Volume 21 Number 1
January 1982

Herbert L. Bacon, Gerald D. Kidd and John J. Seaberg

BILINGUAL education for the American Indian is not a new educational concept. During the years before Oklahoma statehood, several Indian tribes had established educational systems which utilized bilingual instruction in the classroom with the Indian child (see Note 1).

The Cherokee tribe was unique in the area of bilingual instruction among the Indian tribes because they had their own written language. Many historians and educators felt that the Cherokee syllabary facilitated bilingual instruction in the classrooms of the schools within the Cherokee Nation. The pre-statehood educational system of the Cherokee Nation has been recognized as being successful in making Indian students literate in both the Cherokee and English languages.

When Oklahoma became a state, the schools operated by the Cherokee Nation were turned over to the new state government and bilingual instruction was discontinued. After the transition was accomplished, a reduction in the literacy level of the Cherokee people was noticeable in the years that followed. Another noticeable reaction that followed the transition was the increase in the number of Indian students who dropped out of school (see Note 2).

The attitudes and policies of the federal government concerning bilingual education and the use of the Indian languages in the federal boarding schools changed several times during the period between 1860 and 1968. In 1968, the federal government enacted legislation which provided funds for bilingual education in the public schools on a limited basis. Since 1968, the attitude of the federal government has been positive toward bilingual education for the Indian child (see Note 3).

The purpose of this study was to determine if the eighth grade reading and mathematics achievement scores of Cherokee Indian students who received bilingual instruction in grades one through five were significantly different from the reading and mathematics achievement scores of comparable eight grade Cherokee Indian students who did not receive bilingual instruction. A secondary purpose of this study was to determine whether the reading and mathematics achievement of five-year bilingual education participants was significantly different from four-year bilingual education participants.

The Cherokee tribe in northeastern Oklahoma was identified as one of the American Indian tribes with the greatest need for bilingual education. The Cherokees were identified as having a need because of their high dropout rate from the public schools, their language deprivation, and their high unemployment and poor economic conditions.

The Cherokee Bilingual Education Program was funded and began operation in 1969 in four rural dependent schools in Oklahoma. After two years of operation, two of the original schools remained in the program. Because the project schools did not have bilingual teachers, bilingual Cherokee Indian teacher aides were employed to work with the classroom teacher in a team effort to provide bilingual education for the Indian child.

The lack of appropriate bilingual curriculum materials limited the early effectiveness of the program. In 1971, the first classroom curriculum materials developed by program personnel were introduced into grade one and from 1972 through 1975 curriculum materials utilizing the Cherokee language through the Cherokee syllabary and Roman script were placed into operation in grades two through five.

Students who had the opportunity to benefit from the full utilization of the bilingual materials and bilingual instruction were first graders in 1971, and received bilingual instruction through grade five. Students who participated in the program in 1971 as first graders were eighth graders in 1978, and were selected for participation in this study as a treatment group.

Students who entered the bilingual program as second graders in 1972, received four consecutive years of bilingual instruction that was comparable to the bilingual instruction received by the five-year participants in grades two through five. Students who entered and participated in the Cherokee Bilingual Education Program in 1972 as second graders were eighth graders in 1978 and were selected for participation in this study as a treatment group.

Indian students from non-project schools who were eighth graders in 1978 were screened using the following participation criteria: they must have had a Cherokee blood quantum of one-quarter or more; they must have never been a participant in a bilingual project; and they must have never failed a grade level. After screening students, those who met the control group participation criteria, were pooled if they were within the I. Q. range limits of the treatment groups. The control group was selected for participation in this study by a random drawing from the students who were pooled.

The SRA Achievement Series was administered to eighth grade students in the project schools and selected control schools. The achievement test results (reading and mathematics) were used in this study as the dependent variable. The Short Test of Educational Ability was administered in conjunction with the achievement test and was utilized in this study as a control variable. The control variables: age, sex, degree of Indian blood, Indian tribe, and pass-fail record, were obtained for each participant. The control variable, grade point average, was derived from letter grades taken from permanent records. The control variables, speaker of Cherokee and father's educational level, were obtained from the parents or family members through an interview.

Specific criteria were utilized to select participants for the two treatment groups and one control group. Treatment Group I was an intact group that had 17 students who participated in the bilingual program for five consecutive years beginning with grade one. Treatment Group II was an intact group that had 18 students who participated in the bilingual program for four consecutive years, beginning in the second grade. Group III, the control group, was a randomly selected group that had 18 students who did not receive bilingual instruction.

The multiple regression analysis of covariance was utilized as the primary statistical technique in this study as it enables researchers to use intact groups and equate them with the identified control variables. After equating the three groups with the control variables, and "f" value was calculated using the test of regression coefficients to determine if the reading and mathematics achievement scores between the treatment and control groups were significantly different. To determine if the reading and mathematics achievement scores were significantly different between the two treatment groups, a priori comparison method utilizing a "t" ratio formula was used.

An analysis of the data of the SRA Achievement Series revealed the following:

1. When the reading achievement scores of Indian children who received bilingual instruction for five consecutive years were compared to the reading achievement scores of comparable Indian children who did not receive bilingual instruction, the Indian children who received bilingual instruction scored significantly higher on the reading achievement subtest than did those not receiving bilingual instruction.

2. When the reading achievement scores of Indian children who received bilingual instruction for four consecutive years were compared to the reading achievement scores of comparable Indian children who did not receive bilingual instruction, the children who received bilingual instruction scored significantly higher on the reading achievement subtest than did those not receiving bilingual instruction.

3. When the reading achievement scores of Indian children who received bilingual instruction for five consecutive years were compared to the reading achievement scores of comparable Indian children who received bilingual instruction for four consecutive years, there were no significant differences in their performance on the reading achievement subtest.

4. When the mathematics achievement scores of Indian children who received bilingual instruction for five consecutive years were compared to the mathematics achievement scores of comparable Indian children who did not receive bilingual instruction, the Indian children who received bilingual instruction scored significantly higher on the mathematics achievement subtest than those who did not receive bilingual instruction.

5. When the mathematics achievement scores of Indian children who received bilingual instruction for four consecutive years were compared to the mathematics achievement scores of comparable Indian children who did not receive bilingual instruction, the children who received bilingual instruction scored significantly higher on the mathematics achievement subtest than did those not receiving bilingual instruction.

6. When the mathematics achievement scores of Indian children who received bilingual instruction for five consecutive years were compared to the mathematics achievement scores of comparable Indian children who received bilingual instruction for four consecutive years, there were no significant differences in their performance on the mathematics achievement subtest.

When reviewing the conclusions, one should understand that the data we collected for only those selected students who participated four and five consecutive years in the Cherokee Bilingual Education Program. The data collected on the student population in this study support the following conclusions:

The authors recommend the following:

1. Additional study should be given to the relationship between bilingual education when presented in the first through the fifth grades and reading achievement utilizing the student population of this study during their senior year in high school.

2. Additional study should be given to the relationship between bilingual instruction and academic achievement using other measuring instruments.

3. A longitudinal study should be undertaken to correlate reading and mathematics achievement to bilingual education utilizing matched treatment and control groups who are pre- and post-tested in each year of program participation.

4. An effort should be made to encourage the employment of bilingual teachers by schools where language deprivation has been identified among the Indian student population.

Table I
ANALYSIS OF REGRESSION FOR THREE GROUPS IN READING ACHIEVEMENT
FOR GRADE EIGHT UTILIZING THE FULL MODEL

Table II
SUMMARY OF ANALYSIS OF COVARIANCE TEST OF REGRESSION
COEFFICIENTS ON READING ACHIEVEMENT

TABLE III
DEPENDENT AND CONTROL VARIABLE MEANS FOR THREE GROUPS OF BILINGUALLY AND
NONBILINGUALLY INSTRUCTED STUDENTS IN READING ACHIEVEMENT SCORES

*Sex: Male = 1.0, Female = 2.0.

**Speaker of Cherokee: Yes = 1.00, No = 2.00.

***Degree of Indian blood: 1/4 = 2.5, 1/2 = .50, 3/4 = .75, 4/4 = 1.00.

Table IV
ANALYSIS OF REGRESSION FOR THREE GROUPS IN
MATHEMATICS ACHIEVEMENT FOR GRADE EIGHT
UTILIZING THE FULL MODEL

Table V
SUMMARY OF ANALYSIS OF COVARIANCE TEST OF REGRESSION
COEFFICIENTS OF MATHEMATICS ACHIEVEMENT

Table VII
DEPENDENT AND CONTROL VARIABLE MEANS FOR THREE GROUPS OF BILINGUALLY AND
NONBILINGUALLY INSTRUCTED STUDENTS IN MATHEMATICS ACHIEVEMENT SCORES

*Sex: Male = 1.0, Female = 2.0.

**Speaker of Cherokee: Yes = 1.00, No = 2.00.

***Degree of Indian blood: 1/4 = 2.5, 1/2 = .50, 3/4 = .75, 4/4 = 1.00.

CHEROKEE BILINGUAL EDUCATION PROGRAM

Student's name

CHEROKEE BILINGUAL EDUCATION PROGRAM

Student's Name

Date of Birth

Degree of Indian Tribe (JOM Data)

Lives With? (Tell Relationship)

Where is the father employed?

Where is the mother employed?

Father's Occupation?

Mother's Occupation?

Has this student ever failed a grade?

If the answer to the above question is yes, what grade?

Does the student speak the Cherokee language?

Does the student understand the spoken Cherokee language?

Is the student transported to school?

Does the student live in an Indian home built recently (within the last eight years)?

1. Willard Walker, "Notes of Native Writing Systems and the Design of Native Literacy Programs, Anthropological Linguistics, Vol. II, No. 5, 1968

2. Arnold H. Leibowitz, "A History of Language Policy in American Indian Schools," Bilingual Education for American Indians, Curriculum Bulletin No. 3 of the Bureau of Indian Affairs, Washington: U.S. Printing Office, 197 1.

3. Theodore Anderson and Mildred Bayer, Bilingual Schooling in the United States, Vol. 1. Washington: U.S. Printing Office, 1968.

4. Interview with Patricia Hammons, Oklahoma State Department of Education, Oklahoma City, March 1971.

Herbert L. Bacon is Assistant Superintendent of the Claremore (Oklahoma) Public Schools: Gerald D. Kidd and John J. Seaberg are both Associate Professors of Education at the University of Oklahoma.