An investigation of the language skills of students with emotional disturbance served in public school settings.

The purpose of this cross-sectional study, conducted with a random sample of 166 students with emotional disturbance (ED), was to establish, with attention to age and gender differences, the extent to which students with ED served in public school settings experience language skill deficits. This study also examined particular types of problem behaviors related to language skills. Students with ED showed moderate to large language deficits, which appeared to be more pronounced in the expressive language domain. The language deficits of a majority of the students were clinically significant. These language deficits appeared to be relatively stable across age and gender. Furthermore, externalizing behaviors were related to receptive and expressive language skills, whereas internalizing ones were not.

**********

Successful language acquisition is critical for achieving positive social adjustment (e.g., Benner, Nelson, & Epstein, 2002; Kaiser, Hancock, Cai, Foster, & Hester, 2000). Competent language skills are also prerequisites for successful academic learning in all areas, including reading, math, written expression, and content areas such as science and social studies (Baker & Cantwell, 1987). It is hypothesized that both language problems and emotional disturbance (ED) may emerge from the same etiological or environmental factors, such as parent-child interactions (Benner, Nelson, & Epstein, 2002; Kaiser & Hester, 1997; Kaiser et al., 2000). Furthermore, delays in language may worsen ED or vice versa. Although a plethora of research has been conducted with children and adolescents with ED served in clinical settings, only three research studies have been conducted with students with ED served in public school settings (Benner, Nelson, & Epstein, 2002). Thus, this study investigated the language skills of students with ED served in public school settings.

Previous research conducted on children and adolescents served in clinical settings has established the co-occurrence between language deficits and a wide range of problem behaviors (e.g., juvenile delinquency, antisocial behaviors, attention-deficit disorders; Benner, Nelson, & Epstein, 2002). Briefly, children with language deficits are 10 times more likely to exhibit antisocial behaviors than those in the general population (Donahue, Cole, & Hartas, 1994; Warr-Leeper, Wright, & Mack, 1994). The psychopathological problems of children with language deficits tend also to increase as the children age (Baker & Cantwell, 1985). Furthermore, children with pure language deficits (i.e., receptive, expressive, pragmatic) appear to be at higher risk for antisocial behaviors than those with speech disorders or speech-and-language disorders (Prizant et al., 1990; Rutter & Mawhood, 1991; Toppelberg & Shapiro, 2000). The likelihood of children exhibiting antisocial behaviors tends to be higher for those with pure receptive language deficits (Baker & Cantwell, 1985; Cohen, Davine, Horodezsky, Lipsett, & Isaacson, 1993).

A recent review synthesized the literature on the language skills of children formally identified as having ED (i.e., children who met special education or psychological criteria for ED; Benner, Nelson, et al., 2002). This review revealed three primary gaps in the literature on language skills of children and adolescents with ED. First, only three studies have been conducted to date on the language skills of students with ED served in public school settings (i.e., Camarata, Hughes, & Ruhl, 1988; McDonough, 1989; Miniutti, 1991). Furthermore, researchers used small convenience samples (n = 87) composed only of elementary-age children (8-10 years old). Second, researchers generally used a range of cutoff criteria (e.g., 1 standard deviation discrepancy from the mean) to establish the co-occurrence of language deficits and a range of problem behaviors including ED. Differences in the cutoff criteria used by researchers to establish a language deficit resulted in a high degree of variability in estimates of the percentages (i.e., 25%-95%) of children with ED who had language deficits. Finally, it appears that researchers have not investigated the particular types of problem behaviors (i.e., externalizing, internalizing) exhibited by students with ED that are related to language deficits. The failure of researchers to investigate the types of problem behaviors that are related to language deficits is most likely a function of the common use of psychiatric classification systems in clinical settings. Such categorical classification systems make it difficult to identify types of problem behaviors related to language deficits.

This cross-sectional study of the receptive and expressive language skills of students with ED served in public school settings addresses these three identified gaps. First, we sampled randomly from students with ED served in public school settings across Grades K-12. Sampling randomly across the grades provides a more complete picture of the language skills of students with ED. Second, we used mean standardized scores from an individually administered measure of receptive and expressive language skills. The use of mean standardized scores rather than cutoff scores provides a more accurate estimate of the receptive and expressive language skills of students with ED. Finally, we used a dimensional classification to examine the particular types of problem behaviors that are related to receptive and expressive language skills. In contrast to psychiatric classification systems, dimensional ones are designed to measure the degree to which students exhibit particular behavioral syndromes or patterns on a continuum. Dimensional classification systems assume that a number of behavioral traits exist and that all children possess these traits to some degree (Mash & Wolfe, 1999). In the present study, we used the Teacher Report Form (TRF; Achenbach, 1991) to determine the particular types of problem behaviors that are related to receptive and expressive language skills. The TRF is one of the rating scales most commonly used by schools and in research conducted with students with ED (Mattison, 2001).

Based on previous research conducted with children and adolescents with ED served in clinical and school settings (cf. Baker & Cantwell, 1985; Benner, Nelson, et al., 2002; Donahue et al., 1994; Gallagher, 1999; Prizant et al., 1990; Rutter & Mawhood, 1991; Toppelberg & Shapiro, 2000, for reviews of the literature), we expected the following four outcomes. First, we expected that our sample of students with ED would experience moderate to large receptive and expressive language deficits relative to the norm. Second, we expected the students' language skill deficits to be relatively stable across the years because schools typically do not attempt to remediate these deficits. Third, given the pervasive nature of ED, we expected that boys and girls would experience similar language skill deficits. Finally, although it appears that researchers have not studied the particular types of problem behaviors related to language skills, we expected that externalizing ones would be more strongly related to language skills than internalizing ones. We expected this because previous research has revealed that externalizing behaviors are related to academic achievement but internalizing ones are not (e.g., Nelson, Benner, Lane, & Smith, 2004).