Volume 25 Number 1
October 1985

Graham Hurlburt, William Schulz and Lyle Eide

AS INCREASING numbers of American Indians attend high school, they need vocational guidance counselling to prepare them for post-secondary educational programs and career choices. Decker (1979) found that in Yavapai College, the strongest vocational and career interests were in automotive technology, engineering, welding, and business administration. Indian students appear to aspire to occupations largely limited to those visible on or near the reservation in which they lived. Spencer (1973) found higher status occupations included teacher, secretary, nurse, automobile mechanic, medical or dental technician, and member of the armed forces. In a career development program for tribal girls in Wisconsin, Finley (1977) found the counselling program was well received, and the girls changed attitudes, gain knowledge, and made plans for the future.

That career interest testing is important for Indian students may be evident, but are the currently used inventories appropriate? There is a substantial and growing number of studies which indicates that psychological tests normed on non-Indians may be unreliable when administered to Indian students.

Gade, Fuqua, and Hurlburt (1984) found six significant differences on the SDS scales between the Indian high school student sample and the non-Indian sample. In this study it was found that boys from Peguis, Manitoba Reserve scored significantly higher on the Social and Conventional Scale, whereas Swampy Cree boys scored significantly higher on the Realistic Scale. The Swampy Cree girls (from Northern Manitoba) scored significantly higher on the Realistic and Conventional Scales, whereas the Peguis girls were significantly higher on the Realistic scale than the non-Indian group.

Other studies indicate the existence of a difference in learning styles between Indian and non-Indian students which might influence their test taking abilities. Goodenough (1926) found that Indian children score exceedingly well on intelligence tests if they incorporate visual material, such as the Goodenough Draw-A-Man Test. Garber (1968) found Navajo and Pueblo first grade children adept at remembering visual symbols, manipulating pictures and design, and understanding relationships which involved visual associations much better than they could handle the subtests keyed to auditory skills. Some studies show Indian students have a tendency to right hemisphere learning which may not match successful strategies for answering standardized tests. Browne (1984) reported that American Indian children, age 6 to 16 years, had a pattern of WISC-R performance different from that of the standardization population. They had highest performance scores on subtests identified as "right brain". Browne suggested that this characteristically different pattern reflected relatively greater strength in relational, holistic, right hemisphere information processing, and she concluded that Indian students have a preference for right hemisphere information processing as opposed to the linear, sequential, processing of the left hemisphere. The left hemisphere style may be more appropriate for answering psychological tests.

Because there is some doubt about the appropriateness of testing Indian high school students with the Self-Directed Search, the present study examines testretest reliability of the SDS with a sample of Indian high school students at Teulon, Manitoba, Canada.

The sample consisted of Treaty Indian high school students from northern and central Manitoba, and northern Ontario, Canada, who resided in the Teulon Residence and attended the local white public high school. During the first administration of the SDS, 75 students answered the SDS, including 27 females and 48 males. Due to homesickness and loneliness for girl- or boy-friends left behind, a number of students had quit school and returned home by the time of the retesting three months later. At the time of the second testing, 52 students filled out the SDS, including 17 females and 35 males. English was the first language for only 20% of the sample. Cree was the first language for 60% and Saulteaux the first language for 20% of the rest of the sample. The students ranged in age from 15 to 22 years and attended grades ten to twelve in Teulon Collegiate where they composed one-seventh of the student enrollment.

The SDS, 1977 edition (Holland, 1979), was used to collect the data on vocational interests. The SDS is a self-administered, self-scored, and self-interpreted vocational counseling tool. A person obtains a three letter occupational code which is used to match up with suitable occupations. Scores are totaled for activities, competencies, occupations, and self-estimates. These scores are totaled and yield a summary code of six personality types; R-(Realistic) indicating a conforming, materialistic, practical, modest, stable person which matches with outdoor and blue-collar work: 1-(Investigative) indicating a precise, and rational person which matches with investigating and research; A-(Artisitic) indicating an emotional, imaginative, impulsive, independent and nonconforming person which matches with artistic creation and freedom; S(Social) indicating a cooperative, friendly, helpful, kind, and understanding person which matches with teaching and counseling; E-(Enterprising) indicating an attention-getting, ambitious, self-confident and popular person which matches with business and political activities; and C-(Conventional) indicating a conforming, conservative, inhibited, obedient, self-controlled and efficient person which matches with clerical and office procedures.

Correlational patterns among the six scales of the SDS form a crude hexagonal agreement of vocational interests (Cole & Hanson, 1971). Separate norms for 2,384 high school boys and 2,577 high school girls reported in the manual were used for comparison. Corrected split-half (odd-even) reliability coefficients for the six summary scales range from .83 to .95, and retest reliability (time interval 1 to 4 weeks) ranged from .31 to .87. Predictive validity over a 3-year period with a college sample yielded Kappa coefficients of .29 for men and .30 for women.

Permission from the Director of the Teulon Residence for Native Indian students was obtained to administer the SDS during the scheduled evening study period. The SDS was administered separately to the males and females, and one staff member was available for each group to assist the senior author. The SDS was scored by each student and he or she looked up their code in the Canadian Edition of the Occupations Finder (Bedal and Weeks, 1979).

Three months later in December, 1984, the senior author administered the SDS again to the sample of Indian high school students at the Teulon Residence. With a ten minute explanation of terms, retesting with the SDS took place over three evenings with small groups of males and females. A staff member assisted the senior author and the SDS was again scored by each student, who again looked up his or her code in the Canagian Edition of the Occupations Finder.

TABLE 1
Agreement Between SDS Codes in Test and Retest of
Indian High School Students in Teulon, Manitoba

In analyzing the degree of agreement between the three-letter codes of SDS in pre- and post-testing of the Indian Students, the following indexing system was used:

In Table 2 there were significant differences between male and female Indian student first letter codes, particularly in S and R. Sixty-five percent of females scored S for their first letter code compared to 0 percent for females, similar to Holland's male and female norms.

TABLE 2
Differences in Indian Male and Female First Letter Codes

Although no significant differences were found between the first letter codes of the Indian High School male sample and Holland's male norms, there is evidence that Indian males tend to score more in the direction of S and less in R than the high school male norms. These results are presented in Table 4. Twenty-nine percent of Indian male high school students scored R as their first letter code compared to 40 percent for the male norms, and 34 percent of the Indian males chose S for their first letter code compared with 20 percent for the male norms. Because of small sample size, these differences did not reach statistical significance.

In Table 5, the results show that there were no significant differences between test retest of first letter codes with Indian high school students, but a direction of inconsistency was indicated in the A and I code of the male Indian high school students who scored twice as many A's in retesting, and less than a third as many I's.

In Table 6 the results show that with the exception of I for males, the retest reliability coefficients of the SDS Summary Scales and Codes indicate some measure of consistency over a three month period.

TABLE 3
Differences Between Indian High School Female Sample
and Female Norms First Letter Codes

TABLE 4
Differences Between Indian High School Male Sample
and Male Norms First Letter Codes

It would appear that Holland's Self-Directed Search may be a somewhat reliable vocational interest inventory to use with Indian high school students. There is significant stability on all scales except for Social for the Indian females, and the other reliability coefficients are in the moderate to high range. The median retest reliability of SDS Summary Scales and Codes of the Indian male high school students over three months was .72 and for Indian females was .61 compared to .61 for high school male norms (Holland, 1979) and .64 for high school female norms over a 3-4 week period. The retest reliability coefficient of the SDS with the present sample of Indian students indicates some measure of stability of vocational choice or attitude over a three month period and suggests the SDS is probably a reliable and useful measure of Indian students' vocational interests, inchoate though they may be. No statistical evidence was found that Indian male and female high school students' first-letter codes differs from Holland's norms, nor did Indian males and females switch first code letters between test and retest administration of the SDS more than Holland's norms.

TABLE 5
Differences in First Letter Codes Between Pre- and Post-Testing
of the SDS with Indian High School Students

TABLE 6
Retest Reliability Coefficents of SDS Summary Sacles and Codes, N = 52
Teulon Indian High School Students

Scott and Anagon (1980) also found that vocational interest inventories may be used reliably across cultures. They reported that the ACT Interest Inventory results from American Indian college bound students and Caucasian college bound students were similar for both ethnic groups, and concluded that across groups common interpretations were appropriate. In the present study, the absence of English as the first language of 80% of the Indian high school students did not appear to influence the retest reliability of the SDS.

Holland's Self-Directed Search appears to be a reliable vocational interest inventory when used with American Indian high school students. Although no major differences were detected between the Indian sample scores and those of Holland's norms, smaller differences may have gone undetected because of the size (N = 52) of the present sample and the subsequent lower power of the statistical testing used. It is suggested that further research with a larger sample of Indian high school students using the SDS in test-retest administrations may find significant differences between Indian students and Holland's norm which in the current study appear only as directions.

Bedal, C.L., & Weeks, A. (1979). The Occupational Finder. Guidance Centre, Faculty of Education, University of Toronto.

Browne, Dauna Bell. (1984). WISC-R Scoring Patterns Among Native Americans of the Northern Plains. White Cloud Journal, Vol. 3, No. 2, p. 3-16.

Cole, N.S., & Hanson, G.R. (1975). The use and evaluation of interest inventories and simulations. In E. E.Diamond (Ed.), Issues of sex bias and sex fairness in career interest measurement. Washington, D.C.: National Institute of Education.

Decker, Don. (1979). Indian Outreach Program Needs Assessment Survey.

Yavapai College, Prescott, Arizona. (ERIC Document Reproduction Service No. ED 176919).

Finley, Cathaleen. (1977). Choices and Careers: Free to Choose: A Report.

Cooperative Extension Programs, University of Wisconsin-Extension, Madison, Wisconsin. (ERIC Document Reproduction Service No. ED 152447.)

Gade, Eldon, Hurlburt, Graham, and Fuqua, Dale. (1984). Use of the Self-Directed Search with Native American High School Students. Journal of Counseling Psychology, V. 31, No. 4, P. 584-587.

Garber, M. (1968). Ethnicity and Measures of Educability. Unpublished doctoral dissertation, University of Southern California.

Goodenough, F.L. (1921). Racial Differences in the Intelligence of School Children. Journal of Experimental Psychology, 9. p. 388-395.

Graham, Dorothy Marguerite. (1972). A Comparison between Indian and non-Indian Children in Southern Saskatchewan Based on Listening Comprehension, Reading Comprehension, Auditory Comprehension, and I.Q. Unpublished Ph.D. dissertation, University of Northern Colorado.

Holland, J.L. Professional Manual, 1979 Edition, The Self-Directed Search. Palo Alto: Consulting Psychological Press.

Scott, Thomas B., & Anadon, Max. (1980). A Comparison of the Vocational Interest Profiles of Native American and Caucasian College Bound Students, Measurement and Evaluation in Guidance, Vol. 13, No. 1, April.

Sealy, Bruce D., & Riffel, Anthony J. (1984). Easterville Education Review, Winnipeg, January.

Spencer, Barbara. (1973). Occupational Orientations of Choctaw Indian High School Students in Mississippi, Master’s Thesis, Mississippi State University. (ERIC Document Reproduction Service No. ED 154955.)

Graham Hurlburt is a Counselor at the University of Winnipeg, Winnipeg, Manitoba R3B 2E9. He received his Ph.D. from the University of North Dakota. Dr. Hurlburt can be reached by writing him at the University of Winnipeg Student Counselling Services.

Bill Schulz is the Head of the Department of Educational Psychology at the University of Manitoba, Winnipeg, Manitoba, Canada R3T 2N2. Dr. Schulz received his Ph.D. from the University of Wyoming. Lyle Eide is currently Director of the Counselling Service at the University of Manitoba. He received his Ph.D. from the University of North Dakota.