Rural Students at Risk in Arkansas, Louisiana, New Mexico, Oklahoma and Texas
Background Characteristics
Research identifies a number of variables related to a student's family or personal background that appears to contribute to increasing the risk of failure in school. The following are among the most often cited factors.
- Single Head of Household
- Low Socioeconomic Status (SES)
- Minority Group Status
- Limited English Proficiency
- Low Educational Attainment of Parents
- Mobility
- Psychosocial Factors
- Gender
Single Head of Household
In 1955, 60% of all U.S. households consisted of a working father, a housewife mother, and two or more school-age children. In 1985, only 7% fit this pattern. In addition, with over one-half of all today's new marriages slated to end in divorce, we have 15.3 million children living with one parent, the mother in over 90% of the cases. (Davis & McCaul, 1991, p. 22)
From the High School and Beyond research and elsewhere, it is apparent that students from single-parent households tend to drop out of school at a much higher rate than those students who come from the more traditional two-parent family structure (Ferguson, 1992; Bull, Salyer, Montgomery, & Hyle, 1992; McLaughlin, 1990; Wehlage et al., 1989).
According to the Kids Count Data Book (1993), 25 percent of all children now live in single-parent families. This represents a 9 percent increase from just five years earlier. "In 1990, over one-half of African-American children lived in single-parent families; just under one-third of Hispanic children lived in single-parent families" (p. 14).
As stated previously, more than 90 percent of these one-parent families are headed by females. The majority of female-headed households are either the result of divorce or out-of-wedlock births. There is, therefore, a significant correlation between these households and poverty (Davis & McCaul, 1991; Willis, 1987). In fact, children who live in female-headed families have a greater than 50 percent chance of being poor (Ascher & Burnett, 1993). In a 1986 study, it was found that 62 percent of one-parent families (predominantly female) had annual incomes of less than $10,000 (Keough, 1986).
Paying for a child's needs on a single salary is also hard, particularly for women. The average income of single-mother families is 60 percent of single-father families' income. Only 31 percent of mother-headed households receive any child support or alimony. (Kids Count Data Book, 1993, p. 14)
Although the numbers of rural, one-parent households do not begin to approach those in urban areas, the nonmetropolitan family structure is changing rapidly, and this also has implications for the rural at-risk student problem. The incidence of rural female-headed households has increased from 9 percent in 1960 to 20 percent in 1990 (Lichter & Eggebeen, 1992). Further, according to the General Accounting Office (1994), the rate of poor rural children in female-headed families increased faster than similarly situated urban children during the decade of the 1980's. That office provides the following information from the 1990 census:
School-Age Children in Single, Female-Headed Households| 1980 | 1990 | |
|---|---|---|
| Urban | 6,788,605 | 7,274,565 |
| Rural | 1,635,823 | 1,876,503 |
The percent of children living in single-parent households varies significantly among the five states of the Southwestern Region. Yet, all have seen an increase in this family structure from 1985 to 1990. No data were found relative to single parent families specifically in the rural areas of these five states. However, the following overall information is provided by the Kids Count Data Book (1993).
Children in Single Parent Families| Percentage | 1990 Rank | ||
|---|---|---|---|
| State | 1958 | 1990 | (50 States + Washington, DC) |
| Arkansas | 23.3 | 27.2 | 41st |
| Louisiana | 24.7 | 31.9 | 49th |
| New Mexico | 21.7 | 22.6 | 26th |
| Oklahoma | 20.4 | 22.5 | 24th |
| Texas | 20.7 | 22.5 | 24th |
| United States | 22.7 | 24.7 | |
Low Socioeconomic Status (SES)
"It is estimated that of the poor in America, 40% are children" (Davis & McCaul, 1991, p. 21). Approximately 7.6 million school-age children, more than 17 percent of the total student population, live in poverty (General Accounting Office, 1993). Much of the professional literature indicts poverty as a primary factor placing students at high risk of not graduating from high school (e.g., McCaul, Donaldson, Coladarci, & Davis, 1992; Sherman, 1992; Presseisen, 1991; Pallas, 1990). "Students from low-income families are three times as likely to drop out of school as those from more affluent homes" (Kids Count Data Book, 1993, p. 11). Further, female students who come from families in the lowest SES quartile drop out of school at five times the rate of females from the highest quartile. Male students in the lowest quartile drop out at two and a half times the rate of those in the highest quartile (Earle & Roach, 1987). Consequently, the Goals 2000: Educate America Act of 1994 identifies poverty and economic disadvantage as significant at-risk factors.
Ellwood (1988) suggests that nearly half of all poverty exists in nonmetropolitan areas. Estimates suggest the national poverty rate in 1990 to have been at 13.5 percent, with the rural poverty rate at 16.3 percent and the inner city rate at 19 percent (Lichter & Eggebeen, 1992; O'Hare & Curry-White, 1992; Marion, 1992). The General Accounting Office (1993) reports a rural poverty rate of 15.6 percent and an urban poverty rate of 17.6 percent (p. 38).
Clearly, the poverty rate is highest in central cities. However, the largest number of poor people live in rural areas, small towns, and small metropolitan areas. Almost one in three urban children and one in four rural children live in families whose incomes are below the poverty level" (Davis & McCaul, 1991, p. 40).
The poverty problem has also grown for those in rural areas and small communities. According to Alexander (1990), "By the late 1980's, the rural poverty rate...was growing twice as fast as urban poverty" (p. 123). Further, "White children, African American children, children in married-couple families and children in mother-only families are all more likely to be in poverty if they live in a rural area" (Hodgkinson, 1994).
Data relative to the numbers and rates of poor urban and rural school-age children has been made available by the General Accounting Office (1994) from the 1990 census data:
Poor School-Age Children| 1980 Number |
1980 Rate |
1990 Number |
1990 Rate |
|
|---|---|---|---|---|
| Urban | 2,922,623 | 14.3 | 3,188,758 | 16.0 |
| Rural | 806,757 | 18.6 | 1,015,987 | 20.4 |
These data suggest that approximately 75% of all school-age poverty is located in metropolitan settings, while the remaining 25% are rural. The General Accounting Office (1993) has further analyzed this information by ethnicity. They report the following:
School-Age Poverty by Location and Ethnicity| Geographic Area | Percent | (Further broken down by ethnicity) White + Minority = 100%) |
|---|---|---|
| Urban | 75.8 | (32.56% 67.44%) |
| Rural | 24.2 | (67.07% 32.93%) |
The poverty rates of school-age children, whether urban or rural, in the five states of the Southwestern Region are among the most severe in the United States. Further, these rates have increased in each of these states, dramatically in some cases, during the decade of the 1980's.
School-Age Poverty Rates| Rate | 1990 Rank | Total (1990) # of | State | 1980 | 1990 | (50 States) | Poor Children |
|---|---|---|---|---|
| Arkansas | 22.8 | 23.8 | 46th | 107,170 |
| Louisiana | 23.2 | 30.4 | 49th | 267,555 |
| New Mexico | 21.8 | 26.4 | 48th | 82,984 |
| Oklahoma | 15.2 | 19.9 | 40th | 120,018 |
| Texas | 18.5 | 23.4 | 45th | 794,774 |
| United States | 15.3 | 17.1 | 7,571,259 | |
| State | Rate | *Rank |
|---|---|---|
| Arkansas | 26.5 | 41st |
| Louisiana | 36.2 | 49th |
| New Mexico | 30.5 | 47th |
| Oklahoma | 24.4 | 39th |
| Texas | 28.2 | 45th |
*The rank is based on 49 states because New Jersey contains no non-metropolitan counties.
(General Accounting Office, 1993)
From the above information it is apparent that the numbers of poor people are both substantial and growing. However, there are also people who are chronically poor. The term "underclass" has been used to describe persons who not only are poor, but have been firmly entrenched in poverty conditions for a long time (multi-generational) and in areas where poverty is concentrated. Students in these circumstances are surely at quite high risk. Although researchers primarily view the concept of an underclass as an urban phenomenon, O'Hare and Curry-White (1992) analyzed 1990 census data and found a significant underclass in rural areas (2.4 percent in rural areas compared to 3.4 percent in central cities and 1.1 percent in suburban communities). They also found differences between the urban and rural underclass that have implications for the identification of at-risk children in rural areas. For example, 55 percent of the rural underclass are White compared with 17 percent in central cities, and while 49 percent of the urban underclass are African-American, 32 percent of the underclass in rural areas are African-American. They found insignificant differences between percentages of the urban and rural Hispanic underclass.
Sherman (1992) states that "two-fifths of all young rural children with single mothers live on family incomes of less than half the poverty level" (p. 42). He reports that the 1980 census found 28 counties nationwide with child poverty rates exceeding 50 percent. All 28 counties are characterized as rural. Many of them are located in the Southwestern Region--in the Mississippi Delta area of Louisiana and Arkansas and in Texas along the Mexican border.
In contrast to all that has been stated above regarding low socioeconomic status (SES) as an at-risk factor, conflicting research and other cautions have been proposed. Some research suggests that at-risk students may be under-identified if schools are overly dependent on poverty as a criteria. For instance, Bryk and Thum (1989) found a higher dropout rate in high SES high schools than would be expected based on assumptions about the relationship between low SES and at-risk status.
In a study of 2.4 million students in 900 Texas school districts, Ferguson (1991) found that poverty did not have a statistically significant effect on achievement (as measured by the state's minimum competency tests) when the variable "female head of household" was controlled. "In fact, the percentage of children living in poverty is highly statistically significant only when both female headship and students' race variables are omitted; its measured significance becomes very marginal when either of these is added" (p. 479). This finding held true for kindergarten through the ninth grade. Additional research is warranted before any conclusions can be drawn about such findings, but they suggest that there is no simple causal relationship between family income and achievement. Until such research is forthcoming, it is, perhaps, wise to continue to consider children from poorer families at higher risk than those of more affluent ones.
Minority Group Status
Rumberger (1987) notes that Afro-American, Hispanic, and AI/AN [American Indian/Alaska Native] populations, disproportionately represented among dropouts, are increasing at a faster rate than the White majority. This is a critical problem when one considers that in many areas, these populations represent the majority of school-age children and will be the majority in the very near future. (National Education Association, 1991, p. 8)
In general, the dropout rate is higher for minority students (Baumeister, Kupstas, & Klindworth, 1990). Minority group status is, therefore, frequently listed as a risk factor in the literature (e.g., Presseisen, 1991; Davis & McCaul, 1991; Pallas, 1990). Membership in some minority groups, however, does not appear to increase risk (e.g., Asian-American). Further, although minority dropout rates may create an impression that the at-risk problem is essentially a minority problem, the total number of White dropouts is substantially greater than the total of all other dropouts combined (Wehlage et al., 1989). Most importantly, the use of minority group status to identify at-risk students may mask information that could be critical for developing effective responses. Ferguson (1991) suggests that race variables are "stand-ins for factors that are correlated with race but not otherwise represented ... (e.g., peer culture, ethnic idiosyncrasies in grammar)" (p. 479). For example, in his study (described earlier), he found that African-American students did not make significantly lower standardized test scores until the ninth grade. This phenomenon, according to Maeroff (1988) may be influenced significantly by peer pressure, from a strong African-American teen culture, to resist achievement. African-Americans do, however, consistently drop out of school at higher rates than their White counterparts. Certainly, additional research is warranted.
The data on Hispanic at-risk students provides further proof that overdependence on identification by ethnic group is problematic. On the surface, research studies appear to show that, as a group, Hispanic students are at a higher risk of dropping out and/or of lower achievement. An analysis of the High School and Beyond database showed that for the 1980 cohort of high school sophomores, dropouts were more likely to be Hispanic (29.8 percent) than African-American (22.9 percent) or White (20.0 percent) (McCaul et al., 1992). Ferguson (1991) reports similar findings from his analysis of Texas data. According to 1990 census data, 50 percent of all Hispanic adults report not graduating from high school (Census paints portrait, 1993).
McLaughlin (1990) suggests that policy makers should differentiate between native-born Hispanics and recent immigrants. Using census data, he showed that over one-third of the Hispanic population are immigrants (and thus probably are more limited English proficient than their non-immigrant counterparts) and that native-born Hispanics have educational attainment rates slightly less but comparable to native-born Whites. Although McLaughlin's data left many unanswered questions, his position emphasizes the need for appropriate and specific information about why students are at risk before effective programs and policies can be initiated.
Native Americans (both American Indians and Alaska Natives) consistently have the highest dropout rates nationally of all ethnic or racial groups (Reyhner, 1991; Borgrink, H., 1987). The National Education Association (1991) compiled a comprehensive report on Native American dropouts. This report highlights the difficulties involved in determining an accurate national dropout rate for American Indians and Alaska Natives, including:
- no standardized definition of who Native Americans are (though the Bureau of Indian Affairs uses the guidelines of 1/4 "degree of blood" and an official membership in an accredited tribe). Most state education agencies accept self-reports as adequate documentation;
- no standardized operational definitions of who "dropouts" are; and
- no standardization about how dropouts are counted and reported.
Even so, the report documents several facts about the Native American population.
According to Hodgkinson, Outtz, and Obarakpor (1990), in The Demographics of American Indians: One Percent of the People; Fifty Percent of the Diversity, there are approximately 1.7 million persons who trace their lineage to over 500 tribes and native groups (more recent estimates proclaim that there are over two million). Sixty percent of the American Indian and/or Alaskan Native (AI/AN) population are members of ten tribes. One fourth of all AI/AN live on reservations. Over 300,000 Indians live in metropolitan areas... The majority of [American Indian] students (85%) attend public schools, ten percent attend schools funded by the Bureau of Indian Affairs (BIA), and five percent attend mission or private schools. (p. 1)
Reyhner (1991) estimates the Native American dropout rate to be about 30 percent. However, the National Education Association report (1991) suggests that dropout rate estimates vary significantly. "Recent estimates of the problem include dropout rates from about 35.5 percent (NCES, 1988) to over 50 percent (Wells, 1991) and in undocumented cases between 80 and 90 percent" (p. 2).
Data describing the numbers of rural school-age children in the United States, broken down by ethnicity, are provided by the General Accounting Office (1994):
National Data on Rural School-Age Children by Ethnicity| Race/Ethnicity | 1980 | 1990 | Percent Change |
|---|---|---|---|
| White | 9,504,317 | 8,691,783 | -8.6 |
| Hispanic | 479,469 | 557,080 | 16.2 |
| African- American |
1,267,696 | 1,162,640 | -8.3 |
| Asian | 62,184 | 89,399 | 43.8 |
| Native American |
215,518 | 250,819 | 16.4 |
| Others | 7,269 | 7,181 | -1.2 |
| Total | 11,536,453 | 10,758,902 | -6.7 |
In the five states of the Southwestern Region, information about minority children is provided by the Kids Count Data Book (1993).
Minority-Status Children Under 18 as Percent of Total| State | 1990 |
|---|---|
| Arkansas | 24.3 |
| Louisiana | 41.5 |
| New Mexico | 59.9 |
| Oklahoma | 25.6 |
| Texas | 48.9 |
Specific data about ethnicity in the five states of the Southwestern Region were provided for the 1991-1992 school year by the respective state education agencies.
Student Ethnicity| State | White | African American |
Hispanic | Native American |
Other |
|---|---|---|---|---|---|
| Arkansas | 75.6 | 22.7 | 1.1* | <1 | <1 |
| Louisiana | 52.9 | 44.5 | 1 | <1 | 1.1 |
| New Mexico | 41.2 | 2.3 | 45.3 | 10.4 | <1 |
| Oklahoma | 73.5 | 9.9 | 3 | 12.4 | 1.2 |
| Texas | 49 | 14 | 34 | NA** | 2 |
* "Hispanic" is not a distinct category in this Arkansas database. Hispanics may also be counted in other categories.
** Not Available--included in the "Other" category.
Poverty rates are generally higher for minorities, adding an additional at-risk factor for many. An analysis of the 1990 census data (General Accounting Office, 1993) reveals the following:
Poverty Rates by Ethnicity| Ethnicity | Urban | Rural | U.S. Total |
|---|---|---|---|
| Hispanic | 30.75 | 34.84 | 31.15 |
| African- American |
37.22 | 40.76 | 37.70 |
| White | 9.07 | 12.20 | 10.14 |
| Asian | 16.8 | 11.57 | 16.40 |
| Native Amer.
/Other |
29.84 | 59.59 | 34.36 |
A similar, though more dramatic, trend is seen in the rural areas of the five states comprising the Southwestern Region. A further analysis of the 1990 census data reveals the following information.
Percent of Poor, Rural, Minority, School-Age Children| State | Rate |
|---|---|
| Arkansas | 43.8 |
| Louisiana | 64.0 |
| New Mexico | 81.9 |
| Oklahoma | 39.6 |
| Texas | 68.8 |
The reader will note that, although the information previously cited about the percent of minority children in these five states is not the same database (all minority children under 18 versus school-age minority children living in rural areas), a comparison of these two sets of data reflects a significant influence of poverty among rural, minority-status students.
The ethnic percentages of poor school-age children, whether rural or urban, for the Southwestern Region are also derived from the same census database and provided by the General Accounting Office (1993).
School-Age Poverty by Ethnicity| State | White | African American |
Hispanic | Asian | Native Amer. /Other |
Total |
|---|---|---|---|---|---|---|
| Arkansas | 51.3 | 46.2 | <1 | <1.0 | <1.0 | 107,170 |
| Louisiana | 28.6 | 67.8 | 1.5 | 1.2 | <1.0 | 267,555 |
| New Mexico | 17.9 | 2.3 | 58.3 | <1.0 | 21.0 | 82,984 |
| Oklahoma | 56.7 | 18.7 | 6.6 | <1.0 | 17.3 | 120,018 |
| Texas | 20.9 | 21.5 | 56 | 1.2 | <1.0 | 794,774 |
| United States | 41.3 | 32.2 | 21.2 | 3.0 | 2.3 | 7,571,259 |
Limited English Proficiency
The Goals 2000: Educate America Act of 1994 identifies students who are limited English proficient as at significant risk. According to the General Accounting Office (1993), 5.21 percent of all school-age children are limited English proficient (LEP).
As would be expected, children who are not proficient in English face significant challenges in school. Whether they are in bilingual programs or in English-as-a-Second-Language (ESL) programs, some delay in achievement, at least in the early stages, can be expected. Pallas (1990) finds that children of limited English proficiency are also more likely to drop out of school than students who are proficient. He cites a 1987 study by Salganik and Celebuski revealing that students from homes where no English was spoken were twice as likely to drop out as students where English was the primary or only language spoken in the home.
The LEP student population is one that has realized significant growth. The number of LEP students between the ages of 5 and 14 in 1976 was estimated at 2.5 million and projected to grow to an estimated 2.9 million by 1990. Projections suggest that numbers will climb to 3.4 million by the year 2000 (Gingras & Careaga, 1989).
The number of LEP students in the five states of the Southwestern Region are provided by the General Accounting Office (1993).
| State | 1980 | 1990 | Percent of All Students (1990) |
|---|---|---|---|
| Arkansas | 2,309 | 4,142 | 0.9 |
| Louisiana | 16,663 | 16,508 | 1.9 |
| New Mexico | 47,796 | 33,074 | 10.5 |
| Oklahoma | 7,791 | 9,172 | 2.5 |
| Texas | 407,715 | 383,572 | 11.3 |
Low Educational Attainment of Parents
Generally, the potential for being at risk is higher for students whose parents either dropped out or had fewer years of post-secondary schooling. The research suggests that the educational attainment of parents, especially mothers, has a positive and independent effect on student achievement (Smith et al., 1992; Pallas, 1990). Further, the correlation between the level of a mother's educational attainment and family socioeconomic status is such that some research has used both students' enrollment in the free/reduced school lunch programs of their schools and mothers' educational attainment level to define low socioeconomic status for their research (e.g., Griffin, 1992).
Ferguson (1991) found that nearly one-fourth of the variation on Texas' minimum competency test in 1986 was correlated positively with parent educational attainment. An analysis of the High School and Beyond database showed that for those sophomores in 1980 with mothers who had not completed high school, nearly 25 percent dropped out by 1982. Further, the entire High School and Beyond database (including 1984 and 1986 information) revealed that students whose mothers did not graduate from high school were almost twice as likely to drop out as those whose parents did graduate and more than three times as likely to drop out as those whose parents graduated from college (Barro and Kolstad, 1987). Although the average years of formal education has increased substantially since 1960 for rural adults, educational attainment in nonmetropolitan areas still lags behind that in metropolitan areas (Lichter & Eggebeen, 1992).
Earning power and educational attainment are highly correlated. The salaries of those with higher educational attainment are generally greater than those whose educational attainment levels are lower. Consequently, poverty rates are skewed toward those with lesser educational attainment. The following information highlights the number of poor school-age children by the educational attainment of their more educated parent (General Accounting Office, 1994):
School-Age Poverty by Level of Parental Education Attainment| Education status of the more educated parent |
Poor Urban |
Poor Rural |
All Nonpoor |
|---|---|---|---|
| Grade school or less | 754,556 | 228,393 | 996,876 |
| Some high school | 1,502,413 | 571,015 | 2,401,794 |
| High school graduate | 1,490,585 | 753,407 | 9,411,221 |
| Some college/AA degree | 1,084,045 | 438,413 | 12,281,223 |
| BA or more | 265,662 | 90,515 | 10,968,082 |
| Total | 5,097,261 | 2,081,743 | 36,059,196 |
Mobility
"Mobility can foster another kind of instability only rarely discussed in the at-risk literature. Even with a supportive family, students can experience serious disorientation after moving away from a community of peers who provided social identity" (Wehlage et al., 1989). No studies were found to point clearly to mobility, in and of itself, as a significant factor placing students at risk. Rather, mobility is suggested as an additional factor accompanying other at-risk characteristics (Clark, 1991). It appears that some children cope adequately with all the changes that accompany moving while others have more difficulty adjusting.
Bezruczko and Reynolds (1992), from the most recent data of their longitudinal study of Chicago at-risk students, suggest that mobility plays a strong role in grade retention. However, inconsistent data lead them to speculate that mobility may have a stronger negative impact on younger students, but fade as a significant factor in school achievement by the fourth grade (the current grade placement of the students they began tracking in kindergarten). This inconsistency will be studied further as the longitudinal study progresses.
One subgroup affected by mobility are migrant students. The number of migrant students is difficult to determine. According to Sherman (1992), there are more than 780,000 migrant children of school-age, 80 percent of whom are Hispanic. Further, "migrant children are presumed to live mostly in rural areas and to be overwhelmingly poor" (p. 16).
The dropout rate of migrant students is also difficult to calculate, however it is estimated to be about 50% (Salerno & Fink, 1989). "Rural migrant students experience higher rates of the following: child abuse...depression/suicide/low self-esteem, poverty, illiterate parents...teen pregnancy" (Helge, 1991, p. 2). Frequently, migrant students come from families typified by many at-risk characteristics, including poverty, limited English proficiency, minority status, and low parental education attainment. Other factors that may have impact upon migrant families include inadequate prenatal care, poor nutrition, and minimal (or nonexistent) health and social services (Helge, 1991). Still other factors include overage grade placement, gaps in school attendance and inconsistent school record keeping (Salerno, 1991).
The seasonal moves that define the lives of migrant families make it difficult for their children to maintain consistency in academic content or in curricular scope and sequence within content areas. Federal efforts to alleviate some of the unique problems confronting migrant students, including a standardized record keeping system--the Migrant Student Record Transfer System (MSRTS), appear to have helped in limited ways (Helge, 1991). The MSRTS is a service that is contracted by the federal government. The database is currently housed at the Arkansas Department of Education in Little Rock and maintains information on migrant students in 49 states, Washington, DC, and Puerto Rico. Currently, about 30 percent of these school districts utilize the MSRTS (Cahape, 1993).
A second subcategory of mobility is homelessness. The number of homeless families is difficult to assess. However, the numbers appear to be increasing (Klauke, 1989; Shaver & Dornbusch, 1993). Families that have children make up 34 percent of the nation's homeless. One-fourth of those in homeless shelters are people from rural areas (Harrington-Luecker, 1989). Of course, only a fraction of the homeless are housed, even temporarily, in shelters.
Homeless children face unique issues that mitigate against consistent attendance in school. Beyond the obvious concerns about meeting basic needs (physical needs, safety needs, etc.), "problems such as unstable living conditions, unemployment, family problems, and health problems make homeless children particularly at-risk of school failure" (Shaver & Dornbusch, 1993, p. 2). Fatigue, caring for younger siblings, lack of any place to do homework, and other factors impinge upon homeless children to be successful in school (Shaver & Dornbusch, 1993).
One major problem confronting homeless children from attending school on a regular basis is transportation. It is difficult to get to school and back without the availability of personal or free public transportation. Another problem has been the inability to gain access to school. In the past, schools have required proof of residency within school boundary lines before a child was allowed to attend. Leases, electric bill statements, or similar documentation have been usually accepted as proof of such residency. Homeless families are unable to furnish this kind of documentation. Therefore, their children have been barred from enrolling and attending (Harrington-Lueker, D., 1989). However, this action to refuse enrollment has been reduced significantly through the enactment of the 1989 amendment to the Stewart B. McKinney Homeless Assistance Act of 1986 (originally PL 100-77 and amended by PL 101-645), which requires schools to accommodate homeless students.
The McKinney Act also mandates that each state department of education designate a coordinator for the education of homeless children and youth. Among the responsibilities of these coordinators is the task of determining a headcount of homeless children and youth. Because of the inherent difficulties in counting these people and because no clear guidelines exist to standardize how these counts are conducted, the results are, at best, estimates. Some states survey all school districts in their state and accept whatever numbers are returned. Others perform a one-night head count of all known homeless shelters in their state and extrapolate an estimated yearly count. Some states include only those school districts in which Stewart B. McKinney funded programs for homeless students are in existence. Still other states use various combinations of these or other strategies to arrive at their respective homeless headcounts. State agency personnel charged with determining these numbers readily admit that the resulting counts are at best general estimates and are most probably below the actual numbers. With these caveats in mind, the following is a list of the most recent counts compiled by the five state education agencies and submitted to the U.S. Department of Education in December, 1993.
Homeless School-Age Children| State | Annual Headcount (1993) |
|---|---|
| Arkansas | 5,400 |
| Louisiana | 13,000* |
| New Mexico | 2,836** |
| Oklahoma | 4,017 |
| Texas | 123,738 |
* This count is a rounded off figure.
** May be revised upward based upon updated data from New Mexico/Mexico border school districts.
Psychosocial Factors
Personal and family circumstances provide insight into the reasons some students experience difficulty in learning, and research shows that personal and family problems significantly affect school achievement. In a meta-analysis of correlational research into risk factors, Frymier & Robertson (1990) found that family instability, family tragedy, and personal pain were significantly associated with the risk of failing in school. Others report similar personal characteristics and experiences to be related to dropping out (e.g., Clark, 1991; Wehlage et al., 1989; Ediger, 1987; Catterall, 1986). Among these factors were limited experience in the mainstream culture, family crises (e.g., divorce, death, parent's loss of employment), and feelings of alienation or exclusion.
Rural education literature and research also point to some of these variables as significant factors having impact on students in rural areas and in small schools. A survey of rural educators (Bull et al., 1992) suggests that such factors as low self-esteem, depression, suicidal tendencies, unstable family environment or families in crisis, victims of child abuse, and alienation from school are among their most significant concerns about their students. On the other hand, in regard to student alienation and exclusion, Ornstein (1989) suggests that smaller schools afford students with increased opportunities
to participate in extracurricular activities, especially [sic] the high-status ones such as student government, student newspaper, school band, and athletics....The socio-psychological benefits of recognition and affiliation, and the result in terms of students' self-concept and motivation for achievement, are well documented..." (p. 156)
Gender
Some researchers suggest that males are at higher risk of dropping out than females (e.g., Wehlage et al., 1989; Presseisen, 1988, Hahn & Danzberger, 1987). Others have found no dropout rate differences between males and females (McCaul et al., 1992; Hahn, 1987; McDill, Natriello, and Pallas, 1986; Weidman, 1984). Bryk and Thum (1989) found that females are dropping out "at a somewhat higher rate than expected given their social class and at-riskness behaviors" (p. 18). According to Sherman (1987),
Males currently drop out of school at a higher rate than females. However, this appears to be a reversal of the pattern of earlier years. Up through the late 1970's, the dropout rate for females exceeded the rate for males. A rise in the dropout rate for males up through the early 1980's, coupled with a steady decline in the dropout rate for females, has produced lower dropout rates for females. (p. iv)
If it is true that the dropout rate for females is decreasing, one might speculate that a major reason for this shift is a general increase in school-based programs to encourage pregnant females and teen mothers to stay in school. Many schools have made efforts to accommodate teen mothers and mothers-to-be by providing on-campus childcare as well as courses and/or training in parenting skills, child development, sex education, etc.
Although the dropout rate differences between males and females appears to be unclear, some of the reasons given by former students for dropping out do differ significantly. These will be discussed in a later section.
Next Page: Student Behaviors