Michael B. Salzman

Non-Indians, usually Anglo-Americans, have historically had a dominant influence on the educational policies affecting American Indians. An 1880 report by the presidentially appointed "Board of Indian Commissioners," a supervisory group, stated "As we must have him (the Indian) among us, self-interest, humanity and Christianity require that we should accept the situation and go resolutely at work to make him a safe and useful factor in our body politic" (Prucha, 1973, p. 19 1). Captain Richard H. Pratt, soldier turned educator and the superintendent of the Carlisle Indian School stated "All the Indian there is in the race should be dead. Kill the Indian in him and save the man" (Prucha, 1973, p. 191). Roessel (1979) characterized education on the Navajo Reservation as: ". . . the weapon used by non-Navajos to teach Navajo young people to become Anglos--to reject their own heritage and culture and accept the identity and culture of the dominant society . . . For more than a decade following World War II cultural genocide was the deliberate, if not stated objective of most schools teaching Navajo students" (Roessel, 1979, p. 17). In 1969, the Special Subcommittee on Indian Education documented the failure of these assimilationist, culturally destructive policies in what is known as the "Kennedy Report." This situation is changing as Indian communities are gaining political control over their educational institutions and are affirming the foundational strength of American Indian cultures. It is likely, however, that non-Indians will continue to serve in a variety of functions in predominantly Indian school systems. Collaboration across cultural barriers in the service of goals defined by Indian Nations and communities requires a basis in respect and understanding. Such a relationship demands a consciousness that does not confuse differences with deficiencies and assumes a willingness to cooperate on the basis of equality. A non-Indian educator who accepts these essential assumptions may still experience difficulties due to the very nature of intercultural relationships.

The occurrence, form and meaning of a particular behavior in a particular situation may differ from culture to culture due to differences in norms, values, role perceptions and experiences (Triandis, 1972). Consequently, individuals from different cultures bring to cross-cultural interactions different implicit as well as explicit frameworks for interpreting experience. These differences in interpretive frameworks may lead members of two different cultures to view the same situation or behavior differently. The result may be confusion, misunderstanding or conflict (Albert, 1983). Attributions are inferences about the causes of behavior. Heider (1958) indicated that we are constantly engaged in the process of making inferences about causes and motives of behaviors in order to make our worlds more predictable and understandable. Although the behaviors that are or are not performed in any interaction are important, it is the interpretations we give to these behavior that are critical (Albert & Triandis, 1979). Attributions answer the vital "why" questions about a behavior. We, therefore, do not respond directly to the events we experience; we respond to the meanings or interpretations we give to these events. Discrepancies in attributions may result in misunderstandings, low interpersonal attraction, rejection and even conflict. Social psychology (Greenberg, Pyszczynski, & Solomon, 1982; Greenberg & Rosenfield, 1979; Jones & Nisbett, 1971; Ross, 1977) has provided evidence that attributional biases may be based on perceptual, self esteem maintenance and ethnocentric factors. It is suggested, then, that even with the best of intentions, the development of mutually respectful and cooperative intercultural relations can be difficult due to attributional differences.

Cross-cultural cooperation and collaboration can be greatly assisted if the non-Indian educator can learn to interpret behaviors and situations from the cultural/historical perspectives of Indian students, parents and colleagues. This ability may be assisted through intercultural sensitivity training.

The Intercultural Sensitizer (ICS), also called the "culture assimilator" (Brislin, Landis, & Brandt, 1983), has been a carefully researched intercultural training technique. The ICS is based on an "attribution training" approach to intercultural sensitivity which focuses on explanations of behavior from the point of view of the host's culture. It teaches the guests to make attributions isomorphic to those of the hosts concerning the meaning of frequently misunderstood behaviors and situations.

The Navajo Intercultural Sensitizer (ICS) is designed to provide extensive information about the attributional perspectives of Navajo people in an experiential programmed learning format.

Cross cultural research has been described as an attempt to comprehend systematic relationships between cultural level variables and the behavior of individuals (Berry, 1980) where culture serves as the independent variable (IV) and behavior as the dependent variable (DV). The IV of this study is the cultural identification of the respondent. The dependent measure used were the attributions Anglos and Navajos made when asked a question about the causes or meanings of identified critical incidents (Flanagan, 1954). Critical incidents are defined as any activity or situation that is sufficiently complete to permit inferences and predictions.

It was predicted that samples of Navajo and Anglo people would yield significantly (p < .05) different attributions to critical incidents identified according to procedures used in the construction of the Navajo Intercultural Sensitizer.

Generation of Critical Incidents: Critical incidents were gathered at two sites in the Navajo Nation. These incidents were elicited by the question, "Can you think of a situation or incident involving people from the other culture (Anglo people if you are Navajo, and Navajo people if you are Anglo), that you experienced or observed which resulted in some kind of difficulty, trouble, misunderstanding or left you with bad feelings or thoughts about the other group?" Critical incidents were gathered from Navajo teachers, teacher aides, high school students and Anglo teachers in Fort Defiance, Arizona. Incidents were also reported by Navajo middle school students in Lower Greasewood, Arizona. A description of the Navajo and Anglo respondents is shown in Table 1.

TABLE 1
Sources of Critical Incidents

Attribution Elicitation: The Navajo panel offered attributions for each of the 56 incidents selected. They offered their responses (attributions) to the question posed after each incident. The question referred to how and why the Navajo person depicted in the incident might interpret or respond to the behavior or situation described.

The elicitation of Anglo attributions was accomplished by presenting the identical incidents and questions to a subsample of 15 randomly selected respondents from the total Anglo sample of 72. The remaining 57 were assigned to the attribution selection (next) stage. The Anglo sample consisted of students at the University of Arizona who volunteered to participate in the study. The students were enrolled in a senior Education class (Education), a "300 level" educational psychology (Ed. psychology) course, a Masters level Counseling and Guidance (Counseling) course, a seminar of doctoral students in clinical psychology (Clin. psychology).

The "attribution elicitation" questionnaire differed from the empirical test questionnaire (attribution selection) in that it asked for open ended responses and included related questions used for feedback in the final construction of the ICS.

Attribution/Incident Selection: The samples drawn to test the hypothesis that Anglo and Navajo people would differentially attribute meaning to critical incidents were matched according to a method suggested by Campbell and Cook (1979) for extending external validity. The concern of the "impressionistic modal instance model" is to "explicate the kinds of persons, settings, or times to which one most wants to generalize and then to select at least one instance of each class that is impressionistically similar to the class mode" (p. 77). The Navajo sample was drawn from two primary sites in the Navajo Nation. In Fort Defiance respondents were teachers and teachers' aides at Fort Defiance Elementary School (FDES), high school students at Window Rock High School (WRHS), adult students of Navajo Community College (NCC), community members and Chapter House officials at the Fort Defiance Chapter House (Chapter). In Tuba City, the source of respondents were students and staff at Tuba City High School. The Fort Defiance sampling frame was selected by the panel members. In Tuba City, the director of student services recruited volunteers from what he considered to be a good representation of the student and staff population. Table 2 describes the samples used to determine which incidents were differently interpreted by the Anglo and Navajo groups (and would therefore be included in the ICS), and identify those attributions most frequently chosen by the Navajo sample.

The empirical test (see Appendix A) was constructed from the (panel) approved incidents and questions. Four alternative responses were offered in a forced choice format. One of the presented responses was the attribution elicited from the panel. The response of the reporter of the incident was additionally included if it differed from attribution offered by the panel. The remaining choices were developed from the attributions most frequently chosen by the Anglo group that participated in the "attribution elicitation process."

TABLE 2
Attribution Selection: The Empirical Test Samples

The subjective culture (IV) of the respondent was cross tabulated with the attribution chosen (DV) for each incident.

An independent samples chi square analysis was performed for each of the 56 selected critical incidents based on the responses of the Navajo (n 70) and Anglo (n = 60) samples to determine if the respondent's subjective culture was related to the attribution chosen in response to the incident presented. Incidents that yielded significant differences (p. < .05) were included in the ICS along with the corresponding attributions and the percentages of Anglos and Navajos choosing each response. The Cramer's V statistic was computed to indicate the strength of the relationship. This statistic, a slightly modified version of "phi," ranges from 0 to + 1.

A cross validation analysis was performed by constructing an additive scale based on the "predicted Navajo attributions" of the culturally knowledgeable panel. A respondent was given a Navajo Score of 0 + 1 when the predicted Navajo response was chosen (the attributions elicited from the panel). A mean "Navajo Score" was calculated for each group and then compared in a one-way Analysis of Variance to determine if the groups differed on this interval level variable. This analysis excluded cases where more than 26 of the 56 items were declared missing; therefore, the Navajo N = 60 and the Anglo n = 56 for the cross validation analysis. The influence of age on the attributions chosen by the Navajo respondents was also investigated.

The chi square analysis indicate that 46 of the 56 incidents tested produced results that were as predicted. That is, 46 of the 56 incidents yielded a significantly different (p < .05) pattern of attributions from the Navajo and Anglo samples. These results and computations of the strength of association (Cramer's V) between the respondent's cultural identification and attribution selection are partially reported in Table 3 (complete results are available on request).

TABLE 3
Chi Square Test Results of the Relationship Between
Subjective Culture and Attribution Selection per Incident

The "Navajo Score" (M = .52) of the Navajo sample (n = 60) was significantly higher than the "Navajo Score" (M = .48) for the Anglo group (n = 56) in the analysis of variance, F(1, 114) = 7.55, p < .01. A separate Anova compared the "Navajo Scores" of the Navajo respondents under 30 years old (n = 29) and those over 29 years old (n = 27). The score (M = .58) of the older Navajos was significantly greater than those below the age of 30 (M = .48), F(1, 54) = 32.12, p = < .01.

Since the "Navajo Score" (M = .48) of the Navajos below the age of 30 was identical with the "Navajo Score" of the Anglo sample further clarification was sought. A chi-square analysis was done contrasting the attribution selections of the Navajos below age 30 and the Anglo sample for each incident. In this analysis, 45 of the 56 incidents yielded significant differences (p < .05). In contrast to the older Navajos, the younger Navajo respondents and the Anglo sample did not differ significantly on four of the incidents while differing on two. There was correspondence between the older and younger Navajos in comparison with the Anglo group on the other incidents.

The Navajo Intercultural Sensitizer was constructed based on these results and the feedback given by the knowledgeable informants on the panel (see Appendix B for sample).

The primary (chi square) analysis supported the prediction that samples of Navajo and Anglo people would yield significantly different (p < .05) attributions in response to critical incidents identified according to procedures described in the construction of the Navajo Intercultural Sensitizer. The cross validation analysis offered further support for these findings. These differential attributional frameworks have produced confusion misunderstanding and/or "bad feelings" between the Navajo and Anglo participants in these interactions. In addition, the "Navajo Scores" of older Navajos (M < .58) differed significantly (p < .01) from that of the younger Navajos (M < .48) indicating within group age variation. The findings that the "Navajo Scores" of the younger Navajos and the Anglo sample were identical were enlightened by the fact that only four of the incidents were similarly interpreted by the two (younger Navajo and total Anglo) samples. These findings suggest that the younger Navajos share neither the attributional system of their elders or of the dominant culture. It remains to be determined if these results indicate the development of a "third culture" with a consistent and coherent attributional framework that serves the essential function of making sense out of the social environment and thereby prescribing adaptive action.

A Navajo Intercultural Sensitizer has been developed consisting of 48 critical incidents (two were added because of age variable considerations). Differential interpretive frameworks operative in Navajo-Anglo interactions were shown and cultural information relating to interpersonal relationships, relationship to nature, decision making style, beliefs, norms, and taboos derived from Navajo culture were offered. It is reasonable to assume that more such incidents could be identified and used for training purposes. Differences and similarities in the attributional systems between and within tribes could also be explored using this methodology. A general "American Indian Intercultural Sensitizer" could be developed as well as tribal specific ones.

A potentially important area of investigation is a further investigation of the attributional systems of younger Navajo and Indian people in relation to its function of social and behavioral mediation.

It is doubtful that this instrument would assist those who are committed to an ideology based on cultural or racial superiority. Only an alert, community controlled school system could effectively screen out such destructive influences. It was encouraging to note, however, that many prospective Anglo educators responded enthusiastically to an opportunity to be enriched by an appreciation of cultural diversity.

Questions relating to statistical conclusion, internal and external validity exist and need to be examined in subsequent papers.

A Navajo student, attending a "bordertown" public school, was so upset with his teachers (Anglos) that he wished "I wouldn't go to school." It seemed that the teacher would keep "pointing her hands to me." He wanted to tell her to point to other students, but didn't. He reacted very negatively to the teacher pointing at him.

A Navajo student, attending a "bordertown" public school, was so upset with his teachers (Anglos) that he wished "I wouldn't go to school." It seemed that the teacher would keep "pointing her hands to me." He reacted very negatively to the teacher pointing at him.

Please go to the next page

45-1

______________________________________________________________________________________________

45. WHY DID THE STUDENT REACT THE WAY HE DID WHEN THE TEACHER POINTED AT HIM?

Please go to page 45-3

Please go to page 45-4

Please go to page 45-5

Please go to page 45-6

45-2

______________________________________________________________________________________________

You selected a:

No. It is a good answer though. Approximately 26% of the Navajo sample agree with your choice. There is a more popular alternative. You have learned that singling out a Navajo student from the group may be a very uncomfortable situation for him or her. There is something specific about the gesture and the way Navajos tend to interpret it.

Please go back to page 45-1 and try again.

45-3

______________________________________________________________________________________________

You selected b:

Good. This was the second most frequently chosen alternative by the Navajo sample (31.0%). The Navajo consultants confirm that, "Pointing at a person directly is disrespectful unless you are singling out a witch or a skinwalker." Try again.

Please go back to page 45- and try again.

45-4

______________________________________________________________________________________________

You selected c:

Yes! This was the most popular Navajo response (41.4%). Pointing at a Navajo student can cause him or her great discomfort and should not be done. Navajos tend to point with their lips to avoid showing disrespect.

Please continue on to the next incident.

45-5

______________________________________________________________________________________________

You selected d:

No. Not at all. Only one Navajo respondent out of the sample of 70 thought so. You can do better than that.

Please go back to page 45-1 and try again.

45-6