Volume 17 Number 1
October 1977

Rosemarie McCartin and William J. Schill

THE literature on children’s learning is replete with opinions to the effect that individual differences account for different learning patterns or preferences (see Notes 9, 15, and 17). At best, empirical studies have given meager support to the contention that racial minority groups learn in patterns different from the majority white middle class. Lesser, et al. (see Note 10) did find differences in ethnic group membership--Jewish, Negro, Puerto Rican and Chinese--to produce difference in mental abilities (verbal ability, reasoning, number facility, and space conceptualization). The authors contend that ethnic group affiliation strongly affects the pattern or organization of mental abilities in six and seven year-old children.

Although few have studied how the children of a given Indian tribe learn, there is information on how they score on tests. On nonverbal intelligence tests, Indian children have the same average scores and show the same range of performance as white children (see Note 6). In reviewing the literature, Havighurst (see Note 6) found that all of the studies since 1935 which used nonverbal tests supported the conclusion that there is no significant difference in intelligence between Indian and white children. Indian children have the same average I.Q. scores and range of performance between tribes and between communities within tribes as white children between and within communities.

The most commonly employed nonverbal I.Q. test used with Indian children is the Goodenough Draw-A-Man Intelligence Test (DAM). The DAM is a test of mental alertness which does not require language skills. Test results from several studies show normal I.Q.’s for Indian children tested on the DAM (see Notes 4, 7, and 11).

Typically, Indian children possess limited verbal skills or have had less opportunity and incentive to perform well on verbal intelligence tests. At least Indian children score less well on verbal intelligence tests than they do on nonverbal tests. In one study, a group of 30 Sioux pupils was tested one year apart with two different instruments. The average I.Q. score from the Grace Arthur Performance Test of Intelligence, which is a nonverbal test, was 102.8. The average I.Q. score for the same group on the Kulhmann-Anderson Test, a verbal test requiring reading ability, was 82.5 (see Note 8). The difference in scores may be attributed to either the verbal learning or the verbal performance problems experienced in handling standard English, and/or the possible discriminatory nature of verbal intelligence tests used with Indian pupils. Most studies indicate that Indian pupils’ performance on intelligence and achievement tests is better in the elementary grades than in the junior and senior high school (see Notes 3, 11, and 14). This may simply reflect the increasing reliance upon verbal tests as pupils progress through school.

If performance on tests is any indication, it appears that the American Indian child has greater facility in learning when visual methods of instruction are used (see Note 16) and performs less well when tasks are saturated with verbal content (see Note 18). Somewhere between written and pictorial communication lies the Indian’s traditional method of communicating--the oral presentation.

The elementary school in which the students were enrolled is situated on a reservation in many ways typical of North American Indian reservations. The reservation has beaches, rivers, lakes, and forests.

There are slightly more than 1,000 Indians living on, or adjacent to, the reservation. The village, seat of the Tribal Nation, has approximately 600 residents and another nearby settlement of 200. The village has a store, two restaurants, a fire station, public health office, library and Community Action Program building, and an elementary school.

The elementary school, part of the state system of public instruction, houses kindergarten through grade eight and a pre-school program. There is one teacher for each grade and about 20 students per grade. Special funding has permitted the employment of Indian mothers from the village as teacher aides.

The Indian students have no bilingual problems. They have mastered English, and much to the chagrin of a number of the older people, the tribal language has all but disappeared. There are few, if any, unattended health problems. A Public Health Clinic lies within two blocks of the school and the nurse, doctor, and dentist make periodic checks of all the students.

The age-range of the students was somewhat greater than elementary schools throughout the country, although like most elementary schools, a social promotion process was operative.

Achievement was recorded in the form of scores on the Comprehensive Test of Basic Skills (CTBS). Scores were available from tests administered in the spring of 1970, 1971, and again in 1972. Of the 103 children in the study, test scores for 1972 available for 94 resulted in a mean grade placement score of 3.97 with a standard deviation of 1.53. It is not surprising that the mean grade placement is below the expected placement since the CTBS is a verbal test of achievement.

Pearson product-moment correlations were computed for age, grade, and CTRS total scores, point Biserial for sex and CTBS total scores.

*Significant at %5

The age-grade correlation is as expected given the social promotion policy. However, the age and grade correlations with achievement (CTBS) account for less than 25 percent of the variance. The low correlation between sex and achievement is contrary to other studies of elementary students.

Otis-Lehman test scores were available for 89 of the 103 students in the study and although this I.Q. measure is highly correlated with achievement, the data were recorded and are presented here to provide further discussion of the Indian students and their familial setting. To test the interrelatedness of family variables and measured I.Q. and achievement, correlations were computed. The resulting correlations were not sufficiently high to be significant.

The lack of attendance is often used to explain achievement deficits for certain groups of children. Inspecting the attendance records of the students, it was found that their attendance is very good in the grades three through six. In the seventh and eighth grade, the attendance records-become spotty for a few female students.

In summary, the setting and the Indian student characteristics presented seem to indicate that the setting and the pupils involved are representative of American Indian culture. The literature cited suggests that American Indian children have. greater facility in performing when visual methods of instruction are used and perform less well when tasks and tests are saturated with verbal content such as reading. The remainder of this article presents the methods, hypotheses, analyses and conclusions of a study conducted in an all-Indian school.

To assess the relative value of three instructional methods, two conditions had to be met. The information or content of the lessons had to be met. The information or content of the lessons had to be new and of general interest to the pupils at all grade levels. To satisfy these conditions, the study of the nature of cities was selected for its interest appeal and because the content was not available in elementary text form. Chauncy Harris and Edward Ullman (see Note 5) propose the following as types of city functions: 1) Cities as central places performing comprehensive services for a surrounding area; 2) Transport cities performing break-of-bulk and allied services along transport routes; and 3) Specialized-function cities performing one service such as mining, manufacturing, fishing or recreation for larger areas.

Text material was prepared covering the three concepts at the third grade readability level, as measured by Fry formula (see Note 12) for grades three and four. In addition, other text material was prepared for the middle and upper grades covering the concepts of city functions to be presented in three class periods of 30 minutes each.

The literature provided some evidence, though inconclusive, that Indian children may learn best when the learning situation provides visual and oral cues in preference to a written text. Therefore, three methods were selected, each placing emphasis upon either a written, oral or visual mode of presentation. In method one, the text presentation provided for a set of written textual materials and pictures, one for each of the three concepts. In method two, the same material was presented orally by the teacher. Pictures used to clarify the concepts were 10 x 12 inch versions of those appearing in the text. In method three, the teacher used an overhead transparency of the same pictures as those used in the written and oral presentation.

The experiment used a 3 x 3 x 3 design. The three dimensions--grade, concepts, and methods--were each divided into three parts. The six grades were divided into three groups: third-fourth, fifth-sixth, and seventh-eighth.

To guard against teaching-learning interaction, each of the three teachers was randomly assigned across grade levels, concepts and treatments so that each of the three teachers used each method with a grade level and taught each concept to a grade level.

The students were assigned randomly to treatments. So within the third-fourth grade group there was a randomized assignment of the students to text, oral or visual presentations. The same was true for fifth-sixth, and seventh-eighth grade groups.

To assess whether the random assignment resulted in groups that differed in past achievement, the CTBS which had been administered in 1971 was analyzed using analysis of variance across the three assignment groups to test for significance of difference in basic background understanding. There were no significant differences across the groups on any of the sub-scores or on the total CTBS score.

The criterion test of knowledge about central place cities, special function cities, and break-of-bulk cities was graded in terms of total score, scores as they relate to the three concepts, and scores as they relate to the three methods used to present the test questions. The pretest scores were analyzed by analysis of variance and again there were no significant differences across the student assignment groups.

The criterion test was designed to measure the performance of the Indian pupils on three concepts presented across three types of questions. A 21-item test yielded a total score and six sub-scores. The sub-scores were for the concepts: central place, break-of-bulk, and special function. The items were also scored for the method in which the questions were posed to the student, i.e., oral questions, written items and items requiring interpretation and/or production of pictures and geographic symbols.

The administration of the criterion test required 20 to 30 minutes. It was administered as the pretest one day prior to the experiment and again immediately after the third concept had been presented. Of the 103 pupils, 102 took the pretest and 96 were present for the post test. The short exposure to teaching methods, i.e., three sessions of 30 minutes, may have been insufficient to demonstrate significant changes.

The criterion tests were administered immediately after instruction was completed. Therefore, there was no attempt to assess differential retention across teaching methods.

The study was designed and conducted to test one basic hypothesis that grew out of the literature on the education, especially of Indian children.

The hypothesis was as follows: the students who receive instruction using the visual communications method will do significantly better than the students who receive instruction in the same concepts orally, or through reading a text. Further, the students who receive instruction orally will do significantly better than the students who have only the textual material available to them. To restate this, it is a directional hypothesis that says the three groups will differ in their measured achievement, with the visual group being highest, the oral group next highest, and the textual group the lowest. This hypothesis was tested by applying analysis of variance to criterion test scores. The resulting F-ratio was .685.

There is a temptation to talk about the results of the treatment even though there seem to be no significant differences across the groups. Researchers typically avoid the temptation. In this case we elect not to. In one respect, since the study used all of the students in grades three through eight, a total population, it was not a sampling process. Probability statistics legitimately could be avoided and a look at the means of the measured achievement to see if there are differences across the three groups is warranted.

The means of the results from the criterion tests are 8.5 for the group that received instruction by a text method, 9.4 for the group that received instruction by the visual method, and 9.6 for those who received the instruction orally. So even by dismissing probability statistics, and looking only at the mean criterion scores, the hypothesis is still not substantiated because the means are not in the expected direction.

Analysis of student achievement by the individual concepts across methods yields an F-ratio which was not sufficiently large to be significant. The F-ratio on the central place concept was .15, on the special function concept .24, and on the break-of-bulk concept it was 1.1. Again, yielding to the temptation to look at the mean scores even though there are no significant differences, we find that for the central place concept, the mean achievement was as hypothesized test lowest, oral second, visual highest. The same was true on the special function concept, but in the break-of-bulk concept, text was lowest, visual was second, and oral was highest.

Given the differential reading ability and comprehension of the students by grade, it was deemed desirable to look at the effects by treatment and grade level. Using the same basic hypothesis, the analysis of variance was computed for the resulting nine grade-method groups. When the criterion test scores are analyzed by treatment method and by age-grade level, there is a significant difference across the groups for total criterion test score. The F-ratio was 3.58 which is significant at the .05 level of confidence. Looking at the means across the methods within the grade groupings, the scores increase from the text to oral presentation to visual for the third-fourth grade. They increased from text to oral to visual for the fifth-sixth grade, but for the seventh-eighth grade, text is the lowest, with visual the second, and oral highest.

It would be nice to be able to say that the result of many weeks of preparation, three days of instruction, followed by many weeks of analysis, led to at least some trend which would permit us to tell you that visual and oral instruction had been demonstrated to work better with Indian children than textual instruction. We cannot say that with any degree of certitude.

Perhaps current research designs and statistical methods are somewhat inappropriate when comparing the effectiveness of different methods or modes of presentation (see Note 13). When human learners are involved, Snow and Solomon (see Note 19) contend that it is naive to assume that all learners are the same and that random assignment will assure homogeneity. It may be meaningless to use treatment averages when comparing instructional methods.

1. Coleman, J. S. Equality of Educational Opportunity. Washington, D. C.: U.S. Government Printing Office, 1966.

2. Comprehensive Tests of Basic Skills. California Test Bureau, McGraw-Hill Book Company, 1969.

3. Coombs, L. M. The Indian Child Goes to School: A Study of Inter-racial Differences. Washington, D.C.: Department of the Interior, Bureau of Indian Affairs, 1958.

4. Cundick, B. P. "Measures of Intelligence on Southwest Indian Students," Journal of Social Psychology. 1970, 81, pp. 151-156.

5. Harris, C. and Ullman, E. "The Nature of Cities," Annals of the American Academy of Political and Social Science. 1945, pp. 7-17.

6. Havighurst, R. J. "Education Among American Indians: Individual and Cultural Aspects," Annals of the American Academy of Political and Social Science. 1957, 311, pp. 105-115.

7. Havighurst, R. J., Gunter, M. K. and Pratt, I. E. "Environment and the Draw A-Man Test: The Performance of the Indian Children," Journal of Abnormal and Social Psychology. 1946, 41, pp. 50-65.

8. Havighurst, R. J. and Hilkevich, R. R. "The Intelligence of Indian Children as Measured by a Performance Scale," Journal of Abnormal and Social Psychology. 1944, 39, pp. 419-433.

9. Kagan, J. "Developmental Studies in Reflection and Analysis," in A. H. Kidd and J. L. Rivoire (Ed.) Perceptual Development in Children. New York: International University Press, Inc., 1966, pp. 487-522.

10. O. Lesser, G. S., Fifer, G. and Clark, D. H. "Mental Abilities of Children from Different Social Class and Cultural Groups," Monographs of the Society for Research in Child Development. 1965, Vol. 30, 4.

11. Levensky, K. "The Performance of American Indian Children on the Draw-A Man Test," National Study of American Indian Education. ERIC, March 1970, Series 111, No. 2.

12. Maginnis, G. A. "Fry’s Readability Graph: Extended Through Preprimer Level," The Reading Teacher. March 1969.

13. Messick, S. The Criterion Problem in the Evaluation of Instruction. Princeton, New Jersey: Educational Testing Service, 1969.

14. Peterson, S. A. How Well Are Indian Children Educated? Washington, D.C.: U.S. Department of the Interior, U.S. Indian Service, 1948.

15. Riessman, F. "The Strategy of Style," Teachers College Record. 1964, 65, pp. 484-498.

16. Shears, B. T. "Aptitude, Content, and Method of Teaching Word Recognition With Young American Indian Children." Unpublished Dissertation, University of Minnesota, 1970.

I7. Shumsky, A. "Individual Differences in Learning Styles," in L. Sperry (Ed.) Learning Performance and Individual Differences. Scott, Foresman and Co., 1972, pp. 122-125.

18. Snider, J. G. and Coladarci, A. "Intelligence Test Performance of Acculturated Indian Children," California Journal of Educational Research. 1960, Vol XI, No. 1, pp. 34-37.

19. Snow, R. and Solomon, G. "Aptitudes and Instructional Media," A. V. Communications Review. 1968, 16, pp. 341-357.

20. Tyler, L. E. The Psychology of Human Differences (Third Edition). New York: Appleton-Century-Crofts, 1965, pp. 297-302.