Volume 31 Number 2
January 1992

Karen Swisher and Michelle Hoisch

This article is a compilation of research information on American Indian and Alaska Native (AI/AN) dropouts gathered by the Center for Indian Education at Arizona State University for two studies conducted between 1989 and 1992. The first of these two studies began in 1989 with a grant from the National Education Association (NEA). The purpose of the study was to collect existing data on a national level about AI/AN dropouts in grades K-12 from state departments of education; public, private, and Bureau of Indian Affairs (BIA) school systems; and existing studies completed by individual and/or agencies (Swisher, Hoisch, & Pavel, 1991). The intent of the project was to determine a national dropout rate for AI/AN students and to eventually establish and maintain in the Center for Indian Education a much needed clearinghouse for information on AI/AN dropouts. The second study evolved from difficulty encountered during the NEA study in obtaining comprehensive data from the BIA school system on its dropouts. The response was a project, funded by the Bureau of Indian Affairs, to collect data on dropouts and transfers in the BIA school system (Center for Indian Education, Arizona State-University, in press). This study began in May of 1991 and was completed in December of 1991. All of the schools in the BIA system were contacted and asked to provide their records of student enrollment, the number of dropouts identified, and the number of transfers into and out of the school for the previous academic year.

On each study a review of the research literature was conducted. Only those studies and reports that were related to AI/AN dropout issues were included in the final reviews of literature. In this article, we present a comprehensive review of the research studies and reports on AI/AN dropouts in grades K-12.

A small but growing volume of education research illustrates the significant problem of American Indian/Alaska Native (AI/AN) children dropping out of and/or transferring between schools at a higher rate compared to other ethnic and racial groups (Swisher et al, 1991; Office of Indian Education Programs, 1988). Studies have found dropout rates that range from 29% (Deyhle, 1989) to 52% (Wells, 1991) and in some extreme instances rates of 90% or higher are cited (Platero, Brandt, Witherspoon, & Wong, 1986; State of Minnesota, 1988). Concern about high secondary dropout and transfer rates has intensified owing to indications that those individuals who lack a high school diploma are at a distinct disadvantage with respect to their ability to support themselves in today's society. In addition, concerns have been raised that the future competitiveness of the United States economy will be linked to the educational level of its future workers. Thus, where chronic dropout situations exist, the need for monitoring outcomes and implementing effective practice becomes even more paramount. Finally, the national goals established by the Bush Administration include a goal to increase the high school graduation rate to 90% by the year 2000.

In examining studies which have attempted to quantitatively describe the AI/AN dropout rate, several factors emerge that make establishing such a rate difficult. Auerbach, Fuchs and MacGregor (1970) discussed the complications created by the multiple school systems which AI/AN students are eligible to attend. These three systems, federal, state, and private, have not established agreements with respect to maintaining comparable records or using similar methodologies to compute the dropout rate; therefore, combining the data from these three systems into a national picture of the incidence of dropping out among AI/ANs is impossible. To complicate matters further, students often transfer, both between systems and across state lines, making the tracking of student progress difficult. When a student transfers either between systems or between states, the school which he or she left often has no way of verifying that the student did indeed transfer and did not drop out between systems unless a transcript is requested. Thus, dropouts may be misclassified as transfer students and transfer students may be classified incorrectly as dropouts.

In general, methodological differences between studies carried out to determine an AI/AN dropout rate make the results of any one study difficult to discuss with respect to most other existing studies. There has, to date, never been a large-scale study which establishes a comprehensive dropout rate for AI/AN students nationwide across all of the school systems in which AI/AN students enroll. The most comprehensive studies have been regional, rather than national in nature (Owens & Bass, 1969; Platero et al., 1986; Selinger, 1968). The only hope of painting a national picture has been to combine data from several studies in an attempt to create a coherent whole. The lack of uniform definitions for either "dropout" or "American Indian and Alaska Native" and the use of different methodologies to calculate the dropout rate itself are factors that have been identified in the literature as troublesome to anyone trying to discuss the AI/AN dropout rate and using data from more than one study (Auerbach et al., 1970; LeCompte & Goebel, 1987; Morrow, 1986; Swisher et al., 1991). Thus what can be examined with respect to the literature on AI/AN dropouts is a collection of various independent studies, pieces of different jigsaw puzzles which can be partially assembled and examined to give a sense of what is going on in AI/AN education.

Kutsche (1964) examined the dropout rates of Al/AN students attending Cherokee High School, a BIA operated school, in North Carolina in comparison with that of primarily White students attending two neighboring public high schools. The results indicated a three-year dropout rate of 17.9% for Cherokee High School, compared to 8.0% and 7.7% for the largely non-Indian schools, Sylva-Webster High and Swain County High, respectively. Kutsche concluded that Cherokee high school students dropped out at a higher rate in all grades than non-Cherokee students with the exception of twelfth grade. He did not report data about reasons for leaving school, nor was there any speculation about why the dropout rate for twelfth grade was less than other grades. However, because Kutsche's study did not incorporate a sample of AI/AN students attending the two public high schools (assuming there were any) this study might have said more about the difference between the two school systems (public and BIA) than between the two types of students.

Coombs (1970) reported that two studies in the late 1960s, one by Selinger (1968) and the other by Owens and Bass (1969) are the first examples of reliable longitudinal studies that track AI/AN student membership over a specific period. Both studies tracked a sample of students within the public, private, and BIA systems in different regions of the country. Selinger sampled schools in the Northwest and Plains whereas Owens and Bass selected schools in the Southwest and Plateau regions.

Selinger's research (1968) was designed to provide statistical information on Al/AN dropouts in the Northwest, drawing a stratified random sample of students from selected schools in Oregon, Washington, Idaho, Montana, North Dakota, and South Dakota. In this study, students were tracked individually from eighth grade in 1962 to twelfth grade in 1967. In addition to identifying dropouts in BIA, state, and privately-operated schools, the study looked at dropouts by area, state, tribal group, and the majority-minority position of Indian students in attendant schools. Progressive state dropout rates for AI/AN students in all the sample schools ranged from a low of 28.6% in Oregon to a high of 57.6% in South Dakota.

Owens and Bass (1969) examined the AI/AN dropout situation by drawing from a sample of schools in Arizona, Nevada, New Mexico, Oklahoma, Southern Colorado and Southern Utah. Selected eighth grade AI/AN students were tracked individually from 1962-63 to 1966-67 when they were scheduled to graduate as twelfth graders. A comparison of dropout rates by school type (BIA, public, private), state, grade level, and tribe was included in the report. The study reported an overall AI/AN dropout rate of 38.7% for a sample of 1,217 students.

The AI/AN students in Arizona, Nevada, New Mexico, and Oklahoma who stayed in the public system were more likely to drop out than those who stayed in the BIA system (e.g., in New Mexico, 21.7% of the students who began the eighth grade in BIA schools dropped out whereas nearly 30% who began in public schools dropped out). The students who started in the BIA system but later transferred to the public school system were more likely to drop out than those who started in the public system and transferred to the BIA system (e.g., in Arizona 2.8% who started in the public schools dropped out of BIA schools whereas almost 14% who transferred to public school from BIA schools dropped out). The students who began in the BIA schools and transferred to the public system had a lower dropout rate than those who stayed in either the public or BIA system. These findings suggest that AI/AN students who are limited to any one school with no options to transfer fare worse than students who do have such an opportunity.

Another study of AI/AN dropout rates was undertaken in the early 1970s at Ft. Wingate High School, a BIA operated school outside of Gallup, New Mexico (Horton & Annalora, 1974). This study was carried out only for the Fall semester, and students who had transferred out of the school were counted as dropouts, a practice which is not generally followed in calculating dropout statistics. A 26% dropout rate was reported, but, due to the short duration of the study and the fact that transfers were included in this rate, it is of questionable value as an indicator of dropping out within the BIA system.

Latham (1985) examined data from on-site visits, review of literature, questionnaires to schools, BIA offices, state offices of education, and telephone contacts with officials and administrators of various federal agencies and BIA and public schools. He found mean annual dropout rates for high school grades of 32% for students in BIA and BIA contract schools and 24% for AI/AN students of federally recognized tribes enrolled in public school programs. Latham observed that because schools are awarded federal funding based on the number of AI/ANs enrolled during fall count week, considerable extra effort is made at many schools to assure optimum attendance only up to and during this period. AI/AN students apparently transferred frequently after this count week and found placement in a BIA school as their last resort to educational opportunity. Students transferring out of public schools into a BIA school, then dropping out altogether from the BIA school (or vice versa) resulted in a dropout rate which escalated as high as (and even higher than) 50% in some schools. Very few of those who dropped out either acquired a high school diploma or fulfilled high school graduation requirements.

A recognition of the growing problem of Navajo student dropouts provided the impetus for Navajo Students at Risk, a Final Report for the Navajo Area Student Dropout Study conducted by Platero, Brandt, Witherspoon, and Wong (1986). Encumbered by insufficient data and speculative reasons regarding the dropout problem, the Navajo Division of Education (NDOE) realized that preventive programs could not be developed until it knew more about the problem. The purpose of the study was to provide current information on the

Three databases were developed for this study. The student roster database included individual student records from 101 (of 259) schools. The school characteristic database included information from the school characteristics survey returned from 86 (of 259) schools. The student and former student database included surveys from 670 stayers and 219 dropouts. Platero et al. (1986) used a two-stage stratified sample with schools by type of school (BIA, contract, public, and private) as the first unit of analysis, and students as a second unit of analysis. From this vast amount of information, the Navajo dropout rate was determined to be 31%.

In 1984, Deyhle began a four year ethnographic study of Navajo and Ute school leavers (Deyhle, 1989). Her research included a longitudinal study of high school completion and dropout among five separate cohorts of students attending two high schools. The longitudinal dropout rates for the five cohorts, categorized by class membership were as follows; Class of 1984-31%, Class of 1985-29%, Class of 1986-3 1 %, Class of 1987-40%, Class of 1988-50%. Graduation rates were calculated as well, with ninth grade students often tracked beyond the traditional four years to include those graduating later through GED programs or with extra years of schooling. As many as 18% of the students in the study were physically in school for twelve years and still failed to graduate. In addition, over half of the students who dropped out did so in the twelfth grade, "indicating a persistence towards getting a high school degree, and a significant number of the graduates (20%) managed to graduate only through additional years or alternative programs" (p. 38). To ascertain the reasons that these young people give for leaving high school, Deyhle's study includes interviews with students who dropped out.

In 1989, the Center for Indian Education at Arizona State University undertook a project through a grant from the National Education Association (NEA) to determine a national dropout rate for AI/AN students and to establish a repository for information on AI/AN dropouts on a national level (Swisher et al., 1991). The need project stemmed from a need by the NEA and other agencies to report an Al/AN dropout rate to Congressor governmental departments for program justification. In addition there was the realization that no centralized location to serve as a clearinghouse for the information on AI/AN dropouts nationwide existed.

Al/AN students attend public, private, and BIA schools, and very few studies had attempted to combine dropout rates across systems. Those studies that had included all three systems had been restricted to regional samples. Because the individual states which monitor the public school system do not have a system of communication or information exchange for their individual dropout statistics, this project was also designed to centralize this information with respect to AI/AN students. The study also included an examination of some of the current literature on reasons and correlates identified for AI/AN dropouts. Data for the NEA study were gathered from state agencies, national agencies, offices of the BIA, tribal entities, and by examining the literature on AI/AN dropouts. The state agencies targeted were those in the 20 states identified with the largest AI/AN populations according to the 1980 census. Several individual school districts that receive Title V funds also responded individually when they heard that the NEA study was underway.

Data were requested for AI/AN students in grades K-12. The most comprehensive data came from 26 state education agencies, all of whom responded with either comprehensive data or an explanation for why such data were not available. Data from BIA offices and tribal entities was sporadic and difficult to obtain. With respect to quantitative data on AI/AN dropouts, the study focused primarily on what came in from the states. Because the state education agencies computed dropout rates using different, and sometimes incomparable methods, it was ultimately impossible to combine them to compute a national dropout rate for AI/AN students. Instead, individual dropout rates by state were reported with appropriate qualifications to explain why some states appeared to have high dropout rates in comparison to others (some states reported longitudinal rates using cohorts, some reported on grades K-12 on a whole whereas others broke out rates for grades 9-12, etc.). The data available were insufficient to compute statistics for private or BIA school systems.

Largely due to the inability to collect data on the BIA school system for the NEA study the Center for Indian Education undertook a second study, this time focusing only on dropouts within the BIA school system (Center for Indian Education, Arizona State University, in press). Sponsored by the Office of Indian Education Programs (OIEP), the study requested enrollment and dropout statistics from each of the schools in the BIA school system, both by mail and over the telephone. The study also collected information on student transfer, which was speculated to impact heavily on the BIA education system. Responses were solicited from school staff on what contributed most to student dropout and transfer and what they felt were the vital components of effective intervention programs.

The BIA study had a 72% response rate, or 120 out of the 166 schools surveyed. One of the limitations discussed in the study was the fact that without the data from the missing 46 schools, the results could not be accepted as completely representative of education within the BIA system. With this in mind, the study reported a 25% dropout rate in grades 9-12 in the BIA system. Dropout rates were also computed separately for grades 7 and 8. Dropout rates were further separated by grade level (7-9), by type of BIA school, and by state. It appeared that, within the BIA education system, dropouts were not identified before grade 7.

Another possible limitation of the study was that, due to time constraints, the data received from the schools were accepted with no further accuracy verification from the research team. To further explore whether all of the schools counted their dropouts in the same manner or to re-check the numbers provided by each individual school was impossible. In the past the BIA attempted to utilize some type of standardized computer database across schools, but lack of adequate training and the cumbersome nature of the program forced most schools to abandon it in favor of manual or other record keeping systems. This study remains as one of the most current and comprehensive attempts to collect dropout statistics within the BIA system.

The three major student longitudinal studies that the National Center for Education Statistics (NCES) has sponsored over the past two decades are: the National Longitudinal Study of the High School Class of 1972 (NLS-72), the High School and Beyond (HS&B) project, and the National Educational Longitudinal Study of 1988 (NELS:88). The small sample size of Al/ANs in these data bases makes it statistically tenuous to explore the dropout issue. However, the true value of knowing about and analyzing the AI/AN data in NCES student data bases lies in the potential to increase the Al/AN sample size in future studies.

NLS-72, the first NCES sponsored longitudinal student study, gathered data from a sample population of nearly 23,000 high school seniors in over 1,000 schools beginning in 1972. In the base-year and four follow-up surveys, data were gathered on students' secondary school background, academic ability, activities related to work, post secondary experiences, family life, and personal ambitions. Survey data were also collected from school administrators. The fifth follow-up survey in 1986 continued to collect considerable data on work experiences and periods of unemployment, post secondary achievements, family history, and personal ambitions. Through these surveys, NLS-72 compiled a very large volume of data relating to pre- and post-high school experiences of students.

The AI/AN sample in NLS-72 was listed at 242 respondents which represents about 1% of the total sample. Published reports using NLS-72 data never report specific findings on AI/ANs because the standard errors are too large due to the small sample size. It is of interest to note that a descriptive analysis of the unweighted data indicates only 12% of the AI/AN sample were college graduates by 1986 compared to 42% of the Asians, 30% of the Whites, and 22% of the Afro-Americans.

The second NCES-initiated longitudinal study, HS&B, included both a high school sophomore cohort and a senior cohort. These two cohorts were surveyed in order to collect additional data related to high school persistence (the "dropout problem") and the transition to and success in post secondary education. HS&B collected baseline survey data from over 52,000 sophomore and senior high school students in over 1,100 schools in 1980, with subsequent follow-up surveys in 1982, 1984, and 1986. Student questionnaires gathered data on individual and family backgrounds, high school experiences, work experiences, plans for the future, and cognitive tests. A separate survey collected data on school enrollment, staff characteristics, educational programs, facilities and services, and dropout rates (for sophomore cohort only).

A self-reported sample of 351 "Native Americans" was listed in the sophomore cohort, of which 122 were recoded to White, non-Hispanic based on speculation that some respondents thought this category meant being born in America. This left a total of 229 respondents identified as "Al/AN" in the sophomore cohort. The dropout rate of this sample based on a weighted estimates was approximately 23% compared to 7% of the Asians and 14% of the Whites. A sample of 229 Native Americans in the senior cohort was self-reported, and again 20 were recoded to White, non-Hispanic. This left a total of 209 respondents identified as "Native American" in the senior cohort. To extrapolate on the dropout issue for the senior cohort is difficult, but follow-up surveys six years later indicate that over 70% of the AI/AN sample had only a high school diploma or less compared to around 50% of the Whites.

Studies generated from the high school cohort data in NLS-72 and HS&B indicated the need to examine student development and educational experiences at even younger ages. As a result NELS:88 became the third major longitudinal student study the National Center for Educational Statistics conducted. NELS:88 collected baseline data from 24,599 eighth graders in 1988 and will conduct follow-up surveys through middle school, high school, and college or employment. Key research issues include mathematics and science programs, effective schools, academic growth, dropouts, disadvantaged students, bilingualism, transitions to high school and college, and the influence of peer groups.

A self-reported sample of 936 "Native Americans" in the NELS:88 data was identified, of which 639 were recoded to White, non-Hispanic for the same reason stated earlier. This left a total of 307 respondents identified as "Native Americans." Pending this group's anticipated graduation from high school in 1992, the final data for the NELS:88 cohort is not yet complete. The National Center for Education Statistics has released an interim report on the NELS:88 cohort (NCES, 1991). A cohort dropout rate among AI/AN students of 9.2% is reported, as well as a 10.2% rate for the Black, non-Hispanic cohort, 9.6% for the Hispanic cohort, 5.2% for the White, non-Hispanic cohort, and 4.0% for the Asian/Pacific Islander cohort.

Research about dropout rates is useful for illustrating the effectiveness of school systems and the magnitude of the dropout problem. Dropout research, however, cannot and does not stop here. Research which identifies correlates to dropping out and the reasons given by dropouts themselves for their decision to do so is also needed. Such information is integral in the implementation of programs aimed at reducing dropout rates, and assisting educators in both identifying target populations and designing interventions that accurately address the needs of prospective dropouts. Interest in these issues is apparent in much of the AI/AN dropout research performed in the last three decades, particularly in the 1980s.

In 1967, Rosalie H. Wax wrote about the dropout experience on the Pine Ridge Reservation in South Dakota based on data gathered by Robert V. Dumont, Jr. 1962-1963. Titled, Oglala Sioux Dropouts and Their Problems, the article described the experience of dropping out among Oglala Sioux boys. The article was later rewritten and retitled, The Warrior Dropout (Wax, 1967a). Wax suggested that "neither the dropout nor the process of dropping out are well understood" (p. 247). It is assumed that dropouts are alike and drop out of school for much the same reasons; they dislike school and reject school. But, according to Wax (1967), they leave school under different conditions and for different reasons. She stated:

Wax presented a descriptive analysis of how Sioux boys come to drop out of high school. The interview sample included 153 youth between the ages of 13 and 21, 35% of whom had dropped out before the end of the ninth grade. Rarely did the young men mention trouble with studies, but about half found school a lonely place or unbearable for other reasons such as abuse by experienced students. The school population was quite heterogeneous, and Wax took particular notice of the students she identified as Country Indians, i.e., those who were raised out in the country and were more traditional than their town counterparts or those who had spent a great part of their lives in boarding schools. For many of the dropouts, school was "all right," but conforming to regulations was too difficult. Wax stated that most of the young men

These youth are told that they must graduate from high school in order to get a job, but in order to graduate they must develop qualities opposite from those they possess:

Wax's study is significant in that the voices of students were presented. She was able to descriptively tell the story from a student's point of view. In the 1980s, researchers conducting three of the major studies of AI/AN dropouts, Colodarci (1983), Platero et al. (1986) and Deyhle (1989), incorporated similar goals of researching what the dropouts themselves identified as the factors contributing to their decision to drop out.

In 1980, educators in a Montana high school district asked for a different kind of in-service program; they came up with a study that was the first of several studies which examined the factors contributing to high dropout rates among Al/AN students. Under the direction of Coladarci (1983) the dropout rate 60% was examined through interviews with 46 of 224 students who had dropped out during the past three years. The students' responses were categorized into three areas: teacher-student relationships, content of schooling, and parental support. The factors that students said influenced their decision to drop out included:

Teacher student relationships:

The Coladarci study further pointed out the desire to be with other dropouts and/or some degree of peer pressure as salient factors in these students' decisions to drop out. Another element discussed was the students' dissatisfaction with the fact that even though they may have only one or two classes left to complete, they were required to attend school for the entire year. Home problems were also identified in this study as elements that led in part to dropping out, but Coladarci interjects the realization that "home problems may be particularly difficult for educators to address" (p. 21). Interestingly enough, over 90% of the dropouts stated that they would advise other students contemplating dropping out to stay in school. Recommendations are made in this study for educators to examine the curriculum, both its relevance to AI/ANs and the extent to which it reflects Native culture, and to also explore the nature of student/teacher and student/administrator relations in an attempt to address the dropout issue.

Platero and his colleagues (Platero et al., 1986) explored the correlates for dropping out among the Navajo students in their study, and also conducted interviews with both stayers and dropouts. They presented a social/cultural interpretation centered around four main areas: the nature of Navajo dropout behavior, schooling and socialization, the home environment and dropout behavior, and the students themselves and dropping out. Their survey indicated that among Navajo students, academic problems seemed to be a minor factor in dropout behavior. Evidence supporting this conclusion included a lack of a significant difference in grade retention between persisters and dropouts; small difference with regard to doing homework; a self-report among dropouts that their academic performance was average to very good; and the relatively minor role that academics play in the list of factors dropouts gave as to why they dropped out of school.

The Navajo study reported the reasons that students who had dropped out of school gave for their decision. Students reported that the primary reason for dropping out was that they were bored with school. The next most influential factor was problems with other students, followed by retention due to absenteeism and pregnancy/absenteeism. Fifth, and perhaps more significant for boys was discipline problems, either at home or at school. Students who were contemplating dropping out were also surveyed to explore why they were considering this option. They too spoke of boredom in school as the most significant factor influencing their decision. This was followed by being held back by absenteeism, problems with other students, and problems with teachers.

The Platero study recognized that the reasons students gave for dropping out may differ from the reasons that administrators assume lead to the decision to drop out. Because many schools are unable to contact dropouts to find out why they left school, their only source for identifying the reasons that these students left school is the intuition of administrators. Platero and his colleagues questioned school personnel as to what, in their opinion, caused students to drop out of school. The results indicated that school personnel did not understand as well as they thought what motivated students to drop out. Although administrators cited lack of familial encouragement, academic problems and performance, and home and family problems as the three most likely reasons why students drop out; students claim that boredom with school, problems with other students, and being retained a grade due to absenteeism were the top three reasons that led most of them to actually drop out.

Evidence from the surveys show "that dropouts have not acquired the cultural drives and the behavioral molds the school systems wish to develop in their students." So for example, "their boredom with school indicates that they have not internalized the competitive drives for individual self-maximization prevalent in the cultural atmosphere of the school" (p. 74). Parental support and encouragement, communication with and involvement in school activities, parental employment, and a two-parent home encouraged persistence.

Platero and his colleagues note that one of the most interesting aspects of Navajo dropout behavior was that they had not given up on schooling or education. They pointed out that "46 percent of all dropouts expect to return to school and graduate, while another 45.1 percent say 'maybe' when they are asked if they expect to return to school and graduate; only 8.8 percent have no hope or expectation of returning to school or graduating" (p. 84). This indicates that these dropouts did not discount the utility of ultimately obtaining a diploma, but felt unable to continue their education at this time.

Deyhle's (1989) study incorporated the reasons discovered earlier by Colodarci for AI/AN students' decisions to drop out of school and added six other possibilities in exploring the reasons that students might give for dropping out. These included difficulty with classes, difficulty with reading, work needs at home or job, distance from school, unwanted at school, and pregnancy. Dehyle's study included a breakdown of the percentage of A1/AN students questioned who indicated each identified reason for leaving school. The two most common problems identified by Navajo and Ute students in this study were problems at home, followed by difficulty with classes. This was followed by teachers not providing enough assistance with student work, difficulty with reading, lack of parental encouragement, teachers not caring, work needs at home or job, distance from school, which was tied with school not being perceived as important to life, having disagreements with teachers which was tied with feeling unwanted at school, pregnancy, and finally, school being perceived as unimportant to students as American Indians.

Dehyle noted that many of the students who left school felt unwanted there, leading many of them to reject school. Deyhle also discussed the relevance of racial tensions in a student's decision to leave school. Students reported their experience with hostility directed at them from their non-Indian peers as well as from some teachers and community members. The effects of peer and community pressure within AI/AN society are also identified as factors influencing the dropout's decision. This study indicated that students who do well in school often face ridicule from members of the community in the face of their apparent allempts at success in the "white" world.

Other researchers during the 1980s have completed smaller scale research on the reasons that AI/AN students gave for leaving school, as well as the correlates identified for dropping out. Although their samples are smaller and their studies less extensive than the three major studies discussed previously, they still provide valuable insights into dropping out among AI/AN students. They are presented here in chronological order.

The urban Phoenix Union High School District provided the setting for Milone (1983) to survey 31 AI/AN dropouts. Milone asked them about their reasons for dropping out and their attitudes toward school. In follow-up interviews with 15 of the students, she found students to have surprisingly positive attitude toward school. For some, the decision to leave school was involuntary in that they were dropped from the rolls for poor attendance. Other factors such as academic or discipline problems and pregnancy appeared to "push" them out. Many stuudents regretted the decision to leave school and indicated that they would like to return to school, but did not know how. Milone recommended that personal contact from the school would have helped many of the students who dropped out or were contemplating it.

In a case study, Chan and Osthimer (1983) investigated the educational experience of 24 Navajo adolescents in the areas of language minority status, traditionalism, critical markers, distance to school, and future orientation. Thirteen males and 11 females represented six high school dropouts, nine high school seniors with no future educational plans, and 9 high school seniors who were college-bound. Their findings indicated that "the most successful students (college bound graduates) were for the most part Navajo/bilinguals" (p. 34). Traditionalism did not appear to be a negative factor. Given the absence of historical records to conduct a careful analysis, critical markers that could serve as early warning signals of potential dropouts could not be identified. "All students residing on the reservation, regardless of school outcome, reported distance to school and transportation as a negative incentive to go to school. Travel time interfered with school activities, homework time, and recreation time" (p. 35). In terms of future orientation, "Dropouts were found to be quite vague in their future plans and goals, while graduate (sic) and college-bound students had relatively concrete plans and goals for the future" (p. 35).

Giles (1985) explored the cultural factors which affected the decision of AI/AN students in Milwaukee Public schools to remain in or drop out of school. Data were collected by reviewing high school records, looking at participants' neighborhoods and homes, observing participants in their classrooms, and interviewing (when possible) participants' parents, peers, siblings, and teachers. Through eight case studies of four dropouts and four enrolled high school students, Giles found that the school "is failing to meet the needs of Indians. There is widespread dissatisfaction with attempts to remedy the situation and public schools are not preparing students for society" (p. 2).

Following four cohorts of urban AI/AN students, Eberhard (1989) completed a study identifying several phenomena correlated to dropping out. One of the most startling was that among students who were retained between grade levels there was an 88% dropout rate. Dropouts moved twice as much as stayers, but parent status (one or two parents in the home) had no effect on dropping out; however, the trend was for stayers to come from two-parent families. AI/AN students and parents had a high regard for education and preferred math and English. AI/AN stayers were academically equal to other urban district stayers; GPAs and other measures of achievement were higher for stayers than dropouts. AI/AN parents and students rated urban schools higher than reservation schools while expressing the difficulties of making a successful transition from one to another. In this study, Navajo students dropped out less than all other tribes.

As part of the NEA study (Swisher et al., 1991) on AI/AN dropouts, AI/AN dropout information was solicited from the 20 states with the highest Al/AN populations. Most of the states also reported the reasons for dropping out obtained from or assigned to students, but did not separate them out by the ethnicity of the students. The notable exception was New Mexico, the only state which presented the reasons for dropping out specific to AI/AN students. The state of New Mexico, in its report New Mexico Dropout Study: School Year 1988-89 provided a list of the reasons for dropping out by gender and ethnicity. Information was "generally provided by the student" (Ulrich, & Borgrink, 1990, p. 14). The following were the reasons listed for AI/AN students:

State of New Mexico 1988-89 AI/AN Reasons for Dropping Out by Gender, Grades 7-12

According to the information sent by the state, the most common reason for dropping out among AI/AN students (when males and females were combined) was expulsion (removed from school by district), followed by lack of interest, transferring and, for girls pregnancy, for boys adjustment problems. For Asian, White, and Black students the four most common reasons were transfer, lack of interest, failure to re-enroll, and leaving to take the GED. For Hispanic students the four most common reasons were transfer, lack of interest, failure to re-enroll and suspension. Apparently, in New Mexico AI/AN students are more likely than any other ethnic group to become dropouts because, for one reason or another, the district removes them from school, a fact that warrants further investigation given that AI/ANs also have the highest dropout rate in the state.

The recent BIA-sponsored study to collect data on dropping out within its education system (Center for Indian Education, Arizona State University, in press) solicited answers in written questionnaires and over the telephone from school personnel as to what they thought contributed to the student dropout condition within BIA schools. Their responses covered a wide range of possible factors, some pointing to deficiencies within the students and their families, others blaming deficiencies within the educational system. They included:

An acknowledged limitation to the information gathered in this study was that it came only from school personnel. Had students and their parents been questioned, they might have shed light on an entirely different set of contributing factors to dropping out, based on their own experience and viewpoints. The answers that school staff provided, however, did shed light on systemic deficiencies which in their opinion affect the dropout rate.

Twenty-two years ago, The Status of American Indian Education, an interim report of the National Study of American Indian Education, stated that the transfer rate or mobility of AI/AN students was a concern (Auerbach et al., 1970). Today, it is still a concern and for the same reasons; it precludes "accurate follow-up on pupils in the absence of a centralized data collection system for Indian youth." . . . (p. 53). In addition, students who transfer frequently must constantly readjust to new schools which may or may not have complete educational records for that student to ease the academic transition. Transferring often disrupts the continuity of a student's education.

In some arm of Indian Country, transferring is a bigger problem than dropping out. On one northern plains reservation, the student transfer rate is higher than the dropout rate and it is believed to negatively affect the achievement rate. The author of a study (yet to be published) stated in a letter to the tribal council for a northern plains reservation:

On this same reservation, a principal of a Day School indicated that

In an American Indian Education Survey of Oregon City Public Schools, which addressed the major problems that hinder public education for AI/AN students, it was concluded that mobility of the family decidedly plays a negative role in student achievement.

A recent OIEP (1988) report provided several accounts of high transfer rates in several BIA schools. For example, at Wingate Elementary in New Mexico, 25% of the student population transferred to a public school during a single school year, while 20% of the student population transferred into the school from the public school system. Chemawa High School, in Oregon, reported that it lost 50% of its student population annually to transferring. In addition, the principal stated at the time of the report that "78% of the total Chemawa student body . . . were transfers from or have attended another program" during the same academic year (p. 56-57). The high rate of mobility apparently created difficulties not only in the educational deficiencies that it fostered (or maintained), but also in attempting to provide continuity in the student's education.

More recently, Latham's (1989) report of BIA education programs on 17 reservations in eight states cited a mid-year transfer rate in BIA schools ranging from 10% in the elementary grades to upwards of 30% in high schools. He also found that transfers typically have the most academic and behavioral problems. Conversely, academically successful students come from stable homes and tend to enroll and stay in public schools. The inconsistency that multiple transfers create in a student's academic experience is compounded in the BIA school system by the fact that students seem to depend on BIA schools as a sort of academic safety net.

The previously mentioned study of dropout and transfer rates within the BIA school system (Center for Indian Education, Arizona State University, in press) produced several statistics regarding the issue of transferring in BIA schools. The study also included statements from school staff about the cause and effects of what was identified as extensive transferring within the BIA system. The following figures for the transfer rate are reproduced from that report in Figure 1.

According to the qualitative data gathered for the BIA study, transferring into and out of school throughout the school year was a serious problem. Staff noted that students transferring into the school at various times throughout the school year disrupted classroom routine as another student and his or her needs had to be fit into what had already been established. The transferring students themselves also suffered, attempting to catch up and adjust to a new school and make up for deficiencies in curricular experience. Schools that must accommodate mid-year transfers may also be unprepared to meet any special needs of the student, further straining both the school and the transferring student. In addition, students that transfer out of school mid-year often fail to return books and other school-owned supplies, straining the budget of financially strapped schools.

In questionnaires and phone interviews, BIA school personnel identified several factors which aggravate the transfer problem. The staff at the BIA schools point to the lack of policy regarding transfers as a key contributor to the transfer problem. Transferring is not well regulated, and guidelines for transferring students smoothly from school to school do not exist. School personnel also comment that the "count week", which determines how many students each school serves, exacerbates the transfer problem. Many schools actively entice students into attending through count week, and then ignore them once funding for them is secured. These neglected students are then likely to transfer, many from the public to the BIA system, which strains the resources of BIA schools. In addition, the funds that the first school has secured do not follow the transfer student, making things even more difficult for the school to which the student transfers.

Although considerable research documenting high rates of transferring among AI/AN students exists, the research does not go to any great length to investigate its relationship to academic achievement or dropping out. Research does concur that the high rate of student transfer places a heavy burden upon the school for providing consistency and stability to its students (Hopkins & Reedy, 1978; OIEP, 1988). However, reports regarding the incidence of transferring conflict when comparing students who drop out with students who complete school.

Owens and Bass (1969) found that although a slightly higher percent of dropouts than graduates had transferred more than once in their high school careers, the difference was negligible. Eberhard's (1988) work noted that there was more transferring among dropouts than among graduates. In contrast, Platero et al. (1986) indicated that among the Navajo students that they studied, the graduates evidenced more transferring from school to school than did the dropouts.

The question remains whether the act of transferring from school to school is an adaptive coping skill that improves a student's chances at eventually graduating, or an act of denial troubled students utilize to avoid addressing their problems in any one place. The answer to this question is pivotal when examining the educational achievement of Al/AN students, a population statistically at a higher risk for dropping out. Educators need to know whether transferring should be curbed as a disruption or maintained as a way for students to find the place where they can best pursue an education. More extensive research is needed to precisely determine the incidence of transferring within the AI/AN population and how it affects educational achievement or dropout rates.

Many studies over the past three decades have generated concerns with respect to the systems educating AI/AN people. Research consistently indicates that AI/AN students drop out of school at one of the highest rates of any ethnic population in the country. The precise statistics, particularly the establishment of a national dropout rate that includes all of the school systems which educate Al/AN students, would help to illustrate the extent of the problem and guide efforts to reduce it. To aid in this effort, research identifying the factors that influence the student's decision to drop out and the associated correlates for dropping out would enable more effective interventions to be designed and implemented in the populations most at risk. Many small pieces of the research picture have already been put together, but we still await the creation of a coherent and cohesive whole to guide us in our efforts to reduce the incidence of dropping out in the AI/AN community.

New interest has evolved in the area of student transfer, both from school to school and from system to system. There is agreement that an inordinate amount of transferring occurs among Al/AN students, but only speculation as to what causes this phenomena. A more complete understanding of why students transfer in the first place would allow educators to more effectively address this issue. Although agreement that transferring is disruptive to both the student and the educational system can be reached, reliable research into whether transferring improves or hinders a student's chance to eventually graduate from school has yet to be conducted.

Researchers in the past have laid the groundwork to add to our understanding of the educational experience of AI/AN students. What is needed now is an effort to standardize practices and produce research that illustrates the educational outcomes of AI/AN students nationwide. In addition more local research that adheres to a standard that allows comparisons from one region to another is necessary. This seems to be a key in creating an educational system that effectively serves AI/AN students.

Auerbach, H.A., Fuchs, E., & MacGregor, G. (1970). The status of American Indian education. An interim report of the national study of American Indian education to the Office of Education, U.S. Department of Health, Education, and Welfare. University Park, PA: Pennsylvania State University. (ERIC Document Reproductive Services No. ED 039 055).

Center for Indian Education, Arizona State University. (in press). American Indian/Alaska Native membership study: An examination of dropout and transfer rates. Washington D.C.: Bureau of Indian Affairs, Office of Indian Education Projects.

Chan, K.S., & Osthimer, B. (1983). Navajo youth and early school withdrawal: A case study. Los Alamitos, CA: National Center for Bilingual Research. (ERIC Document Reproductive Services No. ED 240 414).

Colodarci, T. (1983). High school dropout among Native Americans. Journal of American Indian Education, 23(l), 15-22.

Coombs, M. (1970). The Indian student is not low man on the totem pole. Journal of American Indian Education, 9(3), 1-9.

Deyhle, D. (1989). Pushouts and pullouts: Navajo and Ute school leavers. Journal of Navajo Education, 6(2), 36-51.

Eberhard, D. (1989). American Indian education: A study of dropouts. Journal of American Indian Education, 29(l), 32-40.

Eberhard, D.R. (1988). Urban American Indian education: A study of dropouts. Denver, Co: University of Colorado.

Giles, K.N. (1985). Indian high school dropout: A perspective. Milwaukee, WI: Midwest National Origin Desegregation Assistance Center: The University of Wisconsin. (ERIC Document Reproduction Services No. ED 270-282).

Hopkins, T.R., & Reedy, R.L. (1978). Schooling and the American Indian high school student. BIA Education Research Bulletin, 6(2), 5-12.

Horton, J.M., & Annalora, D.J. (1974). Student dropout study of Fort Wingate High School. BIA Education Research Bulletin, 2(2), 17-25.

Kutsche, P. (1964). Cherokee high school dropouts. Journal of American Indian Education, 3(2), 22-30.

Latham, G. (1985). The educational status of federally recognized Indian students. Journal of American Indian Education, 25(l), 25-33.

Latham, G.I. (1989). Thirteen most common needs of American education in BIA schools. Journal of American Indian Education, 29(l), 1 -11.

LeCompte, M.D., & Goebel, S.D. (1987). Can bad data produce good program planning? An analysis of record-keeping on school dropouts. Education and Urban Society, 19(3), 250-258.

Milone, D.E. (1983). American Indian student reasons for dropping out and attitude toward school. Tempe, AZ: Arizona State University, Unpublished Masters Thesis.

Morrow, G. (1987). Standardizing practice in the analysis of school dropouts. In G. Natriello (Ed.), School dropouts: Patterns and policies, (pp. 38-51). New York, NY: Teachers College Press.

National Center for Education Statistics (1991). Dropout rates in the United States: 1990. Washington, DC: U.S. Department of Education, Office of Educational Research and Improvement.

Office of Indian Education Programs. (1988). Report on BIA education: Excellence in Indian education through effective schools process. Final review draft. Washington, DC: Office of Indian Education Programs, Bureau of Indian Affairs, U.S. Department of the Interior.

Owens, C.S., & Bass, W.P. (1969). The American Indian dropout in the Southwest. Albuquerque, NM: Southwestern Cooperative Education Laboratory, Inc.

Platero, P.R., Brandt, E.A., Witherspoon, G., & Wong, P. (1986). Navajo students at risk. Final report for the Navajo area student dropout study. Window Rock, AZ: Platero Paperwork Inc.

Selinger, A.D. (1968). The American Indian high school dropout: The Magnitude of the problem. Portland, OR: Northwest Regional Educational Laboratory.

State of Minnesota. (1988). It’s time to break the trend! Annual dropout percentage rate for American Indian and White students in selected urban districts. Minneapolis, MN: MINICRIS data file.

Swisher, K., Hoisch, M., & Pavel, D.M. (1991). American Indian/Alaskan Native dropout study, 1991. Washington, DC: National Education Association.

Ulrich, A.L., & Borgrink, H. (1990). New Mexico dropout study 1988-89 school year. Santa Fe, NM: Assessment, Evaluation and Information Services, State Department of Education.

Wax, R. (1967). Oglala Sioux dropouts and their problems with educators.. In Everett T Keach et al. (Eds.), Education and school crisis: Perspectives on teaching disadvantaged youth (pp. 247-257). New York, NY: Wiley & Sons.

Wax, R.H. (1967a). The warrior dropout. TransAction, 4, 40-46.

Wells, R.N. (199 1). Indian education from the tribal perspective: A survey of American Indian tribal leaders. Canton, NY: St. Lawrence University. Available from the author.