Journal of American Indian Education

Volume 23 Number 1
October 1983


Theodore Coladarci, University of Maine

IN THE FALL OF 1980, several educators in a Montana high school district requested that an alternative to the conventional inservice teacher education program be offered--one that would address important social problems currently confronting their schools. Specifically, these individuals wanted an inservice that would enable them to explore empirically the factors contributing to the high dropout rate (approximately 60%) among Native American students--who constituted roughly 90% of the student body. In addition, implications derived from the study would be presented to the district administration and, it was hoped, would be considered in the formulation of district policy. The author agreed to direct such an inservice, which carried graduate credit through the University of Montana; the course was called "Practicum in Educational Research." This project, as is explicit in the course title, was in the form of a practicum--an "exercise," if you will, in empirical research. Availability of funds, personnel, and time did not allow the problem of school dropout to be investigated in the manner of a formal research organization. To be sure, this was never our intention. Nonetheless, district administrators were encouraged to examine our findings, cautiously and tentatively considering implications for policy. This article presents the conduct, results, and implications of this study of high school dropout. The practical relevance for the participants of this practicum is addressed elsewhere (Coladarci, 1982a).


The practicum involved (a) identifying the population, (b) developing instruments, (c) discussing and practicing interviewing techniques, (d) contacting and interviewing high-school dropouts, and (e) analyzing the resulting data.


Project staff, working from school records, compiled a list of 224 students who dropped out in the previous three years. As many dropouts as possible would be contacted and, with consent, interviewed regarding their reasons for dropping out of high school.


At the first meeting, project staff discussed at length various factors that were thought to contribute to school dropout. Some of these factors were intuitive or conjectural, some were based on personal experience, and others were gleaned from available literature. These factors subsequently were incorporated into a pilot questionnaire that was distributed to a sample of local teachers; a similar questionnaire was distributed to a sample of community members.

A list, compiled from completed questionnaires, represented factors that both teachers and community members tended to believe were important causes of school dropout among high school students in the area. Additionally, frequently reported reasons volunteered by both groups (i.e., factors not explicitly presented in the questionnaire) were included in this list.

The factors in this list served as a basis for 27 items on the questionnaire ultimately to be administered to the dropouts. These items were in the form of "reasons for dropping out," which were read to the dropout by the interviewer. (For example, "You got involved with drugs and alcohol.") The dropout’s response was forced to one of four choices: (1) not important at all, (2) somewhat important, (3) important, and (4) very important. Project staff also developed several open-ended questions, the responses to which were thought to carry additional import for district policy. Further, demographic questions were included concerning, for example, sex and residency. For all questions, particular care was exercised in keeping the language simple. (The questionnaire is available from the author.)


The questionnaire, as mentioned above, was administered in the form of an interview. Project staff believed that general reading ability among dropouts would vary considerably. So that their responses would not be influenced by their reading ability, dropouts were not to be required to read the questions.

Considerable time was spent discussing the "social psychology" of the interview: for example, the importance of establishing rapport and appearing non-judgmental. Also discussed was the judicious use of "wait time" and "probing."

In addition to project staff, several high school students were recruited for interviewing. These students were enrolled in a peer-counseling class taught by one of the participants, where the topic of interviewing was formally discussed. These students practiced conducting interviews with the questionnaire, as did the project staff. These practice sessions also provided an opportunity to pilot the questionnaire, resulting in several minor changes in language that improved the clarity of the respective questions.


Project staff was rather resourceful in attempting to make the initial contact with dropouts. Personal visits to the home were made where the addresses were available. If not, known friends and relatives were asked for assistance in making the contact or information pertaining to the dropout’s whereabouts.

There was considerable difficulty encountered during this phase of the project: Of 224 people on the dropout list, 46 were successfully contacted and interviewed. Approximately 35 dropouts had moved from the area and could not be reached, 12 refused to be interviewed (the refusal took several forms), and 6 were deceased. Thus, roughly 125 dropouts were not interviewed, although (presumably) in the area. There were two principal reasons for this. First, their whereabouts simply were not known by anyone queried by project staff. Second, if the whereabouts were known, repeated attempts to contact the dropouts were unsuccessful—often because streets had no signs and houses had no street numbers or phones. After several months, contact attempts were aborted. Clearly, the dropouts ultimately interviewed did not represent a random sample of the dropout population. One could easily argue that the inaccessible dropouts--those not interviewed--probably differed in significant ways from those who were accessible and agreed to the interview. Though what these differences are is open for conjecture, one nonetheless should be cautious in making generalizations from this study.

For those dropouts who were contacted and agreed to be interviewed, the interviewer read the questions and recorded the responses. For the first 27 items, this involved simply circling the number corresponding to the appropriate verbal designation (e.g., 3 = important). The open-ended questions were handled in one of two ways. If the dropout agreed, the responses were tape-recorded; after the interview, answers were summarized on the questionnaire from the tape-recording. If the dropout (or the interviewer) felt uncomfortable with this arrangement, the interviewer took notes while the subject responded.

Data Analysis

The responses to the first 27 items, expressed on a four-point scale, were collapsed into an artificial dichotomy. If the subject’s numerical response was a 1, he was assigned a 0; 2, 3, and 4 were assigned a 1. A 0 thus represented a factor (not in the statistical sense) that was "not important" in an individual’s decision to drop out, while a 1 now represented a factor that was "important" (to some degree) in an individual’s decision to drop out. Although the two-point scale in one sense was not as informative as the four-point scale, the results were simplified considerably; consequently, trends were seen more easily. The number of dropouts reporting that a particular factor was "not important" or "important" in their decision to drop out was determined for these items. The intercorrelation among the first 27 items were calculated, as well, using the full four-point scale. (Tables are available from the author.)

Responses to the open-ended questions were read carefully by project staff. For each question, categories of responses were identified. Consider, for example, the question "What could have been done to change your decision to leave school?" Eight responses referred to the nature of the relationship between dropout and parent(s) and, therefore, was identified as one category for this question. Once categories were listed for each question and assigned values, two staff members independently coded the responses to these questions.

There was perfect coder agreement for almost all responses. Where there was non-agreement, the particular response was discussed until the nonagreement was resolved. Coded responses for all questions subsequently were keypunched and analyzed at the University of Montana computer facility.

Major Findings

First, a caveat: One should interpret these data cautiously. There is, of course, a distinction in principle between "perception" and "reality." For example, over a third of the cases reported that the uneven application of school rules was a salient factor in their decision to drop out. Were rules, in fact, unevenly applied by school officials? From this study, we cannot say. We do know, however, these dropouts perceived school rules to be applied unevenly. To conclude this was actually the case is a hasty inference. (This is not meant to imply there is no reason for concern. A perception--accurate or not--nonetheless needs to be addressed by school officials.)

Factors pertaining to teacher-student relationships were important in the decision to drop out. Over a third of the dropouts cited as a factor that teachers did not care about them (Question 19). (Interestingly, this weighed more heavily for female dropouts.) The perception that teachers did not provide enough assistance with the student’s work (Q20) was also a salient factor in the dropout decision. Given the high correlation between these factors (r = .85), one may argue it was this perceived lack of assistance that was foremost in the perception that teachers did not care. Additionally, however, the perception that teachers did not care (Q19) is somewhat related (r = .57) to the sentiment that school was not important to Native American culture (Q18). Thus, Q19 may also reflect, to some degree, a (perceived) cultural insensitivity or indifference on the part of teachers.

A related factor, having disagreements with teachers (Q26), was important in the dropout decision in a third of the cases. Q27, getting into trouble at school, is, as a factor, rather nonspecific, although cited by over a quarter of the dropouts as important in their decision. Because it correlates strongly (r = .70) with Q26, one may speculate that "trouble in school" largely took the form of disagreements with teachers.

The importance of teacher-student relationships was also seen in response to several open-ended questions. When asked what could have been done to change their decision to leave school, several dropouts indicated that greater encouragement from their teachers would have been important. Similarly, suggestions for teachers aimed at reducing the incidence of dropout centered on providing more help and showing more concern which, interestingly, were echoed in providing suggestions for administrators.

The content of schooling also emerged as a salient factor in the dropout decision. A little less than half of the dropouts cited as a factor that school was not important for what they wanted to do in life (Q7). Roughly a quarter of the dropouts cited that school was not important to them as Native Americans (Q18). However, given the low correlation between these factors (r = .15), Q7 and Q18 apparently do not represent a concern common to both factors. For example, Q7 may reflect the felt need for job-related skills or general information about different vocations, while Q18 might represent the perception that the curricula did not adequately embrace Native American culture. Q18, further, correlates substantially (r = .89) with Q17, a factor regarding the desire for more Native American teachers. One may infer from this that the "agent" involved in schooling is an important consideration in perceiving the importance or relevance of school to Native American culture.

Lack of parental support (Q22) appeared to be a salient factor for the dropout decision in roughly 40% of the cases and also emerged in response to the question, What could have been done to change your decision to leave school? Suggestions for parents from the dropouts focused on providing more encouragement, communication, and cooperation. Further, over 40% indicated that problems at home (Q24) were salient in the decision to drop out more so, interestingly, for females (48% vs. 35%). One should realize that Q24, relative to Q22, is a more comprehensive factor (i.e., involving variables beyond parental support). These factors do correlate (r = .58), however, suggesting that there is some overlap.

An important outcome of these interviews was finding that over 90% of the dropouts would advise prospective dropouts to either stay in school or, at least, reconsider their decision. Indeed, the majority of dropouts (78%) indicated a considerable change in attitude after having dropped out. Over half reported that, as a consequence of dropping out, only menial jobs were available and their attitude toward life had suffered. These results alone represent a potent "word to the wise."


Insofar as the present study represented a practicum in educational research, resulting recommendations must be made--and entertained by the reader--cautiously and tentatively. Here, we share several observations and raise several questions.

1. The content of schooling was a salient factor in the dropout decision. It appears that this discontent had two general foci. The first pertained to the degree to which the curricula comprised skills and knowledge that were important to succeed in society as perceived by the students. While a high school should not be expected necessarily to assume a vo-tec posture, district officials might examine their curricula for "practical" content. Are the curricula, in fact, irrelevant or nonspecific to the concerns of students about to enter the adult world? Wolcott (1967), in his ethnography of schooling in the Kwakiutl village, made the point well:

Contrast, for example, the typical social studies unit of the elementary school on "Lumber, Our Natural Resource," and its conservation and technology themes, with the specific kind of information an Indian boy needs in order to decide whether to seek a logging job--what will be expected of him, how dangerous the work is, how to cope with the constant order of a White strawboss, or how to get a good name as a logger. (p. 128)

Although Wolcott made these observations some time ago, they doubtless are relevant today to some degree. Insofar as this latter conjecture is plausible, the implications for curriculum development and evaluation in this context are clear (see, e.g., Bergstrom & Goldenstein, 1982).

Second, do the curricula adequately reflect Native American culture? Adolescents who tenaciously identify with a particular culture can easily become dissatisfied with a local institution that appears oblivious or, worse, insensitive, to their identity. Given the cultural homogeneity among students in the area--90% are Native American--this question is particularly important. In his ethnography of the Mopass school, King (1967) found that the local curriculum was simply a direct adoption of the British Columbia Programme of Studies for Elementary Schools, augmented by Yukon Territory supplements. Both of these documents, however, made no reference to Native American children or Native American schools. Instruction, observed King, focused on "reading, writing, and arithmetic, with the first of these featuring Dick, Jane, Spot, Puff, and the all-too-familiar friends, neighbors, pathways and streets of both Canadian and United States textbooks" (p. 49). Educators in schools with a predominantly Native American student body need to determine the degree to which King’s anecdote characterizes their curricula and instruction and, further, the corresponding impact on students’ attitudes toward school. (This point, of course, generalizes to other edinic/cultural groups-)

2. The nature of the teacher-student relationship emerged as a salient factor in the decision to drop out: Dropouts perceived teachers as not caring about them and not providing them sufficient assistance in their work. Is there any basis for these perceptions or, alternatively, are these students simply seeking rationalizations for their desire to drop out? Indeed, this is not an easy question to answer. If the dropouts’ perceptions are realistic, how could teachers demonstrate more caring? What forms might such caring take? To what degree may this not caring involve insensitivity to Native American culture? Is there an adequate support system for students experiencing difficulty with their school work?

Dropouts reported that school administrators should demonstrate more care, understanding, and encouragement. In what context typically do students interact with administrators? Are there opportunities for these two groups to meet, for example, that do not involve disciplinary matters?

3. A finding that was not discussed above is pertinent here. Over a third of the dropouts reported that the desire to be with other dropouts was a salient factor in their decision to drop out. This may be interpreted as a sign of peer pressure, however subtle or forceful this pressure may be. Perhaps prospective dropouts would profit from hearing of the experiences of actual dropouts. Inasmuch as these experiences were reported to be largely negative, the prospective dropout may be given cause to reconsider. A number of dropouts, in fact, recommended that the administration arrange such discussion sessions involving prospective and actual dropouts, with this end in mind.

A further factor implicating peer pressure to some degree is the involvement of students with drugs and alcohol. Approximately a third of dropouts indicated this as a salient factor. Are there any provisions within the district to curb this problem among students beyond, say, a perfunctory unit in the curriculum? What is currently done when a student is suspected of having or developing—a problem in this regard?

4. A frequent complaint was that students had to attend school for the entire senior year, even though they merely needed one or two classes to graduate. Over a third, in fact, reported this as a salient factor in their decision to drop out. What is the basis of such a requirement? Is this requirement necessary?

5. Problems associated with the home are particularly difficult for educators to address. Nonetheless, such problems surfaced in the present study. The very least that could be done by educators, probably, is to develop an awareness of and sensitivity to a student’s home life: separations or divorces, unemployment, alcoholism, child abuse, and so on. If nothing else, such knowledge would help educators interpret the student’s school behavior and, possibly, entertain certain lines of action while ruling out others.


As a practicum in educational research for inservice educators, this project was successful. First, participants, through the practicum, became familiar with the general philosophy of science: problem formulation, the confirmation and refutation of a priori hypotheses, consequent implications. Indeed, participants ultimately acknowledged that the importance of subjecting one’s beliefs to empirical test is the same whether one is an experimental psychologist in the laboratory or a high school teacher in the classroom (see Coladarci, 1982b). Second, participants engaged in planning and executing research--from formulating the research problem to writing policy recommendations.

As a study of school dropout among Native Americans, this project, admittedly, raises more questions than it can answer. A more ethnographic methodology (e.g., Spradley, 1980) doubtless is required to clarify the results obtained in this study and to address many of the questions raised in the preceding section.


Bergstrom, R., and Goldenstein, E. H. Developing authentic curriculum on Native American life. Educational Leadership, 1982, 39, 549.

Coladarci, T. Empirical research as inservice teacher education. Paper presented at the annual meetings of the American Educational Research Association, New York City, 1982. (a)

Coladarci, T. Philosophy of science and teacher education. Paper presented at the annual meeting of the Rocky Mountain Educational Research Association, Albuquerque, New Mexico, 1982. (b)

King, R. A. The school at Mopass: A problem of identity. New York: Holt, Rinehart and Winston, 1967.

Spradley, J. P. Participant observation. New York: Holt, Rinehart and Winston, 1980.

Wolcott, H. F. A Kwakiutl village and school. New York: Holt, Rinehart and Winston, 1967.

Theodore Coladarci is Assistant Professor in Educational Psychology at the University of Maine. He can be reached c/o School of Education, University of Maine, Orono, Maine 04469.

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