Volume 21 Number 1
January 1982

Kathryn A. Bayles and Gail A. Harris

In 1973, the New Mexico State Department of Education conducted a statewide survey to identify handicapped children and the results were presented to the Congressional Subcommittee on Indian Affairs by Nicolais and Joyner (see Note 7). Children with speech disorders accounted for 7.8 percent of the handicapped population, hearing impairments for 3.9 percent, and language disorders for 7.3 percent. Nicolais and Joyner also presented incidence data from a 1974 speech-language screening of 2,454 Navajo elementary school children. Seven percent had speech problems and 67 percent had English language problems. Miles (see Note 6) found 35 percent of 150 Pueblo Indian children in New Mexico Headstart Programs to be in need of speech-language therapy.

Although prevalence data on communication handicaps among Native Americans consistently are higher than for the general population they vary markedly from each other, which may result from differences in screening procedures and diagnostic criteria. Generally, empirical data about the diagnosis and management of speech-language disorders of this ethno-racial minority are scant.

As part of a training program for Native Americans in speech and hearing sciences, speech-language pathologists from the University of Arizona conducted speech-language screenings of 583 children on the Papago Indian Reservation. The purpose of this report is to present the screening results, describe patterns of English usage among this population, and discuss the differentiation of speech-language disorders from dialectical differences

The 583 Papago Indian children screened comprised the total school population of two elementary schools located on the Papago Reservation in southwestern Arizona. The reservation is equivalent in size to the state of Connecticut. Papago, the tribal language, is spoken by an estimated 14,000 individuals,5 and many reservation children are exposed to both English and Papago in their preschool years. During screening, children were asked which language was preferred by members of their families. Sixty-eight percent designated English as the dominant home language; however, a Papago speaker resided in most homes. Twelve percent of the children specified Papago as the dominant home language, and 20 percent reported the languages were used with equal frequency.

Specific measures were used to screen each child's proficiency in fluency, voice, receptive and expressive language in English, and use of the speech mechanism. Hearing screening was not included because recent pure-tone audiometric screening results from a state agency were available in both schools.

To evaluate articulatory proficiency, the "Sounds-in-Words" subtest of the Goldman-Fristoe Articulation Test3 was selected. Through picture naming and questioning, 44 responses were elicited. While common dialectal differences such as placement of stress, de-emphasis of final consonants, and diminished aspiration were not designated as errors, sound substitutions, omissions, additions, and distortions were. The narrative "Jack and Ricky" from the Goldman-Fristoe test was used to elicit expressive language samples. This eleven sentence test is read by an examiner as an accompaniment to a four-picture sequence. Children are subsequently shown the pictures and asked to recount the story and their responses are transcribed verbatim.

Receptive language skills of kindergarten through third grade children were screened with Carrow's Screening Test for the Auditory Comprehension of Language (STACL)1. In this 25-item test of concepts and syntax, a child is instructed to point to that picture, from among three, that best depicts the examiner's verbal stimulus. Because this test has not been normed on a southwestern Indian population, local norms were developed. Means and standard deviations were calculated for kindergarten, first, second, and third grade Indian children. The performance of each child was compared to the class mean.

In the sentence repetition task, children were asked to repeat eight syntactically varied sentences. A record was made of each child's errors and compared to the average error rate of the class.

Students in the fourth through sixth grades, for whom the STACL was inappropriate, read an appropriate selection for their grade from an informal reading inventory developed by the University of Arizona reading department. As with the STACL and sentence repetition task, the performance of each child was compared to the class average. All screening measures were administered in one session by one of three certified speech-language pathologists.

A communication profile containing the following information was constructed for each child: number and type of articulation errors and intelligibility rating of articulation (normal, slight, moderate, and severe), receptive language score on the STACL, or rating on the reading inventory, number and type of errors made on the sentence repetition task, verbatim transcription of the retold story, reported language preference of the home, and the results of hearing screening.

A child exhibiting dysfluency was not described as having a "disorder" before the parents were interviewed and the child's English language skills evaluated. The purpose of the parental interview was to discover the child's exposure to Papago and English and to determine whether or not the dysfluency was present in both languages. Children with poor English skills who were fluent in Papago were not considered fluency "disordered," or stutterers. Unless a child was dysfluent in both languages, she/he was not classified as "disordered."

Distinguishing articulation and fluency disorders was not as difficult as identifying children with language disorders, primarily because the population screened speaks a dialect of English. If English could be assumed to be a second language for this population, the children's development and use of the first language (Papago) could be compared to the development and use of English. However, for many children in this population, English is not a second language. The majority of children are exposed to both languages from birth. Thus Papago children may be less proficient in standard English than monolingual English speakers of the same age, and less proficient in Papago than were their monolingual Papago speaking grandparents at similar ages. Unfortunately a Papago, speaker trained to make such judgments about the children's development and use of Papago was not available.

The method for identifying children with signs of language disorder was to compare the performance of each individual to that of his/her age mates. Children were suspected of having a receptive language problem in English if (a) their scores on the STACL were a standard deviation or more below their class mean, and (b) their reading performance and ability to repeat sentences were below average. Judgments of the adequacy of expressive language skills were made by comparing transcripts of the retold narratives for length, complexity, and logic of the utterances. Each corpus was judged average, below average, or above average.

Eighty children were identified during screening as having speech and language handicaps, or 14 percent of the elementary school population. Delayed English language development was found in 61 children, or 10 percent of the population. Articulation handicaps totaled 29 representing 5 percent of the population, and the five children judged to be dysfluent represented 0.6 percent of the population. Prevalence of speech and language handicaps according to grade is presented in Table 1.

The mean score of 93 kindergarten children, ages five to six, was 14.8 with a standard deviation of 3.07 which falls at the 23rd centile on the STACL. The mean for the 84 first-grade children was 18.6 with a standard deviation of 2.75 which corresponds to the 12th centile of the normative population. The STACL does not provide norms for seven- and eight-year old children, consequently the mean and standard deviation of the Indian child in this age group were compared to those of six- and seven-year olds in the normative group. The 92 children, ages seven to eight, had a mean score 20.8 or the 45th centile for sixand seven-year olds. These findings suggest that Papago children in their reservation schools may be slightly more than a year behind Anglo children in their development of English comprehension skills. Items presenting frequent difficulty were the concepts "pair," "second," and "neither . . . nor," verb tenses "has eaten," "had been cutting," "is chased," the noun "bicyclist," and distinguishing between the direct and indirect object in "She shows the girl the boy."

Table 1
Prevalence of Speech-Language Handicaps by Type and Grade Level

Table 2
Comparison of the Difficulty of Syntactically Varied
Sentences Presented in the Sentence Repetition Task

Table 3
Comparison of Frequency of Articulation Handicaps
Among Papago Children with the General School
Population as Reported by Hull et al. (1976)

Table 4
Centile Correspondences on the Goldman-Fristoe for Papago Indian
Children with Articulation Problems
(Does not include children in the kindergarten
or special education classes)

Of the eight syntactically varied sentences repeated by the students, the tag question and relative clause were most frequently misstated. The sentence with the relative clause was also the longest; thus, length may have contributed to its difficulty. In Table 2, each stimulus sentence is listed with its proportion of the total errors.

Five percent of the Papago Indian elementary school children had misarticulations, a rate considerably lower than that reported by Hull, et al. In a national study of the articulatory proficiency of the general school age population. In Table 3, comparison is made of the frequency rate of articulation handicaps among Papago students with rates reported by Hull, et al. which were specified according to severity.

Like the non-Indian elementary school population, most misarticulations involved the s or z phonemes or an s blend. In more than 75 percent of the cases, the number of misarticulated. sounds was three or less. The centile correspondences of the Goldman-Fristoe Articulation Test for Papago children with articulation problems are presented in Table 4.

Although standard English is spoken by some Papago students, the majority speak a dialect of English that may reveal influences of Papago grammar on English. Mathiot (see Note 5) has discussed the linguistic differences between Papago and English and emphasized the lack of correspondence between aspects of the grammars to explain why certain aspects of English portend difficulty for Papago speakers. First, Papago has a more highly developed inflectional system that uses affixes to convey meanings that are expressed by separate words in English, for example particles and auxiliary verbs. Errors involving the auxiliary verb were frequently found in the language samples in this study:

Tense is not marked by Papago verbs but is expressed by adverbs of time, which may account in part for the numerous verb tense errors evident in sentences like:

In addition, items on the STACL that test knowledge of tense marking were among those frequently missed.

In both Papago and English, adjectives and adverbs are allied with particles to express degree. Formations of the comparative and superlative adjectival forms in English are accomplished by the words "more" and "most" or the addition of the comparative suffix "er" and superlative suffix "est." In Papago, only particles are used to express forms of grading which may explain the frequent omissions of the superlative and comparative suffixes from the adjective base. Often, "more" or "most" was supplied before an uninflected adjective as in "more big" and "most big. " Mixing of the two forms was seen in the addition of the superlative suffix "est" to "most" to produce "mostest."

Nonstandard use of prepositions was common in the language samples. Papago has a word class of post-positions which are functionally equivalent to English prepositions except that post-positions occur either before or after a noun phrase. English prepositions were not omitted by students but often were misused, as in these examples:

According to Mathiot (see Note 5), prepositions may be hard for Papagos to master because they are unstressed in English.

Verbs that are irregular in the formation of the past tense consistently were formed incorrectly by Papago children who added an "ed" suffix. Such overgeneralization suggests that refinements of English grammatical knowledge may occur later in this population, perhaps because of less exposure to standard English. Sentences like the following occurred in most language samples:

In the first example, the child substituted the a vowel for the a, which is appropriate for the past tense form of this irregular verb while adding the --ed" past tense suffix appropriate for regular verbs.

English personal pronouns express gender and case, two aspects which are absent in Papago, and may explain the following sentences:

Certain usage patterns such as topicalization, overuse of conjunctions, incorrect plural formation of irregular nouns, lack of subject-verb inversions and "do" insertion indicate that acquisition of some rules of English grammar occurs later in these children. The following constructions found in the language samples of school age Papago children are similar to those found in language corpuses of preschool monolingual English speakers:

Judgments about language development and diagnoses of disorders pose significant problems for clinicians conducting speech-language screenings of Native Americans. One reason is the paucity of tests that are known to be valid and reliable with this population. Also data on the effect of test variables among different Native American groups are scant. Clinicians do not know the effect of an examiner's ethnic background, sex, race, and age on the test performance of Native American clients, nor the effect of test setting and activity. Further, unless the examiner speaks the tribal language or has access to a trained language informant, comparisons of the linguistic structure of the tribal language with English are not possible.

Until there are empirical data about diagnosis and management of speechlanguage handicaps in Native Americans, and native speakers trained to evaluate language development in tribal languages, clinicians will have to rely on alternative strategies for evaluating Native American children. For example, a clinician can determine a client's exposure to English and the tribal language. To make such a determination, information should be secured about the frequency of English use and the use of the tribal language in the community and home. The assumption can not be made that English is a second language. Many reservation families are in a transitional stage from sole use of the tribal language to the use of English. Such a transition is apparent on the Papago Reservation. Many older individuals are monolingual Papago, while school children prefer English. Frequently, grandparents reside in homes and share responsibility of caring for children which makes knowledge of the composition of the home clinically relevant. If a child lives in a remote village without television and is spoken to in Papago by his parents, his development of English is likely to differ markedly from an Indian child with English speaking parents, a television, and a home near a metropolitan Anglo community.

Seeking assistance of a Native American who is bilingual in English and the tribal language may be a useful strategy. Such individuals can assist clinicians in developing an appreciation of the differences in language acquisition patterns among children who are monolingual in their tribal language, those who are exposed to both English and the tribal language, and those who are monolingual in English. Additionally, a bilingual informant can help analyze the linguistic structure of English relative to the tribal language in order to identify differences in the phonology and grammar of the two languages. Linguistic devices that are present in a dominant language, but absent in a minor, may be hard to master.

To use language tests, clinicians can establish local norms, (a) by administering tests to at least 50 randomly selected children of the same age and (b) by calculating performance means and standard deviations. This facilitates comparisons of each child to his age mates and helps to identify children with language problems. When a child scores below the mean of his age mates, further inquiry should be made about language competencies and caution exercised in labeling the child as disordered.

When language tests are used, the clinician can analyze the most frequently missed test items to discover rules of standard English that are not used in the English dialect spoken by the tribe. Obviously, linguistic expertise is a valuable asset for clinicans serving Native Americans. Additional information is needed on how best to prepare speech-language pathologists for working with individuals from Native American cultures who speak non-standard dialects of English.

1. Carrow E., Screening Test for Auditory Comprehension of Language. Austin, Texas: Learning Concepts, 1973.

2. Evard, B.L., and Sabers, D.L., Speech and language testing with distinct ethnic-racial groups: A survey of procedures for improving validity. Journal of Speech Hearing Disabilities, 44, 271-281 (1979).

3. Goldman, R., and Fristoe, M., Goldman-Fristoe Test of Articulation. Circle Pines, Minnesota: American Guidance Service, Inc., 1969.

4. Hull, F.M., Mielke, P.W. Jr., Willeford, J.A., Timmons, R.J., National Speech and Hearing Survey. U.S. Department of Health, Education, and Welfare, 1976.

5. Mathiot, M., English and Papago compared, in Paul Turner (Ed.) Bilingualism in the Southwest. Tucson: The University of Arizona Press, 1973.

6. Miles, M., Indian Health Service Sensory Disabilities Program: Five-Year Plan. Jamil Toubbeh (Ed.), November, 1978.

7. Nicolais, J. P., and Joyner, J. B., Report Presented to the Indian Affairs Subcommittee an Interior and Insular Affairs, Washington, D.C., September 26, 1975.

8. Toubbeh, J., Indian Health Service Sensory Disabilities Program: Five-Year Plan. November 1978.

Kathryn Ann Bayles holds the Ph.D. degree from the University of Arizona, Tucson, where she is a research associate in the Department of Speech and Hearing Sciences. She has been a tutor-supervisor of master-level graduate students training to become professionals in speech-language pathology and audiology; a consultant in communication disorders for BIA schools; and director of two summer language programs for Indian children trying to learn English. Her published works include a chapter in Linguistics: An Introduction to Language and Communication.

Gail Ann Harris holds the M.S. degree from the University of Arizona where she is program coordinator-American Indian professional training in speech pathology and audiology, also in the Speech and Hearing Sciences Department. She has also been consulting speech-language pathologist to the I I affiliated tribes of Arizona through Program Development for Pre-School Handicapped Indian Children. Her published work includes test and models in speech.

The authors express their appreciation to Richard Curlee and Steve Boney for their help in the preparation of the manuscript. This research was supported by Bureau of Education for the Handicapped Grant G007901329. Requests for reprints should be addressed to Kathryn Bayles, Department of Speech and Hearing Sciences, University of Arizona, Tucson, Arizona 85721.