In focusing on the direct and indirect effects of certain school learning variables on the academic achievement of African American 10th graders, a model was proposed. The model considered those variables associated with student background characteristics (i.e., gender and socioeconomic status); the school (i.e., students' perceptions of the school environment, teachers, and teaching); the family (i.e., parental expectations and involvement); and students (i.e., student educational aspirations and motivation). It was tested using data from African American students who participated in the National Educational Longitudinal Study of 1988. A revised model was developed after only socioeconomic status, prior academic achievement, and students' perceptions of teachers and teaching quality were found to have statistically significant effects on achievement. INTRODUCTION The academic achievement of high school students has long been of concern to parents, academicians, policymakers, and the general public. Many models have been developed to explain variability in student achievement (Donovan, 1984; Keith & Benson, 1992; Keith & Page, 1985; Martinez-Ponz & Zimmerman, 1990; Reynolds, 1989). For minority students, research historically has focused on their background characteristics alone. That is, attempts have been made to predict or explain minority students' academic achievement based on the background characteristics of their families (i.e., socioeconomic status [SES] or home possessions as indicators of affluence); the students themselves (e.g., ethnicity or gender); and/or characteristics ascribed to them (typically ability, self-esteem, locus of control, and motivation) (Burlew, 1979; Curry et al., 1978; Flaugher, 1971; Frye & Coe, 1980; Heussenstamm & Hoepfner, 1971; Johnsen & Medley, 1978; Lloyd, 1967; Suchman, 1968). Other studies have sought to explain varying minority graduation rates by looking at factors that inhibit or enhance their persistence in high school and college (American Council on Education [ACE], 1988, 1989; Astin, 1982; Fleming, 1984; Ogbu, 1990; Spring, 1989). From these and similar studies, the term has been applied to students who were more likely than others to drop out of school. More often than not, at-risk status has been ascribed to students by virtue of their membership in an ethnic minority group (i.e., typically African American or Hispanic American) and/or a social minority group (i.e., teenage females considered at risk of dropping out of school due to pregnancy), or their belonging to impoverished inner-city families (ones & Watson, 1990). Although background and SES undoubtedly factor into at-risk status for poor and minority students, these variables may only partially explain the level of academic achievement attained. Furthermore, they may play merely an indirect role in explaining the at-risk phenomenon's relationship to other causal factors. Thus, models focusing on background characteristics of students alone often cannot explain why students possessing one or more risk characteristics do not drop out of school, nor can they explain why students possessing none of these characteristics drop out. Moreover, reliance on these models seldom leads to improvements in educational service delivery because student backgrounds are difficult, if not impossible, to change. Many researchers have used national data sets to look at achievement patterns and their relationship with variables such as school experience, motivation, self-concept, and aspirations (Rasinski, Ingels, Rock, & Pollack, 1993; Rock, Owings, & Lee, 1994; Rock & Pollack, 1995). For example, Rasinski et al.'s research noted a narrowing of the gap in mathematics achievement between European/Asian American students and African/ Hispanic American students. Tate (1997) later concurred with this finding, maintaining that while the former two groups' mathematics achievement continues to be higher, the latter two groups closed some of the gap by making greater gains. …
Read full abstract