and lacking crucial background knowledge to correctly interpret the text (Scarborough, 2001). The role of language as a component, or underlying factor, of reading skills such as vocabulary knowledge and text comprehension has been described as strong and reciprocal, but is not fully understood (Beck, et al., 2002; Gersten, et al., 2001; McKeown & Beck, 2004).

Continuing with the reciprocal nature of the relationship between language and other areas of disability, it is documented in the literature that students with language deficits are at risk for reading disability and behavior disorder (Catts, 1993; Catts, 1997; Catts, Fey, Tomblin, & Zhang, 2002; Tomblin, Zhang, & Buckwalter, 2000) and that students with LD and students with ED demonstrate language skill deficits (Benner, Nelson, & Epstein, 2002; Lyon, et al., 2001; McArthur, Hogben, Edwards, Heath, & Mengler, 2000; Nelson, Benner, Neill, & Stage, 2006; Scarborough, 2001; M. Wagner, Kutash, Duchnowski, Epstein, & Sumi, 2005; R. Wagner, Francis, & Morris, 2005). However, the generality of this statement and the scattered approach to investigating language deficits from the various perspectives do not provide a clear picture of which specific language skills are impacted for each population.


This study was designed to identify the relationship among language, history of suspension, academic, and cognitive constructs, or overall performance as measured by observed events (available test scores), for students with school-identified ED and LD using structural equation modeling (Kline, 2005). Like students with ED, students with LD have educational performance deficits that are not the result of cognitive impairment, making the two groups ideal for comparison to identify similarities and differences in language skill, academic performance, cognitive ability, and behavior. We utilized school archival records from the districts electronic database and developed a statistical model to address the following hypotheses based on the literature:

* Hypothesis 1: Students (with ED and LD) with greater language deficits will have more incidence of suspension.

* Hypothesis 2: Students (with ED and LD) with greater language deficits will also display lower performance in both cognitive ability and academic performance.

* Hypothesis 3: Students (with ED and LD) with higher cognitive ability will have higher academic performance.

* Hypothesis 4: Students' disability status (ED versus LD) will significantly impact their performance in the areas of language, academics, and history of suspension

Figure 1 provides a visual depiction of the hypothesized model.



We conducted a secondary analysis of an extant database of a medium sized Midwestern city's student assessment records from the 2008-2009 school year. Total student population for the district was 17,419 students, with 2,316 students receiving special education services. Of the 2,316 students receiving special education, this study only analyzed data for students who were identified by the district as LD and ED and for whom there were complete records for the measures used in the analysis.


Application of the inclusion criteria resulted in a total sample of 142 students, 17.6% (n = 25) with ED and 82.4% (n = 117) with LD. The low sample of students with ED was primarily related to a lack of available language assessment data for students with ED. Although the sample was disproportionate, the analysis was conducted because (1) the model utilized disability as a dichotomous predictor variable and the model fit was not indicate error of prediction, and (2) the standard deviation of the ED sample across all measures was very close to the standard deviation of the LD sample, suggesting an absence of overt bias. Students with ED ranged in age from 5-13 years, with a mean age of 8.84; students with LD ranged in age from 6-14 years, with a mean of 9.92. Table 1 provides a breakdown of other key demographic descriptive statistics of the samples. Of note is the large percentage of females in the ED sample (40%). In an analysis of national data sets, Wagner and colleagues (2005) found that for students with ED, males typically represent more than three-fourths of the population. In addition, 60% of the current ED sample was African-American, which is a much greater percentage than found in the national averages for students with ED (Wagner, et al., 2005). Socio-economic status (SES) measures of the ED sample are concomitant with the results of Wagner, et al. (2005). These sample limitations will be further discussed below.

Construct Definitions and Measures

Three standardized measures were used to represent three of the four constructs (language skills, cognitive abilities, and academic performance) and the remaining measure, used to represent behavior, was developed from the database. Each of the constructs represents core areas of interest for understanding similarities and differences between students with ED and LD.

Language skills. The construct of language skills refers to one's ability to understand and use words effectively orally and in writing (Nelson, Benner, Neill, & Stage, 2006b; Owens, 2001). Although other language assessments were used in the district, the most prevalent standardized measure was the Clinical Evaluation of Language Fundamentals, Fourth Edition (CELF-4; Semel, Wiig, & Secord, 2003). The CELF-4 is an individually administered test for use with children and youth between the ages of 5 and 21 years of age that takes approximately 30-60 minutes to complete. Composite scores are provided for core language, receptive language, expressive language, language structure, language content, language memory, and working memory. Test-retest reliability coefficients for the CELF-4 ranged.88 to.92 for composite scores and split-half reliability ranged from.87 to.95 for composite scores. Due to incomplete data, only the expressive and receptive composite scores were used in this analysis.

Cognitive ability. The cognitive ability construct in this study refers to one's ability in reasoning and problem solving. The most widely used assessment of cognitive ability within the sample district was the Cognitive Abilities Test (CogAT; Lohman & Hagen, 2001). The CogAT is designed to assess abilities in reasoning and problem solving using verbal, nonverbal, and quantitative skills. The CogAT provides scores for three individual test batteries: verbal, non-verbal, and quantitative. Administration of all three test batteries takes approximately 90-minutes, and results in an overall score jointly scaled using a Rasch model for construct verification that was updated in 2000 on a standardization sample 115,133 subjects. This study utilized all three test battery scores and the composite score.

Academic performance. The academic performance construct in this study refers to one's performance in acquiring and using curricular content. To assess academic performance, we examined students' performance on the Missouri Assessment Program (MAP). The MAP is a state-mandated grade-level assessment based on the Missouri state standards for all Missouri students in grades 3-8 and 11. The MAP is administered at each of these grade levels within the content areas of Communication Arts and Mathematics. Primary content areas measured in Communication Arts include speaking/writing Standard English, reading fiction/poetry/drama, reading nonfiction, writing formally/informally, and combined reading. Primary content areas measured in math include number and operations, algebraic relationships, geometric and spatial, measurement, and data and probability. Mean grade reliability coefficients (Cronbach's Alpha) for the Communication Arts test across grades was 0.92 and for math was 0.92 (Missouri Department of Elementary and Secondary Education, 2009). The MAP scores were converted to levels in the database to represent the performance descriptors that are common across all grade levels, and these were used in the analysis. Levels included below basic, basic, proficient, advanced.

History of Suspension. History of suspension refers to student behavioral performance in school as measured by number and severity of incidents of challenging behavior that resulted in suspension, either in-school or out of school. Although a standardized measure of behavior would have been ideal, suspension data was the only behavior data available from the school district database. The variable represents the presence of a consequence for a problem behavior in school and, using the type of suspension (in or out), the variable approximates a measure of severity of a problem behavior broadly defined. The history of suspension measure was created by assigning a value of 0 for students who were not suspended during the 2008-2009 school year, a value of 1 for students who received in-school suspension, and a value of 2 for out-of-school suspension. The values were then multiplied by the number of days of the suspension. For example, a student who received in-school suspension for a total of 2 days was assigned a history of suspension score of 2 (1 x 2 = 2). This calculation provided greater weight to out-of school suspension as it is typically assigned for more significant behaviors, such as fighting or threats to the safety of others.

Table 2 provides descriptive statistics for each of the measures for the ED and LD samples. In addition, mean differences (T-tests and Chi-Square) were calculated for each of the measures. Significant differences between the sample of students with ED and LD were only evident on the history of suspension measure.

Analysis Procedures

Structural equation modeling (SEM) was used to assess the interrelationships among the latent constructs (language, academic performance, cognitive ability, and history of suspension), and the influence of disability on each of the four constructs. This technique allows for the simultaneous examination of latent construct relationships and identification of the magnitude of effect each construct has on the other based on the hypothesized relationships (Kline, 2005). For example, SEM can analyze different constructs independently, such as cognitive ability and language skill, as well as identify how the constructs are related to each other. This allows for more complex examination of data and relationships, not just analysis of the observed measures as is typical in many statistical procedures. SEM models combine measurement (i.e. the combination of observed measures that represent a construct) and path models (the relationships between and among constructs) to define direct and indirect effects of observed and latent variables. The measurement model in this study was based on the measures used and the constructs they are designed to measure. For example, the CELF-4 measures both receptive and expressive language, which are combined to create a measure of overall language skill. The decision about what variables to include in the model and how they are related are made by the researcher based on a theory supported by literature. In this study, the SEM model is based on the four hypotheses outlined above, which were rooted in the literature for students with ED and LD. The analysis was conducted using AMOS (Arbuckle, 2006), and interrelationship of the constructs and observed variables assessed were analyzed based on the hypotheses of the study.


Means, Standard Deviations, and Correlations of Observed Variables

Descriptive and correlational data for the whole sample are presented in Table 3. Means for the CELF and CogAT are presented as standard scores (M = 100, SD = 15), and for the MAP, the mean is from the 4 levels of performance (l=below basic, 2=basic, 3=proficient, and 4=advanced). The mean value of the dichotomous disability variable (1=ED, 2=LD) represents how the sample is divided. If there were equal numbers of students with LD and ED, this value would be 1.5. The calculated mean of 1.82 indicates that more of the sample is LD. The mean values for the MAP scores indicate that the students in both categories are scoring between below basic and basic. These results suggest that the students with ED and LD are performing very poorly in comparison with their peers. Overall, only 9% of students in the sampled district scored below basic on the MAP-CA and 12% on the MAP-M (http://www.dese.mo.gov/schooldata/ftpdata.html).

Correlations of interest include the significant (p=.033) relationship between disability and the CogAT Quantitative measure, between disability and history of suspension (p<.000), and the significant correlations (all p<.000) between CELF Expressive and CELF Receptive and all academic skill and performance variables, with the non-significant correlations between the language measures and either history of suspension or disability. The correlation matrix results suggest that disability and history of suspension are not related to cognitive ability, academic performance, or language skills.

Structural Equation Model

The SEM results are presented in Figure 2. The Chi-Square test for model fit was [x.sup.2] (21, N=142) = 30.22, p =.088, verifying that the specified model fits the data because the [x.sup.2] value is not significantly different from 0. Model fit was additionally assessed using Hu and Bentley's (1999) joint criteria of Tucker-Lewis Index (TLI) and comparative fit index (CFI) [greater than or equal to] .96, standardized root mean squared residual (SRMR) [less than or equal to] .09, and root mean squared error of approximation (RM-SEA) [less than or equal to] .06. For the specified model in this study, goodness of fit indices results were as follows: TLI =.96, CFI =.98, SRMR =.04, and RMSEA =.05. All fit indices met the joint criteria and, combined with the Chi-Square results, indicate a good fitting model.


The model in Figure 1 presents the standardized path coefficients for observed (rectangle) and latent (ellipses) variables. The coefficients between the latent constructs and their observed predictors (measurement model) were all significant at p <.000. These results, along with the goodness of fit statistics, further confirm the tenability of the model. For the purposes of this study, the path coefficients between cognitive ability, academic performance, language skills, disability, and history of suspension were designed to statistically represent the study hypotheses. As expected, language skills have a significant effect on both academic performance ([beta] =.195) and cognitive ability ([beta] =.733), and cognitive ability had a significant impact on academic performance ([beta] =.509). Language skills did not have a significant effect on history of suspension ([beta] = -.074). Disability did not have a significant effect on language skills ([beta] =.102), but did have a significant impact on history of suspension ([beta] = -.238). The hypothesized relationship standardized path coefficients between disability and the latent constructs were additionally tested for their direct, indirect, and total effects. The results in Table 4 identify that, within this model, disability did not have an effect on cognitive ability, academic performance, or language skills. These results suggest that students with ED and LD perform similarly in regard to the constructs as tested in this study. Stated another way, in this study, there were no significant differences between the two groups of students with regard to cognitive ability, academic performance, or language skills. There were significant differences between the disability groups on the history of suspension measure. In the next section, these results are discussed in relation to the four hypotheses.


This study was designed to examine the interrelationship among language, cognitive ability, academic performance, and history of suspension for students with ED and LD. As noted above, language skills may be a significant predictor of academic and behavioral performance for students with ED and LD (Tomblin et al. 2000), contributing to etiological understanding of both constructs for each disability group. Our results suggest that language deficits were significant predictors of cognitive ability and academic performance, but not for behavioral performance in school. Additionally, the study found no significant differences between students with LD and ED on any of the measures except history of suspension, suggesting that the key distinction between the groups was behavioral performance in school.

Using structural equation modeling, four hypotheses were tested to examine the relationship among the language, history of suspension, academic, and cognitive constructs. Hypothesis one posited that there would be a significant negative correlation between language and history of suspension for both groups of students. The results indicated that, although the relationship was negative (-.074), it was not statistically significant, which did not support our hypothesis. This suggests that, for both groups, language skills were not predictive of history of suspension. This result is contrasted with the results reported by Nelson and colleagues (2006) and does not support a potential causal relationship between language and history of suspension.

Hypothesis two proposed a significant positive relationship between language and cognitive ability and academic performance for both disability groups, meaning a student with greater language skills will have greater cognitive ability and academic performance. The results supported our hypothesis, finding that both cognitive ability (.733) and academic performance (.195) were significantly related to language skills. These results support Nelson, et al.'s (2006) findings that, for students with ED, language skills significantly impact academic performance.

Similarly, hypothesis three posited that cognitive ability would be significantly related to academic performance. This hypothesis was also supported in our study, finding that the greater the cognitive ability, the greater the academic performance (.509). However, referring back to the descriptive statistics, the academic performance for both groups overall was quite low, in line with much of the research for both disability groups (Reid et al., 2004; Fletcher et al., 2003).

The last hypothesis suggested that a significant difference would be evident between the disability groups on all of the measures. This hypothesis was not supported. As noted, the only difference between the ED and LD groups was on the history of suspension measure, indicating that the groups were the same in regard to language skills, cognitive ability, and academic performance. Table 4 documents how small the group differences were, finding total effects of.10, .09, and.01 for language, cognitive ability and academic performance, respectively. The significant difference in history of suspension as measured by the history of suspension measure intimates that, although the two disability groups are similar in language, academic performance, and cognitive ability, students with ED are suspended much more often. However, given the definition of ED versus LD, this differential pattern of behavioral performance is expected. The findings of this study empirically support the disability distinction on behavioral performance -- students with ED are attributed more difficulties with behavior than students with LD.

Overall, the full model provides a picture of similarities and differences between the disability groups on the constructs measured. The results support the findings that students with ED have both language (Benner et al, 2002) and academic performance deficits (Reid et al., 2004), and that language skill is a significant predictor of academic performance for both disability groups (Benner et al., 2009). The model in this study provides additional support for the Nelson and colleagues (2006) contention that public school professionals should provide language assessment and intervention for students with ED. Additionally, the full model suggests that students with ED need the same level of academic support as students with LD, in addition to the social and behavioral interventions currently recommended.


The findings from this study have several limitations. The first limitation was related to generalization of the findings, specifically small sample size, students all from the same district, and the demographic characteristics of the sample. The sample size was limited by the low frequency of language assessments provided to students with ED in the district sampled, thereby reducing generalization. Although statistically accurate based on the model fit results, the study did not include enough students with ED for broad implications therefore the results should be replicated further and viewed as exploratory in nature. Additionally, the students were all sampled from the same Midwest district, which may not be comparable to other districts in other geographic locations. Additionally, the high percentage of African-American and female students with ED was a concern for the generalization of findings. These high percentages indicate that more African-American and female students with ED are provided a language assessment in this district. Whether a similar pattern is evident in the population is unclear. The inclusion criteria that a language assessment was necessary resulted in the skewed ED sample characteristics and limit generalization of the findings.

Second, there were limitations from the use of an extant database. The measures used in this study were not ideal based on the limitations of using an extant database. The history of suspension measure only addressed in and out of school suspension, not more specific behavioral manifestations, including internalizing and externalizing behaviors. The academic performance measure was limited by its scaling. A standardized measure, such as the Woodcock-Johnson, would have been more ideal as it would have allowed a more fine-grained examination of academic skills. While the CELF-4 is an ideal measure of language skills, a more developed model would examine differences along expressive and receptive constructs.


The results of this study support the other findings that language skills do impact academic performance for students with ED and need to be assessed in this population (Nelson, et al., 2006). Second, our results provide support for academic and language-based interventions to be used with students with ED. As noted previously, language skill is a significant predictor for academic performance. The current model provides additional evidence that students with ED require the same level and extent of academic and language-based support as students with LD, in addition to needed behavioral and social interventions. An integrated approach to intervention may have positive effects on the performance of students with ED (Kauffman, 2010).

Future Research

As mentioned, this was an exploratory study that used data from one school district. Replication studies should involve data from numerous districts in a variety of geographical locations to allow for greater generalization of the results. Future studies also should examine more complex models based on more specific constructs (e.g. expressive and receptive language; reading and math performance; internalizing and externalizing behaviors) using standardized measures. This will allow for greater knowledge into how the constructs interact at a more specific (e.g. receptive language) level. Additionally, future studies should focus on how interventions and supports in language, academics, and history of suspension could be integrated into an efficient and successful support system for students with ED.

As noted above, students with ED often have negative short and long term outcomes, necessitating early and effective intervention practices and systems of support. This study highlights a potential area for further study in both etiology and targeted intervention research. If interventions targeting language skills impact academic and behavioral performance, perhaps they can be integrated into students' with ED systems of support. As future research develops, so, too, will the potential for targeted interventions to concomitantly ameliorate problem behaviors and increase academic performance for students with ED.


Arbuckle, J. L. (2006). Amos 7.0 user's guide. Chicago, IL: SPSS.

Beck, I., McKeown, M., & Kucan, L. (2002). Bringing Words to Life: Robust vocabulary instruction. New York: The Guilford Press.

Benner, G. J., Mattison, R. E., Nelson, J. R., & Ralston, N. C. (2009). Types of language disorders in students classified as ED: Prevalence and association with learning disabilities and psychopathology. Education and Treatment of Children, 32(4), 631-653.

Benner, G. J., Nelson, J. R., & Epstein, M. H. (2002). The language skills of children with emotional and behavioral disorders: A review of the literature. Journal of Emotional and Behavioral Disorders, 10(1), 43-59.

Bradley, R., Doolittle, J., & Bartolotta, R. (2008). Building on the data and adding to the discussion: The experiences and outcomes of students with emotional disturbance. Journal of Behavioral Education, 17(1), 4-23.

Catts, H.W. (1993). The relationship between speech-language impairments and reading disabilities. Journal of Speech and Hearing Research, 36(5), 948-958.

Catts, H.W. (1997). The early identification of language-based reading disabilities. Language, Speech, and Hearing Services in Schools, 28(1), 86-89

Catts, H.W., Fey, M.E., Tomblin, J.B., & Zhang, X. (2002). A longitudinal investigation of reading outcomes in children with language impairments. Journal of Speech, Language, and Hearing Research, 45(6), 1142-1157.

Fletcher, J. M., Morris, R. D., & Lyon, G. R. (2003). Classification and definition of learning disabilities: An integrative perspective. In H. L. Swanson, K. R. Harris & S. Graham (Eds.), Handbook of Learning Disabilities (pp. 30-56). New York, NY: Guilford Press.

Gersten, R., Fuchs, L., Williams, J., & Baker, S. (2001). Teaching reading comprehension strategies to students with learning disabilities: A review of research. Review of Educational Research, 71(2), 279.

Gallagher, T. M. (1999). Interrelationship among children's language, behavior, and emotional problems. Topics in Language Disorders, 19, 1-15.

Hodge, J., Riccomini, P. J., Buford, R., & Herbst, M. H. (2006). A review of instructional interventions in mathematics for students with EBD. Behavioral Disorders, 31, 297-311.

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55.

Hummel, L. J., & Prizant, B. M. (1993). A sociometric perspective for understanding social difficulties of school-age children wit language disorders. Language, Speech, and Hearing Services in Schools, 24, 216-224.

Kaiser, A. P., Hancock, T. B., Cai, X., Foster, E. M., & Hester, P. P. (2000). parent-reported behavioral problems and language delays in boys and girls enrolled in Head Start classrooms. Behavioral Disorders, 26(1), 16-41.

Kauffman, J. M. (2010). Commentary: Current status of the field and future directions. Behavioral Disorders, 35(2), 180-184.

Kauffman, J. M. & Brigham, F. J. (2009). Working with troubled children. Verona, WI: Full Court.

Kauffman, J. M. & Landrum, T. J. (2008). Characteristics of emotional and behavioral disorders of children and youth (9th ed.). Upper Saddles River, NJ: Prentice Hall.

Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford Press.

Kostewicz, D. E., & Kubina, R. M. (2008). The national reading panel guidepost: A review of reading outcome measures for students with E/BD. Behavioral Disorders, 33(1), 62-74.

Kuder, S. (2008). Teaching Students with Language and Communication Disabiltities (Third ed.). Boston: Pearson Education, Inc.

Lane, K. L. & Menzies, H. M. (2010). Reading and writing interventions for students with and at risk for emotional and behavioral disorders: An introduction. Behavioral Disorders, 35(2), 82-85.

Lohman, D. F., & Hagen, E. P. (2001). Cognitive Abilities Test (CogAT). Rolling Meadows, IL: Riverside Publishing.

Lyon, G., Fletcher, J., Shaywitz, S., Shaywitz, B., Torgesen, J., Wood, F., et al. (2001). Rethinking learning disabilities. Rethinking special education for a new century, 259-287.

Maag, J. W., & Katsiyannis, A. (1998). Challenges facing successful transition for youths with E/BD. Behavioral Disorders, 23(4), 209-221.

Mann, V. A. (2003). Language processes: Keys to reading disability. In H. L. Swanson, K. R. Harris & S. Graham (Eds.), Handbook of Learning Disabilities (pp. 213-228). New York, NY: Guilford Press.

Mattison, R. E. (2008). Characteristics of reading disability types in middles school students classified as ED. Behavioral Disorders, 34(1), 27-41.

McArthur, G., Hogben, J., Edwards, V., Heath, S., & Mengler, E. (2000). On the "specifics" of specific reading disability and specific language impairment. Journal of Child Psychology and Psychiatry, 41(7), 869-874.

McCabe, P. C, & Meller, P. J. (2004). The relationship between language and social comptence: How language impairment affects social growth. Psychology in the Schools, 41(3), 313-321.

McKeown, M, & Beck, I. (2004). Direct and rich vocabulary instruction. In J. F. Baumann & E. J. Kame'enui (Eds.), Vocabulary Instruction: Research to practice (pp. 13-27). New York: The Guildord Press.

Missouri Department of Elementary and Secondary Education. (2009). Missouri Assessment Program: Grade-level assessments technical report. Jefferson City, MO: Author.

Nelson, J. R., Benner, G. J., & Cheney, D. (2005). An investigation of the language skills of students with emotoinal disturbance served in public school settings. The Journal of Special Education, 39(2), 97-105.

Nelson, J. R., Benner, G. J., Neill, S., & Stage, S. A. (2006). Interrelationship among language skills, externalizing behavior, and academic fluency and their impact on the academic skills of students with ED. Journal of Emotional and Behavioral Disorders, 14(4), 209-216.

Owens, R. E. (2001). Language development: An introduction (5th ed.). Needham Heights, MA: Simon and Schuster.

Pullen, P. C, Lane, H. B., Ashworth, K. E., & Lovelace, S. P. (2011). Learning disabilities. In J. M. Kauffman & D. P. Hallahan (Eds.), Handbook of Special Education (pp.187-197). New York, NY: Routledge.

Rogers-Adkinson, D. L. (2003). Language processing in children with emotional disorders. Behavioral Disorders, 29(1), 43-47.

Ruhl, K. L., Hughes, C. A., & Camarata, S. M. (1992). Analysis of the expressive and receptive language characteristics of emotionally handicapped students served in public school settings. Journal of Childhood Communication Disorders, 14, 165-176.

Scarborough, H. (2001). Connecting early language and literacy to later reading (dis) abilities: Evidence, theory, and practice. In D. K. D. Susan B. Neuman (Ed.), Handbook of early literacy research, Volume 1 (Vol. 1, pp. 97-110).

Semel, E., Wiig, E. H., & Secord, W. A. (2003). Clinical Evaluation of Language Fundamentals-Fourth Edition (CELF-4). San Antonio, TX: Pearson.

Tomblin, J., Zhang, X., & Buckwalter, P. (2000). The Association of Reading Disability, Behavioral Disorders, and Language Impairment among Second-grade Children. Journal of Child Psychology and Psychiatry, 41(4), 473-482.

Trout, A. L., Nordness, P. D., Pierce, C. D., & Epstein, M. H. (2003). Research on the academic status of students with emotional and behavioral disorders: A review of the literature from 1961-2000. Journal of Emotional and Behavioral Disorders, 11,198-210.

Wagner, M., Kutash, K., Duchnowski, A. J., Epstein, M. H., & Sumi, W. C. (2005). The children and youth we serve: A national picture of the characteristics of students with emotional disturbances receiving special education. Journal of Emotional and Behavioral Disorders, 13(2), 79-96.

Wagner, R. K., Francis, D. J., & Morris, R. D. (2005). Identifying English Learners with learning disabilities: Key challenges and possible approaches. Learning Disabilities Research and Practice, 20(1), 6-15.

Woodcock, R. W., McGrew, K. S., & Mather, S. (2001). Woodcock Johnson III test of Achievement. Itasca, IL: Riverside.

Correspondence to Lisa Goran, 303 Townsend Hall, University of Missouri, Columbia, MO 65211; e-mail: lgg4fe[congruent to]mail.mizzou.edu.

Lisa G. Goran and Nicholas A. Gage University of Missouri
Table 1 Student Characteristics

                      ED      LD

                    n   %   n     %

Gender              0   0   0     0

 Male              15  60  69    59

 Female            10  40  48    41


 White              8  32  54  46.1

 African-American  15  60  47  40.2

 Hispanic           0   0  11  09.4

 Asian              0   0   3   2.6

 Multi-Racial       2   8   2   1.7


 Not Free/Reduced   5  20  35  29.9

 Reduced            1   4  12  10.3

 Free              19  76  70  59.8

Table 2 Construct Measures

                             ED(n=25)          LD(n=118)

                            M      SD         M        SD    p

 CELF Expressive           80.05  16.12      83.32  14.83   0.32

 CELF Receptive            84.74  11.50      87.61  13.70   0.33

Cognitive Ability

 CogAT Nonverbal           88.71  14.52      93.22  12.13   0.10

 CogAT Quant               84.12  11.14      85.13  11.83   0.70

 CogAT Verbal              82.76   9.10      86.44  10.83   0.12

Academic Performance

 MAP Comm Arts              1.62   0.46       1.54   0.50   0.77

 MAP Math                   1.62   0.54       1.66   0.52   0.16

History of Suspension

 Suspension                 1.28   1.14       0.50   1.20  0.00*

Note: *p<.00

Table 3

Mean, Standard Deviations, and Correlations of Observed

Measure    1      2      3      4      5      6      7      8      9

1. DIS        -

2. CA     -0.06      -

3. MA     0.03   0.52      -

4. REC     0.08   0.29   0.39      -

5. EX     0.08   0.34   0.33   0.53      -

6. VERB   0.13   0.23   0.34   0.53   0.41      -

7. NONV   0.13   0.36   0.49   0.43   0.29   0.49      -

8. QUAN   0.03   0.32   0.48   0.57   0.37   0.63   0.68      -

9. SUSP  -0.25  -0.09  -0.06  -0.10  -0.07  -0.01   0.02  -0.00     -

M         1.82   1.55   1.65  87.11  82.74  85.79  92.43  84.95  0.63

SD        0.38   0.49   0.52  13.35  15.06  10.61  12.65  11.68  1.22

Note: Correlations [less than or equal to].23, p<.01;
[less than or equal to].29, p<.000; Dis - Disability, CA=MAP
Communication Arts, MA=MAP Mathematics, REC=Receptive Language,
EX=Expressive Language, VERB= CogAT Verbal, NONV=CogAT Nonverbal,
QUAN=CogAT Quantitative, and SUSP=History of Suspension.

Table 4 Direct, Indirect and Total Effects of Disability on Constructs

                      Direct  Indirect  Total

Language                 .10       .00    .10
Cognitive Ability        .02       .08    .09
Academic Performance    -.06       .07    .01
Gale Copyright:
Copyright 2011 Gale, Cengage Learning. All rights reserved.