Donald E. Bebeau

On April 25, 1968, the first phase of this research was begun. A test of English as a foreign language, the Michigan Test Battery (see Note 1) was administered to 116 Indian high school students on the Rosebud Sioux Reservation. The students tested included 37 ninth grade and 18 twelfth grade pupils from Todd County High School, a public school in Mission and 35 ninth grade and 26 twelfth grade pupils from St. Francis Mission, a Catholic school at St. Francis. On, March 14, 1969, an additional group of 50 ninth grade and 25 twelfth grade pupils from Red Cloud Indian School, a Catholic school at Holy Rosary Indian Mission, Pine Ridge, were administered the same test.

Reasons for administering the test were:

The Michigan Test Battery was designed to be used in estimating whether a student whose native language is not English is able to pursue academic study in an English language college or university, and how much study he might be able to undertake at his present level of proficiency in English. The battery includes the Michigan Test of English Language Proficiency, a Test of Aural Comprehension, and an impromptu 30-minute written composition.

Part I, the grammar subtest is designed to show how well one can recognize and use English grammatical structures. It consists of 40 multiple-choice items, based on short dialogues. In each item part of the dialogue has been omitted and must be completed by the examinee by his selection of one of four "fillers." According to the test manual, "This technique has been selected to favor grammatical production by the examinee rather than simple recognition or parsing; also to avoid reliance upon his knowledge of the nomenclature of grammatical analysis." Part II, the vocabulary subtest, consists of 40 multiple-choice, substitution-type and selection-type items. Thirty items test words of the 4,000 to 6,000 most commonly used according to the Thorndike-Lorge general word count. Part III, the reading comprehension section, consists of four reading passages of 139 to 217 words, each followed by five multiple-choice items. The reading passages are non-technical in nature and have been selected to provide variety in style and subject matter.

The TAC is a 40-minute test to estimate a student’s ability to comprehend basic English structures presented orally as in a lecture situation. To administer the test, the examiner reads aloud to the students a series of 60 short sentences and paragraphs. The student demonstrates his comprehension by selecting from three pictures or phrases the one which best corresponds to the material read aloud by the examiner. The test was designed for students at any level of English proficiency. Native speakers of English at the University level make very high scores on it, as they properly should. The scores of students whose native language is not English range from 0 to 100 in a nearly normal distribution curve.

The 30-minute written composition shows whether a student is capable of handling English well enough to indicate his knowledge of subject matter in a course examination situation. It is not a comprehensive composition test. The instructions to examinee states that "Your composition will be graded primarily on the amount of information you can express in 30 minutes. . . . Errors of vocabulary, grammar, spelling and punctuation are not important if your intent and meaning are clear."

The only criterion established for selection of students to be tested, other than being ninth or twelfth graders, was classification as Indian by the schools they attended. At St. Francis and Holy Rosary, every member of the ninth and twelfth grade classes was present in school on the day testing was included, some of whom are less than 1/8 Indian. The students tested at Mission were selected from the integrated student body by the guidance counselors.

Approximately one-half of all students tested indicated English as their first language. All but one of the rest claimed the Lakota dialect of Sioux or "Indian" as a mother tongue (the one exception is a native speaker of Spanish).

One problem in determining accurately the number of native speakers of Lakota is that some students are reluctant to admit that English is not their native language. Unfortunately, such a statement is, to some, an admission of inferiority. According to teacher estimates, over 60 percent of St. Francis students do not learn English until they begin school. Fifty percent is a safe estimate of the number of non-native speakers at Todd County (Mission). A breakdown of test scores according to first language is provided below for Holy Rosary students, about 45 percent of whom (by their own statement) are native speakers of Lakota.

A breakdown of test scores in terms of degree of Indian blood is provided below for Pine Ridge students only. As will be noted, there is no significant pattern of differences between full bloods and others, although those of less than one-fourth Indian extraction do score the highest.

The differences among the groups tested have been suggested above. In general, Pine Ridge students represent the greatest number of mixedbloods and native speakers of English. St. Francis includes the largest number of full-bloods and non-native speakers. The St. Francis group is also the most isolated from towns of any significant size. The Todd County students are somewhere between the other two groups in terms of number of full-bloods and non-native speakers but attend an integrated school.

Because of the very high dropout rate among Indian high school students (estimates range as high as 50 percent in South Dakota), it can be assumed that the twelfth graders represent the most able of Indian students--or at least those who have managed to survive. The ninth graders represent a more general cross-section; this group includes some potential dropouts.

Table 1 summarizes test results showing mean scores and standard deviations. In comparison with foreign applicants, the twelfth graders are somewhat above average in grammar scores, somewhat below average in vocabulary scores and far below average in reading. The total of these three tests is just about at (for Todd County) or a bit above (for St. Francis and Pine Ridge) the foreign applicant mean.

Conclusions about ninth grade students are tenuous at best because the test was designed for a college or university applicant population. However, native speakers of English at the ninth grade level could be expected to do better than foreign applicants on all areas of the test except reading comprehension, which represents college level expectations. In the tables below only twelfth graders are compared with foreign student groups.

Table 3 is a breakdown by percentile comparing Indian student results on each subtest with results of foreign students representing various language groups.

Note. One native speaker of Spanish not included.

Table 5 presents a breakdown of mean scores on the entire battery in terms of degree, of Indian blood. No definite pattern appears other than, that those of less than one-fourth Sioux extraction do slightly better than those with a higher percentage of Indian blood. One would expect a significant pattern of differences in text scores on the basis of greater isolation of full-bloods, fewer native speakers of English among them, and closer involvement with their Sioux cultural heritage and much less involvement in, the surrounding non-Indian culture. For this not a good predictor of group, degree of Indian blood is apparently in English language proficiency.

Table 6 shows mean scores on each subtest of the MTB by grade and mother tongue. The table indicates that the areas in which native speakers of Lakota are most handicapped are vocabulary and aural comprehension. Note the Lakota speakers scored better than the native speakers of English on the composition section of the test. The reason for this is unknown, but one might guess that the Lakota-speaking students are more conscientious and therefore made a greater effort and used more of the time allowed for writing on the composition than did the other group. At least one examiner observed this to be the case.

Note. One native speaker of Spanish is not included.

1. In general the competence level of twelfth grade students tested in basic phonological and grammatical structure is more limited than that of the native speaker but more extensive than that of the foreign students for whom the test was designed. This is true despite the fact that 50 percent of the students tested indicated English as their native language.

John A. Upshur, Director of Testing and Certification at the English Language Institute, comments: "The ability (of the Indian students tested) to use this somewhat limited knowledge for communication is relatively high in comparison with foreign applicants as indicated by high aural comprehension and composition scores. Lower scores in vocabulary and reading seem to indicate a narrow range of experience in the use of English."

2. Reading comprehension is the area of greatest weakness of all students tested. Compared with foreign students, the twelfth graders tested would be at about the 23rd percentile; ninth graders would be below the 10th percentile.

3. Vocabulary acquisition is a significant problem for most of the students tested, especially for those who learned Lakota as a first language.

4. Those students who claim English as a first language scored somewhat higher than those whose first language is Lakota.

5. All students scored rather high on aural comprehension and composition sections in comparison with foreign students, but not as high as would be expected of native speakers of English.

6. For those whose first language is Lakota, vocabulary and aural comprehension are the most significant areas of difficulty and the areas which most significantly differentiate their proficiency in English from that of native speakers.

7. Differences in degree of Indian blood are not, at least for the group tested, significant indicators of English proficiency, although those who have the largest percentage of Indian blood are least likely to acquire English as a first language.

8. Unfortunately, attendance at an integrated school with a significant non-Indian population is apparently not a solution to lack of English proficiency on the part of Indian students. The twelfth graders at Mission (Todd County) scored lower, in fact, than either of the other groups tested. Why this is so is not immediately obvious. Examiners noted, however, that in integrated classes Indian students are reluctant to compete with their white classmates. They tend to sit in the back of the room and function as observers rather than participants, despite the teacher’s efforts to involve them. As a result the teachers seem naturally to direct their teaching to those who respond--the non-Indian students in the front of the class. Although Indian students are seldom competitive, this situation cannot occur in the mission schools, in which all are considered Indian and there is no apparent "out" group.

9. Techniques characteristic of the teaching of English as a second language seem to be indicated by the test results. It is probably in the lower grades, however, that such instruction is most needed, especially in the areas of vocabulary acquisition and aural comprehension. Reading comprehension is also an area needing special attention in the lower grades.

10. The most obvious implication of test results is for teachers to be aware that in some areas of English proficiency their Sioux students are more like non-native speakers of English than like native speakers.

It should be pointed out in conclusion that language problems among Indian students are often only surface reflections of underlying sociological and psychological problems; that English instruction is only one of many variables affecting English proficiency; and that proficiency in English as one of many factors related to success in school. Environmental factors are also significant to the test results. It is quite possible that non-Indian students from rural South Dakota would show the same lag in proficiency demonstrated by Indian students.

1. The Michigan Test of English Language Proficiency, Form C (1962) and Test of Aural Comprehension, Form A, English Language Institute, University of Michigan.