Louis C. Bernardoni

Guidance Specialist, Division of Indian Education

Arizona State Department of Public Instruction

Attempts to test the intelligence of Indian children are usually frustrating in that linguistic and cultural differences contribute to low scores on most standardized tests of intelligence. Havighurst (see Note 1) in a review of the literature identified three phases in testing the intelligence of Indian children. Pioneers in the field first used verbally loaded tests and concluded that Indian children were less intelligent than other children. Later, various types of performance and non-verbal tests were administered revealing a wide variation between the verbal and non-verbal abilities of Indian children with the non-verbal test results approaching nearer to the norm of non-Indian children. More recent studies with the Goodenough Draw-A-Man test and the Grace Arthur Performance Test revealed that Indian children of various groups score near the mean achieved by non-Indian children with some groups scoring above national norms.

These recent studies have resulted in most educators accepting the theory that Indian children are as intelligent as other children. Most schools do not have personnel trained in the administration and scoring of tests such as the Grace Arthur Performance Test and the Goodenough Draw-A-Man Test, however, and are still faced with the problem of assessing intelligence with tests that would be less time consuming than these tests. Several schools on the Navaho and Apache reservations have used Cattell's Culture Free Intelligence Test which utilizes geometric drawings. The unpublished results of these studies reveal that Navaho and Apache students score very low on this test.

Norman and Midkiff (see Note 2) administered Raven's Progressive Matrices and Goodenough's Draw-A-Man tests to 96 beginning and primary Navaho students at the U. S. Indian School at Albuquerque, New Mexico. Below are the results of their testing:

An inspection of these results would indicate that these Navaho students performed at a much lower level on the Progressive Matrices than on the Draw-A-Man test. The Progressive Matrices tests the ability to identify patterns which are consistent with the theme of thought developed by the stimulus pattern.

The recent publication of the TOGA (S.R.A. Tests of General Ability) was viewed with interest by schools having large Indian populations. Since the test is divided into two parts, one cultural and one non-cultural, and can be given in any language, it appeared to be a test that would be appropriate for Indian children. The problem of this investigation was to determine the appropriateness of the TOGA with first grade Indian children by determining if:

As it was felt that first grade children are the most difficult to test in a group situation, the groups were limited to first grade students to subject the TOGA to a difficult task. The groups selected were first grade students at Poston and Kayenta, Arizona, the exhibit varying degrees of cultural deprivation. Below is a description of these three groups:

Group I. These children live in the Poston community which is about thirteen miles from Parker, Arizona. A paved road to Parker provides them with ample opportunity to visit this town and other towns in this area. Most of these families live in frame or masonry homes and most of the parents either farm or work in the Parker area. All of the children in this group spoke English before entering school. The children were mostly Mohaves and Chemehuevis with a few Hopis, Navaho and other tribes represented. This group was considered as the least culturally deprived group.

Group II. These children live in the town of Kayenta which is one of the most isolated areas of Arizona. Until recently, there were no paved roads into Kayenta and the nearest city of any size is about 150 miles away. Most of the children have had little opportunity to visit other areas. Some of the parents are employed by governmental or private industries in Kaventa. The homes are either small frame homes or hogans. All of the children in this group spoke both Navaho and English. These children were rated on a ten point scale as to ability to speak English and averaged a rating of 6.1. The rating was done by the coordinator who speaks both Navaho and English and knows the children well. On this scale, a rating of 1 indicated little or no knowledge of English and ratings of 10 indicated the ability to understand English as well as the "average" non-Indian child. This group while culturally deprived, did understand English.

Group III: This group lives in the country outside of Kayenta and are transported to school by bus. These children speak Navaho and are in the process of learning English. All of these families live in hogans and obtain a subsistence living by raising sheep and periodically obtaining wage work in Kayenta or other areas distant from their home. When rated by the coordinator on ability to speak English, this group obtained an average rating of 3.0. This group was considered to be very culturally deprived.

To offset any effect that school attendance might have on the results, the tests were administered to all three groups during the week starting October 17, 1960. Group I was given to TOGA in English by the school counselor assisted by the classroom teacher. Group 11 was given the TOGA in English by the classroom teacher assisted by the principal. Group III was given the TOGA in Navaho by the coordinator assisted by the classroom teacher.

To minimize any bias that may have been injected due to the TOGA being administered by different persons, each person studied the directions carefully and followed standardized procedures in administration. The greatest difficulty encountered was in the translation of the directions into the Navaho language.

As the Navaho language does not traditionally contain words for all those used in the TOGA, it was necessary to use the Navaho translation which often was a descriptive phrase. For example, the Navaho translation for awning is "shade that is used over doors, windows or porches." Such a translation gave away the answer to the question. On the other hand, there is no Navaho word for violin so the translator used the word "instrument." This made the question too ambiguous to be answered logically. Similarly, there are no Navaho words for maple or magnet or many phrases used. In general, it appeared that the descriptive nature of the Navaho translation tended to help the students determine the answer to many questions.

In computing I.Q. scores, it was discovered that 29 percent of the students scored below the norms given in the booklets. Rather than extrapolating or computing actual I.Q.'s, these students were arbitrarily given an I.Q. score of 60, the lowest l.Q. listed in the table of norms. This would tend to make the results spuriously high.

Group	Location	Directions	N	Age	Mean Total IG	Total I.Q.	Verbal G.E.	Reasoning G.E.
Group I	Poston	English	25	6-4	.028	69.84	.14	.06
Group II	Kayenta	English	18	6-9	.005	69.72	.03	.31
Group III	Kayenta	Navajo	19	7-0	.226	74.74	.51	.17

Although the standardization of the test does not provide for I.Q. scores for the Verbal and Reasoning sections in isolation, Verbal and Reasoning I.Q. scores were obtained by utilizing Table #3 in the Manual. This was done by equating the Part Score norms with the Total Score norms. Following are the results of this procedure:

Although the above procedure may violate the standardization of the test, it does give a rough picture concerning the relative contribution of the two sections of the test to the total I.Q. score.

The TOGA is described as a test of general intelligence that does not require any form of school achievement, Part I of the test (Verbal) measuring the child's general knowledge of his surroundings, the American culture and Part 11 measuring powers of abstract reasoning that present an equal challenge to all children regardless of their cultural background.

Since the three groups chosen for this experiment were all culturally impoverished to some degree, one would predict that they would score below average on the verbal section (Part I) and score close to 100 on the reasoning section (Part 11) since it is purported to be culturally fair. The results obtained reveal that two groups scored higher on the Verbal section than on the Reasoning section. This would raise grave doubt that the Reasoning section does provide an equal challenge regardless of cultural background. There is no research evidence available that indicated that all cultures prepare its children equally well to detect differences in geometric drawings. The test results obtained would be an indication that these groups are as handicapped in detecting differences in geometric drawings as they are in knowledge of American culture. Perhaps the low score on Part 11 can be attributed to lack of ability to work independently, however. More research would need to be done to discover the reason for the low scores on Part 11.

The catalog advertising the tests indicate that the tests can be given to children from foreign speaking backgrounds simply by translating the examiners directions into the appropriate language. Since Group III was the most culturally impoverished, they should have scored lowest on Part 1, yet they achieved the highest score. As indicated before, the Navaho language is different in structure' and becomes descriptive when used in describing concepts in English. This is undoubtedly the reason for the high score for this group but raises grave doubts that the directions can be given in another language without re-standardizing the test. Perhaps the statement should be modified to indicate that the directions may be translated to languages that are similar in structure and content. Evidence should be included in the technical report that would indicate that the test may be translated without loss of validity.

Although the three Indian groups were chosen so that it included one of the best acculturated groups as well as the least acculturated groups in Arizona, the results of the TOGA indicated little difference. To test the significance of the difference, F tests were computed for the three groups. None of the F tests were significant at the 5 per cent level of confidence. If these groups are as different in degrees of acculturation as their outward appearance indicates, it would appear that the TOGA does not pick up this difference in acculturation.

As Groups II and III were rated as to ability in English, these ratings were correlated with the I.Q. and the derived verbal and reasoning I.Q. with the following results:

These results would indicate that there is a positive correlation between ability to speak English and results on the TOGA with the exception of Group II where there is no correlation between ratings in ability to speak English and the Reasoning I.Q. (derived). Perhaps. the test is succeeding in giving some students who are verbally handicapped the opportunity to perform at a higher level.

On the basis of this experiment, several recommendations and areas for further study are discussed. The experimenter feels that a great deal of further experimentation is needed before the TOGA could be recommended for first grade Indian children.

The efforts of the personnel of the Parker and Kayenta elementary schools and Mr. Vern Watkins of Science Research Associates in this study are gratefully acknowledged.