Journal of American Indian Education

Volume 27 Number 1
October 1987

ANALYSIS OF NAVAJO ADOLESCENTSí PERFORMANCES ON THE RAVEN PROGRESSIVE MATRICES

Craig Sidles, James MacAvoy, Carolyn Bernston, Anne Kuhn

This study was conducted under the auspices of the Northern Arizona University Native American Research and Training Center, funded by the National Institute on Disability and Rehabilitation Research (NIDRR), Office of Special Education and Rehabilitative Services, Department of Education, Washington, DC (Grant No. G0083CO095), and the Northern Arizona University Organized Research Program.

One hundred eighty-three adolescent Navajo students from schools in Arizona and New Mexico were administered the Raven Standard Progressive Matrices (RSPM). These regular classroom students from BIA and public schools ranged in age from 13 to 15 years. Raven norms were established for this population. In addition, comparisons in RSPM performance were made between students having different primary language (Navajo vs. English) and school location (on reservation vs. off reservation). No significant differences were found between these groups suggesting that neither language nor geographical factors play a role in the RSPM performance of Navajo students. These findings would also support the contention that the Raven Standard Progressive Matrices is a "culture fair" measure of non-verbal mental processing with secondary level Navajo students.

One current area of research for the Native American Research and Training Center at Northern Arizona University addresses the question: What assessment instruments are available for measuring the intelligence of Native American adolescents which utilize a "culture fair approach"? Studies involving the intellectual assessment of Native American students have illustrated the effect of testing such populations in their secondary language using verbal scores. Krywaniuk and Das (1976) found that Navajo children earned depressed verbal scores while obtaining significantly higher Performance scale scores on the WISC-R. Similarly, Zarske and Moore (1982) found average WISC-R Verbal-Performance scale discrepancies of over 25 points in their study involving regular classroom, learning disabled, and educationally disadvantaged Navajo students.

Sattler (1982) points out that language-related factors often appear to confound attempts to accurately measure the intelligence of persons from various cultures. These same language factors likely contribute to academic deficits exhibited by Indian students, especially those attending inner reservation schools where the least acculturated pupils can be found (Coombs et al., 1958; Edington, 1969). In view of this research, the authors believe that tests should be used which minimize language and cultural factors.

Recognizing that identification of possible culture-fair tests of intelligence and/or mental processing has been going on for some time, the present research team looked at existing instruments which might be utilized. Those most commonly mentioned in the literature, according to Martin (1986), including The Leiter International Performance Scale, the Draw-A Person, Cattellís Culture-Fair Intelligence Test and the Raven Standard Progressive Matrices (RSPM). Of these, the most widely used instrument has been the Raven Standard Progressive Matrices (RSPM) which has been in existence for 50 years (Martin 1986). These measures share in common a non-verbal testing format which has been recommended for use when attempting to obtain a least biased assessment of intelligence in American Indian populations (West & McArthur, 1964; Wiltshire & Gray, 1967; Hartlage, Lucas, & Goodwin, 1976; Cattell, 1979; Sattler, 1982).

The present researchers reasoned that the visually oriented RSPM would be especially relevant for research with Navajo Indian populations because of their strong visual reasoning capabilities (Krywaniuk & Das, 1976) and their well known expertise in jewelry making, rug weaving, painting, and pottery making. The Raven also had the advantage of being non-timed (an important consideration for the Navajo culture) and available for use as a group administered instrument.

Purpose of the Study

This study was conducted in order to examine the performance of adolescent Navajo students on the RSPM. First, the studentsí RSPM scores were compared on the basis of their primary language (Navajo vs. English) and school geographical location (on reservation vs. off reservation). By analyzing the effects of these intragroup variables on test performance, it was felt that conclusions regarding the cultural fairness of the RSPM could be made with respect to this population. Secondly, norms were developed for this population with descriptive data relating to the following variables: chronological age, primary language, sex, and school geographical location.

Procedure

Three Bureau of Indian Affairs (BIA) schools located on the Navajo Reservation provided the 124 "on reservation" sample subjects for this study. These schools are administratively represented by the Ft. Defiance Agency, and represent a geographical cross section of the reservation. Permission for testing was obtained from administrators at the local school level. The subjects ranged in age from 13 to 15 years of age. Whereas male and female students were equally represented, approximately one third of the subjects were primary English speakers and two-thirds identified Navajo as their primary language. Students with prior grade retention or special education placement were excluded from this sample.

The "off reservation" adolescent Navajo sample was selected from public and BIA schools in the Farmington, New Mexico area. These 59 students also ranged in age from 13 to 15 years of age. Similar permission procedures were obtained and special education students were again excluded from the sample.

For on-reservation students, a primary language questionnaire was developed to ascertain the language performance of each subject. It was a goal of the team to determine whether primary language as a variable contributed to variance in the subjects' RSPM scores. If the studentsí scores did not vary significantly as a result of different primary language status (Navajo vs. English), this finding would lend support to the cultural fairness of the RSPM. The questionnaire developed by the team included five items addressing the first language of the child as well as the language used at home, school, and with friends. The items, which were answered by the subjects, were read to them in Navajo and English to ensure their comprehension and proper response. The language endorsed by the child on three or more items was accepted as his or her primary language.

The RSPM was administered to small groups of subjects using standardized procedures outlined in the test manual. An untimed testing format was used. Data related to individual test scores and categorical variables (age, sex, primary language, geographical location) were encoded into a computer file. Descriptive and inferential analyses were calculated by means of the Statistical Package for the Social Science (SPSS). Statistical significance was set at the p = . 01 level.

Results and Discussion

The initial analyses addressed the effects of primary language and geographic location upon the subjects RSPM performance. A t-test for independent samples was computed to compare the performance of Primary Navajo speaking students (x = 39.51) with that of the Primary English speaking subjects (x = 39.86). There was no significant difference (t (113) = 2.63, p>.01) found between the means for these two groups suggesting that primary language does not play a prominent role in the RSPM performance of adolescent Navajo students. This finding is not surprising in view of the non-verbal/visual reasoning format utilized in the RSPM.

A t-test analysis was also computed to examine the effect of geographical location of school attended upon the RSPM performance of the Navajo subjects. Specifically, this analysis involved comparing the mean RSPM scores of students attending schools on the Navajo Reservation (x = 39.86) and those enrolled in schools off the reservation (x = 38.49). The difference in mean scores between these two groups was found not to be significant (t (181) = 2.30, p>.01). This finding would suggest that the RSPM is appropriate for use with Navajo students attending school on or off the reservation.

In summary, the RSPM appears to be valid for use with the majority of adolescent Navajo students regardless of their primary language status or the location of their present school. The fact that the RSPM scores of these students remained relatively constant in spite of intragroup differences would lend support to the cultural fairness of this instrument.

Normative Data for the RSPM

A second objective of this study was to establish current RSPM norms for adolescent Navajo students. In Table 1, scores are listed which reflect the mean and standard deviations obtained for the Ft. Defiance school population.

 

TABLE 1
Mean and Standard Deviation Scores for Students
from Ft. Defiance Schools (on reservation)

 

Chronological Age

Sex

 

13

14

15

Male

Female

Mean

37.56

39.84

41.00

40.00

39.90

SD

3.43

6.35

5.50

5.59

6.43

Total Mean: 39.86; Total N: 124; Total So: 6.08

Table 2 provides percentile ranks corresponding to a range of RSPM scores obtained by Ft. Defiance students.

Listed in Table 3 are descriptive statistics reflecting the performance of the Farmington School sample on the RSPM.

Table 4 contains the percentile rankings for the RSPM performance of Farmington students.

The means and standard deviation scores listed below represent parameters for the entire subject population. The authors recommend that practitioners refer to the following descriptive data when needing to estimate the performance of Navajo adolescents on the RSPM in the future.

Percentile ranks corresponding to the combined RSPM performance for the Ft. Defiance and Farmington schools are listed below in Table 6.

Implications

The norms of these Navajo regular classroom 13, 14 and 15 year olds may prove helpful to psychometricians and educators working with this population. The finding that primary language (Navajo or English) and school location did not influence Raven scores speaks to the applicability of the Raven Progressive Matrices in cross-cultural settings for special education and gifted assessment; the test is "culture-fair" to the extent that primary language and geographical location (on-reservation vs. off-reservation) did not affect test scores in this study.

TABLE 2
Percentile Ranks Corresponding to Raven Raw Scores
for Ft. Defiance Schools (N = 124)

Raven Raw

Score

Cumulative Percentile

Ranking

55

100.0

52

99.2

51

98.4

50

97.6

49

96.0

48

93.5

47

91.9

46

86.3

45

82.3

44

75.0

43

71.0

42

63.7

41

54.0

40

50.8

39

44.4

38

40.3

37

36.3

36

31.5

35

25.8

34

20.2

33

12.9

32

11.3

31

8.9

30

8.1

29

5.6

28

4.0

26

3.2

TABLE 3
Mean and Standard Deviation Scores for Students
from Farmington Schools (off reservation)

 

Chronological Age

Sex

 

13

14

15

Male

Female

Mean

32.50

39.66

37.42

38.12

39.16

SD

1.73

7.79

7.61

8.56

6.92

Total Mean: 38.49; Total N: 59; Total So: 7.65

Future research might concentrate on a comparison of the present Raven findings with a comparable group of Mexican-American, Anglo, Asian and other cross-cultural adolescent groups. This would enable researchers to know more about what the RSPM does measure, and its possible cross-culture applicability as a culturally sensitive measure of intelligence.

TABLE 4

Percentile Ranks Corresponding to Raven Raw Scores
for Farmington Schools (N = 59)

Raven Raw

Score

Cumulative Percentile

Ranking

54

100.0

51

96.6

50

94.9

49

89.8

48

86.4

46

84.7

45

81.4

44

79.7

43

76.3

42

69.5

41

64.4

39

55.9

38

54.2

37

50.8

36

44.1

35

35.6

34

27.1

32

25.4

31

18.6

30

10.2

28

8.5

27

6.8

26

5.1

25

3.4

17

1.7

 

 

TABLE 5
Mean and Standard Deviation Scores for Combined Student
Populations for Ft. Defiance and Farmington Schools

 

Chronological Age

Sex

 

13

14

15

Male

Female

Mean

36.00

39.78

39.95

39.42

39.64

SD

3.81

6.84

6.31

6.64

6.57

Total Mean: 39.42; Total N: 183; Total So: 6.64

 

TABLE 6
Percentile Ranks Corresponding to Raven Raw Scores
for Ft. Defiance Schools (N = 183)

Raven Raw

Score

Cumulative Percentile

Ranking

55

100.0

54

99.5

52

98.4

51

97.8

50

96.7

49

94.0

48

91.3

47

89.6

46

85.8

45

82.0

44

76.5

43

72.7

42

65.6

41

57.4

40

52.5

39

48.1

38

44.8

37

41.0

36

35.5

35

29.0

34

22.4

33

16.9

32

15.8

31

12.0

30

8.7

29

6.6

28

5.5

27

4.4

26

3.8

25

1.1

17

.5

 

 

 

REFERENCES

Cattell, R. B. (1979). Are culture fair intelligence tests possible and necessary? Journal of Research and Development in Education, 12, 3-13.

Coombs, L. M., et al. (1958). The Indian Child Goes to School. Washington, DC: U. S. Department of Interior, Bureau of Indian Affairs.

Edington, E. D. (1968). Review of recent research on American Indian Student: Academic Achievement. Journal of American Indian Education, 8, 10-15.

Hartlage, L.C., Lucas, T.L., & Goodwin, A. (1976). Culturally biased and culture-fair test correlated with school performance in culturally disadvantaged children. Journal of Clinical Psychology, 32, 658-660.

Krywniuk, L.W. & Das, J.P. (1976). Cognitive strategies in native children: Analysis and intervention. Alberta Journal of Educational Research, 22, 271-280.

Martin, W.E., & Sidles, C.W. (1986). Presentation at Council for Exceptional Children Symposia on Ethnic and Multicultural Concerns. Dallas, Texas, USA. November, 1986.

Raven, J. (1983). Manual for Ravenís Progressive Matrices and Vocational Scales, Section 3, 1983 edition, p. 2 1. The Psychological Corporation, 757 Third Avenue, New York, NY.

Reschly, D.J. (1982). Assessing mild mental retardation: The influence of adaptive behavior, sociocultural status and prospects for non-biased assessment. In C.R. Reynolds & T.G. Butkin (Eds.), The Handbook of School Psychology. New York: John Wiley & Sons.

Sattler, J.M. (1982). Assessment of childrenís intelligence and special abilities. (2nd edition). Boston: Allyn and Bacon, Inc.

West, L.W., & MacArthur, R.S. (1964). An evaluation of selected intelligence tests for two samples of Metro and Indian children. Alberta Journal of Educational Research, 10, 17-27.

Wiltshire, E.B., & Gray, J.E. (1969). Draw-A-Man and Raven Progressive Matrices (1938) intelligence test performance of reservation Indian children. Canadian Journal of Behavioral Sciences, 1, 119-122.

Zarske, J.A., & Moore, C. (1982). Recategorized WISC-R scores for non-handicapped, learning disabled, educational disadvantaged and regular classroom Navajo children. School Psychology Review, 11, 319-323.

Craig Sidles, Ph.D., is a Professor of Educational Psychology in the Center for Excellence in Education at Northern Arizona University. He received his doctorate in the Division of Counselor Education at the University of Iowa in 1968. He is a research associate with the Native American Research and Training Center at Northern Arizona University, and presently is involved in a National Institute on Disability and Rehabilitation Research study of culturally sensitive measures of intelligence. He is also a member of the Arizona State Department of Education Task Force for selection of statewide pupil progress measures.

James MacAvoy was a doctoral intern in Educational Psychology in the Center for Excellence in Education at Northern Arizona University at the time of the study. He has since received his doctoral degree, and is a School Psychologist with the Flagstaff, Arizona Public School District. Dr. MacAvoy is a former staff member of the Institute for Human Development at Northern Arizona University. His dissertation study examined the relationship of primary and secondary language presentation and Navajo childrenís performance in multi-trial auditory memory assessment.

Carolyn Bernston is a doctoral student in Educational Psychology in the Center for Excellence in Education, and at the time of the study, held a Doctoral Assistantship with the Native American Research and Training Center at Northern Arizona University. She has previously taught special education classes in Alaska.

Anne Kuhn is an Educational Diagnostician with the Farmington, New Mexico Public Schools. She is involved in current norming exercises with the Raven Progressive Matrices in collaboration with Dr. John Raven of the Scottish Council for Research in Education.

 

 
 
[    home       |       volumes       |       editor      |       submit      |       subscribe      |       search     ]