Volume 2 Number 2
May 1962

Herbert D. Peters

The measurement of intelligence has made rapid progress since the advent of Binet'soriginal scale. Tests have been both refined and multiplied, yet the difficulty of interpretation of the results is still with us. The task of interpretation is magnified whenever a non-representative group, such as Indian children, is being tested.

Previous studies of the Hopi Indian child have produced unrealistic, or at least uninterpretable, results. Dennis', Havighurst, Gunther, and Pratt2, and Havighurst and Hilkevitchs studied Hopi children at First Mesa and Oraibi. Since the present study is limited to Third Mesa, only the Oraibi results will be discussed. The two Havighurst studies are both based on the 1942-43 testing program and therefore may be considered as one. The results of the Dennis and Havighurst studies are summarized in Table 1.

Table 1

Summary of Three Studies of the Performance of

Hopi Children on Two Intelligence Tests

*Dennis' data includes both First Mesa and Oraibi Hopi.

**Only five tests from the Arthur Performance Scale were used.

Using the Goodenough Draw-A-Man Test, Dennis reported mean I.Q.'s of 116.6 and 99.5 for boys and girls, respectively, with a mean of 108.3 for the total sample of 152 First Mesa and Oraibi children. Havighurst, Gunther, and Pratt' reported means as high as 110.5 for 46 6-11 year old Oraibi children on the Draw-A-Man Test and a mean of 112.2 on the Arthur Performance Scale. Since the Dennis and Havighurst samples overlapped, it was possible to compare their results directly. Havighurst, Gunther and Pratt2 report a correlation of .52 on the Draw-A-Man Test for 64 cases. Dennis reported that the girls' scores decreased with age while the boys' scores increased. Havighurst, et al.2 , however, report that boys' scores remained constant but that the girls' scores decreased. The latter reported a correlation of .21 between the Draw-A-Man Test and the Arthur Scale.

While the results from the three studies are interesting, their interpretation is most difficult and their usefulness questionable. Two plausible explanations for the results obtained on the Draw-A-Man Test given by Havighurst, et al.2 are that (1) the Bureau of Indian Affairs was encouraging Indian art work in the schools, and (2) there is a definite cultural bias or training of boys in artistic expression in the Hopi community. Both of these explanations are well attested.

A different factor, test manipulation, must be taken into account with respect to the administration of the Arthur Performance Scale. This scale was limited to the five scales which seemed to have the least cultural bias. Havighurst and Hilkevitch defend the deletion of the Healy Completion Test because, "this test proved decidedly unsatisfactory for use with our subjects."' This procedure merely substitutes one kind of cultural bias for another.

The Problem. The study was designed to test the feasibility of assessing the intelligence of Hopi children with four available instruments.

Subjects. The subjects were the 59 Hopi children attending the Hopi Mission School at Oraibi. They came from the four Third Mesa villages of Oraibi, Old Oraibi, Bacabi, and Hotevilla. Their ages ranged from 5-15 years.

Tests.

All of the tests were administered by the writer who was personally acquainted with all the children. The testing procedures described in the test manuals were strictly followed.

Findings. Table II presents the mean results for each test and sub-test for each age level from 5 to 11 and 13-15 combined and the mean and standard deviation for the entire school. The highest and lowest I.Q. scores on each test are also reported. Figure 1 is a graphic presentation of the data found in Table 11 after smoothing.

Table II

Performance of Hopi Children on Four Intelligence Tests

Figure 1. Performance by fifty-nine Hopi children on four intelligence tests. The data presented in Table II were smoothed.

With the exception of the CTMM, performance on non-language scales compared favorably with the test norms. The school mean on the CTMM Language was slightly higher than the CTMM-Non-language mean. The results from the CTMM may intimate that the sub-titles are misnomers. The results are certainly at variance with those obtained on the Otis Alpha and the WISC. Performance of any task which required language skills was severely depressed. The correlations between the four tests were all higher than .60 and significant at the .01 level.

These findings appear to be much more in line with the actual school performance of these children. Performance seems to reach its peak in Grades III and IV after which learning appears to be more difficult.

In contrast to the findings of Havighurst, et al. and Dennis, there appears to be a definite curvilinear relationship between age and I. Q. scores. The 7 to 9 year old children obtain higher scores on the four tests and their respective sub-tests than the younger and older children. Depending upon the test used, peak performance was reached at the 7, 8, or 9 year level.

A single explanation may be an oversimplification but an obvious one seems to be available. The Hopi have become partly acculturated, especially in the most general aspects of the American culture. The younger children are quick in learning these most obvious elements of the culture in school, movies, radio, and through occasional visits to the city. However, for the older child, the broad cultural experiences with their varied nuances are not available. Futhermore, the increasing Hopi cultural emphasis denies the child the same opportunities available to the children in the general population. This should not be construed to be a value judgment but merely affirms the fact that each person acquires his own unique culture.

The results indicate that it is possible to obtain a fair assessment of the Hopi child's intelligence with the Otis Alpha non-verbal and the WISC Performance scales. The other tests are heavily weighed with verbal and cultural content which handicap the Hopi child. However, these verbal scores may be a fair indicator of the child's probable achievement in academic subjects.

While the results cannot be compared directly with general norms, it may be feasible to prepare local norms for at least one of the tests. The Otis would appear to have good prospects. These new norms would be very useful for intra-reservation comparisons while the general norms could be used for prediction of success at non-reservation schools. The general discrepancy between the findings in this study and the general norms supplied by the test publishers may be a fair indication of the degree of acculturation having taken place.