Printer Friendly

Assessment considerations in the evaluation of second-language learners: a case study.

To provide optimal services to the large population of children who have learned little or no English by the time they enter school, educators must differentiate children with temporary limited English proficiency from children with language disorders (those who have difficulty leaming any language). The needs of these two groups are different. Children with limited English proficiency should eventually attain proficiency after receiving adequate bilingual education or English as a Second Language (ESL) instruction. Those with language disorders will need long-term special education services. Unfortunately, many students with limited English proficiency have been misdiagnosed as having language disorders--or remain undiagnosed; and children who are second-language learners are reportedly overrepresented in certain categories of special education (see Kretschmer, 1991, for a review of the literature).

The differentiation of children with language disorders from those with limited English proficiency presents perplexing problems to even the most skilled members of a child study team. The problem in diagnosis occurs primarily because norms are not readily available for all the languages of diverse populations and cultures, and because standardized tests in English are of little value in assessing many second-language learners who are not yet proficient in English. The criterion of limited communicative competence in both languages is often used for determining the presence of a language disorder (American SpeechLanguage-Heating Association, [ASHA], 1985), but arrested language development or language loss may negatively affect the child's performance in the native language so that using this criterion exclusively may result in false positive identifications.


Standardized Tests in English

Standardized language tests in English have generally been norreed on monolingual English speakers, and if children from non-Englishspeaking or limited-English-proficiency (LEP) homes are included in the normal sample, they are usually underrepresented (see Mattes & Omark, 1984, pp. 51-70). Children from homes in which English is not spoken or only limited English is spoken cannot be expected to perform as well as children who come from homes in which standard English is spoken. Because English is acquired as a second language and there has not been sufficient opportunity to hear and use it (in comparison with an English-monolingual child), a normal second4anguage learner is expected to score lower on standardized tests for some time. Age-appropriate cognitive/academic language proficiency (CALP) may take up to 7 years to develop (Cummins, 1980).

Standardized Tests in Other Languages

Federal mandates (Public Law 94-142 and Title VII of P.L. 95-561) require that school assessments be conducted in the client's primary language. Consequently, it is recommended that non-English-speaking children orthose with LEP be tested in their native languages to determine if their native4anguage development is delayed (ASHA, 1983, 1985; Langdon, 1983; Mattes & Omark, 1984). Although standardized tests in Spanish are available, some have been normed on monolingual English-speaking children and then translated into Spanish. Even when tests have been norreed on a Spanish-speaking population, often the normed sample consists of monolingual Spanish speakers rather than speakers who have learned Spanish as a first language and are subsequently exposed to English.

Cultural and Linguistic Differences

When tests have been normed on "bilingual" speakers (i.e., other children who are learning English as a second language), one cannot assume that the tests have been normed on Spanish speakers who come from a particular client's culrural or- dialectal background. Cultural and linguistic differences need to be considered. In the Hispanic population, for example, cultural and linguistic differences exist among Puerto Ricans, Cubans, Mexicans, and Hispanic groups from Central and South America. Further, the language spoken in the home may not be the standard Spanish from that country that is used on the test. Therefore, determining whether a child has a language disorder by comparing his or her language performance to that of peers is problematic. To meet this need, school systems have been urged to develop local norms or adapt test instruments (Kayser, 1989; Mattes & Omark, 1984). Because of the heterogeneity of the Spanish-speaking population, however, even within a school district, this may be an impossible task.

Informal Measures

To obtain a more complete picture of the child's communicative competencies, the American Speech-Language-Hearing Association ( 1983, 1985) has recommended the use of informal methods. These informal methods include language-sampling analyses and informal probes or questionnaires. Such methods are useful, but do not completely solve the problem. To get a representative sample, educators should collect language samples in both languages when the child is communicating in a naturalistic context with a native speaker of each language (Kayser, 1989; Mattes & Omark, 1984). These samples should include information about content (complexity of thought encoded) and use (interactional skills), as well as form (vocabulary, syntax, and grammatical morphemes) (Lahey, 1988). Interpreters and translators are recommended when the staff speech-language pathologists are not proficient speakers of the child's language (ASHA, 1985). However, the child's language must still be compared to some standard that subsumes both cultural and linguistic differences.

Although language samples are useful for analyzing children' s communication behaviors, vocabulary and morphosyntactic errors cannot be used to determine a language disorder without doing a norm-referenced evaluation. Staff members must still make a comparison to other children from the same cultural and linguistic background to distinguish a language disorder from normal language difficulties encountered in the learning of a second language. The child from a Spanish-speaking home may seem to have limiited vocabulary and may be making syntactic errors for one of several reasons:

* He or she may not have had enough experience with the second language (English).

* He or she may be code switching.

* He or she may be using the dialect or combination of languages heard in the home (Zentella, 1990).

* He or she may have a language-learning problem.

Therefore, the evaluation team must make a differential diagnosis based on aspects of the behavior other than the form of language. However, many tests used to evaluate perceptual and cognitive deficiencies are language dependent and may be culturally biased (Chamberlain & Medinos-Landurand, 1991).

Consideration of Arrested Development or Language Loss of the First Language

The ability to distinguish between communication differences and communication disorders is further complicated because arrested language development or language loss can occur in the primary language when the child focuses attention on learning a second language (L2). It is not unusual for some children with LEP to be temporarily delayed in both languages (Hamayan & Damico, 1991; Schiff-Myers, 1992). This lack of proficiency occurs primarily within the group of children who learn L2 outside the home in a "school-like" setting during the preschool or early childhood years, rather than in a natural communicative context. Kessler (1984) reported that if children are put on the defensive by being required to perform too early in the second language, the process can be interrupted or slowed down. Some children may go through a silent period as they attempt to process L2, and it may appear that these children have problems. Kessler has contended that children acquiring a second language during the school years have an advantage because they have developed some metalinguistic skills. They know what language is and can focus on learning language form, whereas the preschooIer focuses on meaning.

Cummins (1984) proposed threshold and developmental interdependence hypotheses to explain this bilingual language delay. Cummins postulated that the level of competence in L2 is partly a function of the competence developed in the primary language (L1) at the staff of exposure to L2. Therefore, a threshold of competency in L1 must be obtained before exposure to L2 if problems are to be avoided. These hypotheses suggest the following:

1. When a child is introduced to L2 before. the

full development of his or her native language,

the development of the native language may

be arrested or ngress while the child attends

to the development of L2.

2. The development of competence in L2 is a

function of the level of competence previously

developed in the mother tongue.

Therefore, if L1 has not been allowed to fully develop, it in turn may affect the extent of L2 development and can result in "semilingualism" (Miller, 1984). The child will not become literate or proficient in either language. The regression or language loss of L1 (e.g., Spanish) seems to occur most often when L1 culture and language is not viewed by the child to be as prestigious as L2, the dominant language of the community (Lambert, 1977, 1981).

Although cases of language loss in the Hispanic population have been reported elsewhere (Merino, 1983), the possibility is not often considered even among speecManguage pathologists who have received the Bilingual Certificate of Assessment Competence (Langdon, 1989). In a study that reviewed practices of bilingual speech-language pathologists who held this certificate, Langdon found that the possibility of language loss was rarely considered by these welltrained professionals, In addition, few reports contained information on developmental issues, family and sociocultural factors, and classroom performance, but considerable importance was given to test scores.

The purpose of this article is to present a case study of a child, Sandy, who was classified as having a language-learning disorder by a child study team. These deficiencies proved to be temporary. A major problem that child study teams encounter is the differentiation of students who have temporary difficulties nounally encountered in developing school-language proficiency (CALP) from persistent language-learning disorders resulting from other problems that interfere with language learning (e.g., cognitive, perceptual, emotional deficits) (Hamayan & Damico, 1991, p. viii). Therefore, an examination of the procedures used--and not used--when the child was classified should be revealing. The major criterion that was used to classify this child at the end of her kindergarten year was limited language ability in both Spanish and English. The child was tested in English, but not in Spanish, because her English was reportedly better than her Spanish. However, the team did not consider the possibility of arrested language development or language loss in Spanish to explain the limitations in her first language.

This child fits the description of a child whose deficiencies in Spanish can be explained by Cummins' threshold and developmental interdependence hypotheses because she was exposed to English for the first time in nursery school before her first language was fully developed. An analysis of a comprehensive case history and language samples might have made team members suspicious that this child's problem was related to L2 learning, and a different kind of service program could have been recommended. However, no analysis of language samples in either language was reported.

It was the analysis of language samples approximately 6 months later that led the investigators to question the classification. A case history was collected by the child study team, but was used only to determine that English was the dominant or preferred language of the child. Key factors that might have signaled arrested development or language loss would be (a) normal developmental milestones, (b) responses that identified the child as one who acquired English as a second language in a school-like setting during the preschool years before competency in Spanish had been achieved, (c) preference for speaking English when spoken to in Spanish (Conklin & Lourie, 1983), and (d) the caretakers' appraisal of the problem.

The child was evaluated by the authors when she was in the 1st grade. Two graduate students in speech-language pathology did the assessment under supervision. One was a native Spanish speaker who evaluated the child in Spanish; the other graduate student evaluated the child in English. The methods used were those recommended by ASHA (1983, 1985) and included both informal and formal methods. Informal methods included language sampling and analysis and informal probes and questionnaires (the case history). Formal procedures included both standardized English tests and Spanish tests. At that time, the language samples did not appear to be typical of a child with language disorders. They were qualitatively different, and content and use were far superior to form. The results of subsequent observations and testing (when the child was in the 3rd grade) yielded essentially normal results, although there were still some minor errors of form. Assessment procedures are described in the following case study.



An Hispanic girl, whom we will call Sandy, was 6 years 7 months and in the 1st grade when initially seen by the authors. Both parents had 1 year of study beyond high school before emigrating to this country from Uruguay. Sandy was born in the United States and is the oldest of two children. Although the mother could speak some English, all communication between the parents and to the children was in Spanish. Spanish was the language used in the community and in the church. Sandy said her first words in Spanish when she was approximately 1 year old and began combining words between the ages of 18 and 24 months. She was reportedly a quiet child, but all developmental milestones were normal, and her parents considered her to be normal in every respect.

At the age of 3 years and 6 months, Sandy was sent to nursery school 3 days a week so that she would learn English. The nursery school found Sandy to be "nonverbal" and told the mother they thought Sandy might have a language problem. During her kindergarten year, Sandy received ESL instruction in a group of three students for 30 min each school day. In March of that year, when Sandy was age 5 years and 7 months, she exceeded the exit score on Level 1 of the Language Assessment Battery (LAB) (1982), a test developed by the New York City Public School System and required by the state of New Jersey to assess the language proficiency (reading, writing, listening, and speaking) of all nonnative speakers of English whose English-language proficiency may be limited. Sandy passed the minimum English language proficiency level at that time. However, the standards for minimum proficiency in English changed 2 years later, and Sandy's score would not now be considered to be a passing one. Therefore, if today's standards were used, she would have been retained in the ESL program and would not have been referred to the child study team.

Classification of Communication Disability

According to all reports, Sandy's English was better than her Spanish, and because she was not fully proficient in English, she was referred to the child study team for further evaluation. Table 1 shows the results of this evaluation, which occurred when Sandy was between the ages of 5 years 7 months, and 5 years 9 months. The differential between Verbal and Performance IQ, as indicated by the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) (Wechsler, 1967), was significant; there were 2 standard deviations difference between the two. The only receptive measures on which Sandy scored below 1 standard deviation from the mean was the visual association portion of the Illinois Test of Psycholinguistic Abilities (ITPA) (Kirk, McCarthy, & Kirk, 1968) and the Picture Vocabulary subtest of the Test of Language Development (TOLD) (Newcomer & Hammill, 1977). All other scores in subtests that were below normal were expressive measures. The mean scaled score for all subtests on the ITPA was 30.9, considered to be within normal limits (within 1 standard deviation), while the language quotient for the TOLD was 76 (considered to be below average (more than 1 standard deviation below the mean). The Peabody Individual Achievement Test (Dunn & Markwardt, 1970) was administered for mathematics, reading recognition, reading comprehension, spelling, and general information. Only reading comprehension was reportedly below grade norms. Sandy was classified as having a communication disability, with recommendations for promotion to the 1st grade, but with resource room tutoring in math, reading and language arts, and speech-language therapy.

Results of Speech-Language Assessment--1st Grade

All testing and informal observations occurred during the spring of the 1st grade when Sandy was between age 6 years 7 months, and 6 years 8 months. Sandy was receiving speech-language therapy from one of the investigators and was in the lowest reading group in her class. She was attending the resource room for math, reading, and language arts, and the teachers reported that Sandy was shy and did not speak often in class.

Case History Report

A comprehensive case history form was developed that contained questions concerning family background and the language(s) used in the home and community (see Mattes & Oreark, 1984, pp. 111-112), as well as developmental history, age, and manner of exposure to English. Questions concerning contacts with the homeland, attitudes toward both languages, and the parents' perception of the problem were also included. All questions were asked orally by the Spanish-speaking investigator so that the mother's communication could be evaluated and her answers fully explored.

The parents were the primary caretakers for this child. The mother stated that although comprehension in both languages was excellent, Sandy chose to speak in English. In response to questions concerning which language the family preferred Sandy to speak, the mother responded that although she wanted Sandy to be proficient in English so that she would do well in school, she also wanted her to be proficient in Spanish so that she would maintain her language and her culture and would be able to communicate with relatives when periodic visits were made to Uruguay. The family intended to stay in the United States permanently. Subsequently, at the end of the 2nd grade, Sandy's mother placed her in an after-school Spanish program.

When asked about the nature of Sandy's speech and language problem, the mother responded that Sandy's problem was that she confused the two languages. She was concerned that the school had considered her to be "handicapped." In response to a question concerning family members who have speech or language problems, the mother reported that a cousin who (like Sandy) had attended an English-speaking nursery school, had received speech-language therapy for a similar problem, but now is fine. This child's older sister, who emigrated to this country when she was 8 years old, presumably after she had mastered Spanish, had no difficulty with either language.

Test Results

The Bilingual Syntax Measure (Burt, Dulay, & Hernandez-Chavez, 1976) was administered in both languages to determine linguistic dominance. The results of the Bilingual Syntax Measure showed English to be dominant, with a Level 2 score in Spanish and a Level 4 score in English. A Level 2 score in Spanish indicated that Sandy did not communicate her ideas fully in Spanish. A Level 4 score in English for a child who is 7 years old indicated that English was not as proficient as native English speakers of that age. The test authors recommended additional ESL instruction for children at this level, focusing on conversational skills, reading, writing, and vocabulary. These results clearly indicated syntactic dominance in English and the need for additional language training.

The following standardized tests were administered: Peabody Picture Vocabulary Test-Revised (PPVT) (Dunn & Dunn, 1981) and the Spanish adaptation Test De Vocabulario En Imagenes Peabody (TVIP) (Dunn, Padilia, Lugo, & Dunn, 1986); Test for Auditory Comprehension of Language (TACL) in Spanish and in English (Carrow, 1973); and Northwestern Syntax Screening Test (NS ST) (Lee, 1971 ) and its Spanish counterpart Screening Test of Spanish Grammar (STSG) (Toronto, 1973). The Test for Auditory Comprehension of Language-Revised (TACL-R) (Carrow-Woolfolk, 1985) was also administered to compare the results of the more recent version with those of the earlier test.

Test results indicated that Sandy's comprehension of both English and Spanish syntax was normal, whereas her expressive syntax in both languages was not (based on the results of the NSST and the STSG). If the 10th percentlie and 1.5 standard deviation below the mean are used as the criteria for determining a deficit, then receptive vocabulary in Spanish and English would also be considered to be within normal range. If 1 standard deviation is used as the criterion, then the results of the TVIP indicate a receptive vocabulary deficit in Spanish (see Table 2). Sandy did better on the Spanish translation of the TACL than on the English test, and she did better on the receptive portion of the STSG (Spanish) than she did on the receptive portion of the NSST (English). All these tests are measures of comprehension of syntax. However, her receptive vocabulary scores were better in English than in Spanish, based on the results of the PPVT and the TVIP.

Sandy's percentile rank was 20 percentage points higher on the TACL-R than it was on the older TACL, but both were clearly within normal limits.

Language Sample Analyses

Four half-hour language samples (two in Spanish and two in English) were recorded as the child interacted in a "naturalistic" play environment. Narrative samples were also collected. The Spanish samples were collected with a native Spanish speaker who had learned English when she had entered the public schools at the age of 6, and the English samples were collected with a native English speaker who was the child's speech language clinician. Both investigators were graduate students and were supervised by the first author, an experienced researcher of children's language. An investigator transcribed the language samples and the related contextual information independently, and two other listeners verified them. An utterance was considered to be unintelligible and therefore was not considered for analysis if two of the three listeners did not agree on the transcription. Only 3 utterances were eliminated from the English sample and 19 from the Spanish sample. A total of 330 intelligible utterances were analyzed from the English sample and 192 utterances from the Spanish sample.

Utterances were categorized according to the meanings encoded (content or semantics), the forms used (syntax and morphology), the functions for which they were used, and discourse characteristics (use or pragmatics) (Lahey, 1988). For utterances to be categorized, agreement of the three investigators was required. A particular content-form category was considered to be productive if a minimum of four different spontaneous utterances occurred. A level of development for the three aspects of language (content, form, and use) was established for both languages, using guidelines from "A Content/Form/Use Goal Plan for Language Development Goals" (Lahey, 1988).

Analysis of Content-Form Interactions itt English. Analysis of content-form interactions in English (Lahey, 1988) showed Sandy to be talking about a variety of subjects and coding complex sentence structures with two clauses. She was productively coding in English all semantic categories with the appropriate forms through Phase 8, with the exception of adversative C'I would like to color but I don't have any crayons"). This category was coded only once in the samples. In addition, Sandy did not ask many questions at any phase of development; only two "what" questions were asked for existence in the English sample (e.g., "What is that?"); more complex questions in English occurred in the Spanish sample in utterances in which Sandy code switched (e.g., "What's esto? How you put on esto?").

Analysis of Content-Form Interactions in Spanish. In Spanish, Sandy was using predominately simple one-clausal structures; she was productively coding all content-form interactions through Phase 3 with sporadic coding of semantic categories in more advanced phases. Many of the forms that carry meaning in English are not coded in Spanish. To compensate, credit was given for coding the form even if it was not required in Spanish. For example, the agent of the action need not be included in Spanish (e.g., "Va asi--lt goes like this"). In cases such as these, Sandy was given credit for coding three constituents. Content-form interactions, such as the use of the verb "will" to code intent is often not needed in Spanish (e.g., "Venemos tarde okay--We will be late okay?"), and negatives are coded with "no" rather than forms such as "can't" or "didn't." These 'forms are necessary for categorization in the later phases of development of negation in English. These content-form interactions could not be assessed, and a comparison with English content-form interactions was impossible. Nevertheless, Sandy was clearly not coding two clausal sentences in Spanish, and she was coding them in English. Code switching occurred often in the Spanish sample, especially when Sandy tried to express complex ideas (e.g., "Let me see what she gusta para poneF') and sometimes as vocabulary items ("I go la escuela Sunday/hoy is Friday"). Code switching never occurred in the English sample.

Analysis of Form. To further define a level of grammatical complexity, we determined a mean length of utterance (MLU) in each language (Brown, 1973). Sandy's MLU in English was 6.03; in Spanish, 3.1. Utterances that involved code switching between the two languages were not included in the calculation of MLU. These differences in MLU indicated that Sandy was coding complex ideas in complex utterances in English, but not in Spanish. Some code switching in the Spanish sample may have occurred because Sandy was aware that her listener was bilingual and viewed code switching to be acceptable communicative behavior. However, her more complex sentence structures were not coded in Spanish without using English vocabulary and syntax to complete them (e.g., "Let me see what she gusta para poner" ; "Dos ninos went to the haunted house for see bats and ghosts"). Therefore, Sandy's expressive English vocabulary and syntax appeared to be superior to her Spanish.

An analysis of form in the English sample revealed numerous errors. In a total of 259 sentences (those utterances containing subject-verb-object constructions), 88 (34%) contained a syntactic or morphological error. These errors included incorrect subject-verb agreement ("He have a bed"), word finding or word substitution errors, misuse of the irregular past tense, overuse of the preposition for, errors in use of the preposition off, and sporadic omission of the following forms: plural s; third person singular s; past tense ed; the verb will; auxiliary verbs; the copula; present progressive ing; and syntactic connectors in complex sentences (what, when, where). The missing forms were emerging, but had not yet reached the criterion of 90% in required contexts (Brown, 1973). Other errors of form occurred as isolated instances; therefore, there was no identifiable pattern of error.

Evidence of Linguistic Interference. Within the sample were errors that are not typically found in the language of monolingual language-disordered children, such as omission of ing with the inclusion of the auxiliary verb (e.g., "We're all eat"; "I was so cold yesterday and I'm feel sick"). A number of errors seemed to result from the influence of Spanish syntax (e.g., the overuse of articles--"My brother play with her the Friday" from the Spanish--mi hermano juega con ella el viernes; "He work a lot in the night" from the Spanish en la noche instead of at night; the use of for in "We need to call the doctor first for see we're open" from the Spanish use of para--para ver si estd abierto; the use of double negatives as in "We don't bought nothing new" from the Spanish--no compramos nada nuevo,' the omission of third person singular s in "She live there" from the Spanish---elle vive alli, etc.).

Analysis of Use. Clearly, Sandy had numerous problems of form. However, despite these errors and teachers' reports that she was nonverbal in class, she was a delightful communicator in a one-on-one situation. An analysis of use (function and conversational ability) revealed that Sandy used language for a variety of functions, often spoke spontaneously to inform and report, and used language to create complex episodes in pretend play. The functions of her utterances and her ability to converse were typical of a child whose content-form interactions were at least in Phases 6-8 or beyond (Lahey, 1988) and appeared to be age appropriate. (Example--Sandy: "When my daddy take my tooth and then not come off he get a paper and took it off. And my brother say don't be afraid and he help me every day." Clinician: "Well, that was nice of him." Sandy: "Yesterday he help me"). Sandy's utterances were contingent on those of others and on her own, and she took several turns around a topic so that she added information to each topic. When she was not understood, she attempted to repair her utterances with lexical substitutions or additions and, occasionally, syntactic revisions. An example of a conversation follows:

Clinician: Is Tina back in school now?

Sandy: Nope, she go to Argentina now.

Clinician: She's going where?

Sandy: Argentina where she live.

Clinician: Oh?

Sandy: I saw all the beautiful picture where

she live.

Clinician: ls it pretty?

Sandy: Like a garden, lots of flowers. I like it.

Clinician: Have you ever been to Argentina?

I bet you'd like to go there.

Sandy: I went like five times to Uruguay when

I was five or three.

Clinician: When you were five or three?

Sandy: Uh huh.

Clinician: And who lives in Uruguay? Do you

have relatives there?

Sandy: My cousin' s grandma live there and

Natasha. Is a little baby.

Clinician: Uh huh.

Sandy: She's only two year old.

Clinician: Yeah.

Sandy: And/cause/sometime my more um

write a letter.

Clinician: To Natasha or her Mommy?

Sandy: For her sisters.

(Conversation is interrupted but then continues.)

Sandy: Susie live there. She's my cousin.

Clinician: Susie's your cousin? How old is


Sandy: I don't know.

Clinician: Is she older than you?

Sandy: Older like my cousin Jackie. She live


Clinician: Jackie lives here, yeah. You always

talk about your cousin Jackie. She must

be nice.

Sandy: Cause she love me a lot. She haves a

pictures of me,

Clinician: She does?

Sandy: She's a babysitter and a ballerina

dance, too. And her mother' s a teacher.

Clinician: Does she teach little kids like you?

Sandy: No--bigger kids.

In summary, the analysis of content, form, and use of both samples revealed that content and use were far superior to form. Some errors of form in English appeared to result from Spanish syntactic interference or influence. This was true despite the fact that English was clearly the dominant and preferred language used for communication.

The analysis of the English sample included an evaluation of the seven "pragmatic" criteria identified by Damico, Oller, and Storey (1983) to be predictive of language disorders as differentiated from those with errors that resulted from "normal interlanguage processes." These pragmatic categories were linguistic nonfluencies, revisions, delayed responses, nonspecific vocabulary, inappropriate responses, poor topic maintenance, and need for repetition. (For definitions and procedures, see Damico et al.) For this analysis, "Any meaningful communication unit carrying a distinct and unified propositional meaning, even though it may not appear between clear phonological junctures" was regarded as an utterance (Damico et al., p. 387). Damico et al. found that the number of errors of form did not predict which children would be classified as language disordered. Children who were considered to have normal language potential had numerous errors of form but had pragmatic difficulties in no more than 20% of utterances in the dominant language, whereas those that were classified as disordered had difficulties in 30% or more of their utterances.

Our analysis indicated that only 12% of Sandy's utterances revealed what Damleo et al. (1983) termed pragmatic errors. Therefore, her score fell below the criterion for identification of a language disorder. The majority of her pragmatic difficulties were within the categories of linguistic nonfluencies, poor topic maintenance, revisions (described as "dead-ends"), and nonspecific vocabulary. This latter category was difficult to determine because Damico et al. excluded nonspecific vocabulary usage of words such as this, that, and thing when they reflected insufficient lexical knowledge, but included these same nonspecific terms when they incorrectly reflected what the speaker assumed the listener knew (e.g., the use of a pronoun when no antecedent could be inferred) (Damico et al., p. 388). As in all analyses, the reliability of this categorization was obtained by agreement of two of the three investigators.

Academic Performance

The California Achievement Test was administered by the school in the spring of the 1st grade. At that time, Sandy scored in the 88th percentlie for math computation, but in the 26th percentile for math concepts and the 28th percentile for reading. She was receiving resource room help for math, reading, and language. Written reports from her homeroom, math, and reading teachers were collected in the spring of the following year when Sandy was in the 2nd grade. At that time, Sandy was doing well academically and was at least on grade level for all subjects except reading. She was still in the lowest reading group but reportedly was making excellent progress. Her teachers indicated that she was more self-confident and was no longer considered to be "nonverbal.'' All were of the opinion that Sandy did not have a language disorder, but was delayed in reading because she was not yet fully proficient in English.

Sandy transferred out of the school district at the end of the 2nd grade. Information about 3rd grade performance was obtained from written reports and report cards given to the mother. In the 3rd grade, Sandy was on grade level in all subjects, including reading, and was receiving a grade of A in all academic areas except reading, in which she obtained a B. In the middle of the year, she was dismissed from the resource room for math and reading tutoring, but continued to receive speech-language therapy.

Results of Speech-Language Assessment--3rd Grade

Sandy was tested at the age of 8 years 5 months. The Peabody Picture Vocabulary Test-Revised (Dunn & Dunn, 1981) and the Test of Auditory Comprehension of Language-Revised (CarrowWoolfolk, 1985) were administered, and Sandy continued to score within the normal range on both tests. The subtests that involved expressive language from the ITPA (Kirk et al., 1968), in which Sandy had previously scored 1 or more standard deviations below the mean, were readministered (Auditory Association, Auditory Memory, and Grammatic Closure). These were now within the normal range. The Test of Language Development (Newcomer & Hammill, 1977) was readministered, and Sandy scored within the normal range on all subtests, except Sentence Imitation. Her language quotient was now 103, within the normal range, as compared to 76 when first administered by the child study team.

Informal observations during conversation revealed that Sandy had a marked interdental lisp that required intervention, but her articulation otherwise was excellent. Her spontaneous communication still contained some morphosyntactic errors. These mistakes occurred when she was giving involved explanations (e.g., a description of how lava is formed), but were not noticeable in ordinary conversational interchanges.


The possibility of language loss or arrested development in the primary language should be considered when evaluating all children with limited proficiency in both their first and second languages. This case study revealed the following possible diagnostic indicators that morphosyntactic errors reflect limited proficiency in English rather than a language disorder:

1. The child had normal developmental milestones.

2. The child was a second-language sequential

learner (rather than a child who was acquiring

both languages simultaneously) and had pre-

school/early childhood exposure to English in

a "school-like" rather than a naturalistic set-


3. The child reacted to second-language expo-

sure with a long, silent (nonverbal) period

when at school.

4. The child preferred to speak English rather

than Spanish in the home.

5. The child presented fewer than 20% errors on

the pragmatic criteria (Damico et al., 1983)

found to be related to academic success.

6. The child received normal comprehension

scores on standardized tests in English and in


7. Content and use of language as revealed by

analyses of language samples were better than

form. Sandy was an interesting conversation-

alist who tried to make herself understood.

Good content was indicated by the variety and

complexity of the utterances that she was try-

ing to communicate. Although Sandy's En-

glish utterances contained numerous errors

and omissions, her MLU was over 6.0. Be-

cause English clearly was the dominant lan-

guage, complexity of thought was maintained

in the Spanish sample by language mixing.

8. There was evidence of Ll confusion or inter-

ference in the English-language samples.

There were utterances in the English sample

that contained Spanish syntactic structure.

9. The caretaker identified the problem as one of

confusion of the two languages.

Primary language interference (e,g., the use of Spanish syntax with English vocabulary) may also occur in language samples of children with language disorders, and not all mothers are as astute about their children's problems as Sandy's mother. Normal comprehension and normal content and use of language can also be found in children with specific language impairments who demonstrate only a deficit of form. Therefore, a possibility of a language disorder cannot be entirely ruled out on these bases, and children like Sandy should be followed.

There is little doubt that Sandy benefited from the language tutoring she received, but this tutoring did not have to be done by a speech-language pathologist. Further, the terminology used in classification of language disorder, impairment, or communication disability carries with it the connotation of "not normal." Bilingual children with limited English proficiency are normal language learners. They do not have disorders, impairments, or disabilities. The dilemma that school systems must address is how to provide services to this group of children who are normal, but are experiencing primary language loss and temporary limited proficiency in English. As in Sandy's case, some of these children are seen at a point in development when their English is better than their Spanish, but not yet on grade level. If school systems do not provide adequate ESL support or bilingual support in the classroom, another avenue for services is necessary that would ensure that this group of normal children are provided with the help that they need without the communication disability or special education classification.

The results of this case study suggest that all assessments should include a comprehensive case history. In addition, an analysis of language samples should involve some determination of language interference and a comparative assessment of the levels of content, form, and use. When content and use are better than form, limited proficiency in English should be considered, even when English is the dominant and preferred language of the child.

More information on children with primary arrested development or language loss is necessary to determine whether normal comprehension occurs with any regularity in those children who eventually develop linguistic proficiency and achieve academic success. A comparison of other children like Sandy, who eventually catch up, with children with known bilingual language disorders should reveal additional patterns of differences.


American Speech-Language-Hearing Association. (1983). Position of the American Speech-Language Hearing Association on social dialects. Asha, 25(9), 23-25.

American Speech-Language-Hearing Association. (1985). Clinical management of communicatively handicapped minority language populations. Asha, 27(6), 29-32.

Brown, R. (1973). A .first language, the early' stages. Cambridge: Harvard University Press.

Burt, M., Dulay, H., & Hernandez-Chavez, E. (1976). Bilingual Syntax Measure. New York: Psychological Corporation.

Carrow, E. (1973). Test for Auditory Comprehension of Language. Austin, TX: Learning Concepts.

Carrow-Woolfolk, E. (1985). Test for Auditory Comprehension of Language-Revised. Allen, TX: DLM Teaching Resources.

Chamberlain, P., & Medinos-Landurand, P. (1991). Practical considerations for the assessment of LEP students with special needs. In E. V. Hamayan and J. S. Damico (Eds.), Limiting bias in the assessment of bilingual students (pp. 111-156). Austin, TX: ProEd.

Conklin, N. F., & Lourie, M. A. (1983). A host of tongues: Language communities in the United States. New York: The Free Press.

Cummins, J. (1980). Psychological assessment of immigrant children. Logic or intuition? Journal of Multilingual Multicultural Development, 1(2), 97- 111.

Cummins, J. (1984). Bilingualism and special education: Issues in assessment and biology. Austin, TX: Pro-Ed.

Damico, J. S., Oller, J. W., Jr., & Storey, M. (1983). The diagnosis of language disorders in bilingual children: Surface-oriented and pragmatic criteria. Journal of Speech and Hearing Disorders, 48, 385-393.

Dunn, L., & Dunn, L. (1981). Peabody Picture Vocabulary Test-Revised. Circle Pines, MN: American Guidance Service.

Dunn, L. M. & Markwardt, F. C. (1970). Peabody Individual achievement Test. Circle Pines, MN: American Guidance Service.

Dunn, L. M., Padilla, E. R., Lugo, D. E., & Dunn, L. M. (1986). Test de Vocabulario en Imagenes' Peabody. Circle Pines, MN: American Guidance Service.

Hamayan, E. V. & Damico. J. S. (Eds.). (1991). Limiting bias in the assessment qf bilingual students. Austin, TX: Pro-Ed.

Kayser, H. (1989), Speech and language assessment of Spanish-English speaking children. Language Speech & Hearing Services in Schools, 20(3), 226241.

Kessler, C. (1984). Language acquisition in bilingual children. In N. Miller (Ed.), Bilingualism and language disability: Assessnzent and remediation (pp. 26-54). San Diego: College Hill Press.

Kirk, S., McCarthy, J., & Kirk, W. (1968). The Illinois Test of Psycholinguistic Abilities (rev. ed.). Urbana: University of Illinois Press.

Kretschmer, R. E. (1991). Exceptionality and the limited English proficient student: Historical and practical contexts. In E. V. Hamayan and J. S. Damico (Eds.), Limiting bias in the assessment of bilingual students (pp. 1-38). Austin, TX: Pro-Ed.

Lahey, M. (1988). Language disorders and language development. New York: Macmillan.

Lambert, W. E. (1977). Effect of bilingualism on the individual. In P, A. Hornby (Ed.), Bilingualism: Psychological, soc'ial and educational implic'ations (pp. 5-27). New York: Academic Press.

Lambert, W. E ( 1981 ). Bilingualism and language acquisition. In H. Winitz (Ed.), Native language and foreign language acquisition: Vol. 379. Annals of The New York Academy of Sciences (pp. 9-22). New York: New York Academy of Sciences.

Langdon, H. W. (1983). Assessment and intervention strategies for the bilingual language disordered student. Exceptional Children, 50, 37-45.

Langdon, H. W. (1989). Language disorder or difference? Assessing the language skills of Hispanic students. Exceptional Children, 56, 160-167.

Language Assessment Battery, (1982). New York: New York City Public Schools.

Lee, L. (1971 ). Northwestern Syntax Screening Test. Evanston, IL: Northwestern University.

Manes, L. J., & Omark, D. R. (Eds.). (1984). Speech and language assessment for the bilingual handicapped. San Diego: College Hill Press.

Merino, B. (1983). Language loss in bilingual Chicano children. Journal of Applied Developmental Psychology, 10, 477-494.

Miller, N. (1984). Language problems and bilingual children. In N. Miller (Ed,), Bilingualism and language disability. (pp. 81-103). San Diego: College Hill Press.

Newcomer, P. L., & Hammill, D. D. (1977). The Test of Language Development. Austin, TX: Empiric Press.

Schiff-Myers, N. (1992). Considering arrested language development and language loss in the assessment of second language learners. Language Speech & Hearing Services in Schools, 23, 28-33.

Toronto, A. (1973). Screening Test of Spanish Grammar. Evanston, IL: Northwestern University.

Wechsler, D. ( 1967 ). Wechsle r Preschool and Primary Scale oflntelligence (WPPSI). New York: Psychological Corporation.

Zentella, A. C. (1990). Integrating qualitative and quantitative methods in the study of bilingual code switching. In The Uses of Linguistics: (Vol. 583)Annals of the New York Academy of Sciences (pp. 7592). New York: New York Academy of Sciences.


NAOMI B. SCHIFF-MYERS, Professor of Speech-Language Pathology, Communication Sciences and Disorders, Montclair State College, Upper Montclair, New Jersey. JANINE DJUKIC, Speech-Language Pathologist, Hackensack Hospital, Hackensack, New Jersey. JANINE MCGOVERN-LAWLER, SpeechLanguage Specialist, Montclair Public Schools, Montclair, New Jersey. DAISY PEREZ, Speech-Language Specialist, Paterson Public Schools, Paterson, New Jersey.

For correspondence contact: Naomi SchiffMyers, Montclair State College, Upper Montclair, NJ 07043.

Manuscript received September 1991; revision accepted November 1992.
COPYRIGHT 1993 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Schiff-Myers, Naomi B.; Djukic, Janine; McGovern-Lawler, Janine; Perez, Daisy
Publication:Exceptional Children
Date:Dec 1, 1993
Previous Article:Acceptability and feasibility of classroom-based social interaction interventions for young children with disabilities.
Next Article:Observations of students with learning disabilities in general education classrooms.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters