This article explores the relationship between nations' level of economic development and the influence of adolescents' social backgrounds on their academic achievement. Using data from the Third International Mathematics and Science Study (TIMSS), the authors found that the positive effect of higher parents' education on middleschool students' mathematics test scores is remarkably consistent among the 34 nations examined. In contrast, the relative advantage of living in a traditional family for mathematics achievement varies systematically between nations, being significantly greater in those with stronger economies. Although the influence of socioeconomic status on educational stratification does not appear to change, the deepening academic disadvantage of living with only one natural parent in more developed nations may result from marginalization of families in these societies.
Key Words: academic achievement; cross-national studies; family structure; parents' education; social background.
During the past 25 years, many sociologists and policy makers became concerned with the relatively poor academic performance of American school children, which has been partially attributed to an increasing number of nontraditional famdies (Astone & McLanahan, 1991; McLanahan & Sandefur, 1994). The assumption is that children living with only one parent, or a stepparent, lack access to social and economic resources vital for academic achievement and success in school. The rising numbers of children living with single parents in other Western nations has increased international interest in the relationship between family structure and children's academic success (Cochran, Larner, Riley, Gunnarsson, & Henderson, 1993; Pong, 1996). However, most studies of family structure are limited to only a few nations and do not explore the impact of economic development, culture, or public policy on the relative disadvantage of living in a nontraditional family.
In contrast, comparative education research has a long tradition of exploring cross-national differences in the effect of children's socioeconomic background on their academic achievement and educational attainment. One continuing debate concerns the impact of economic development on the association between parents' socioeconomic status and their children's academic success (Baker, Goesling, & LeTendre, in press; Blossfeld & Shavit, 1993; Heyneman, 1980; Simmons & Alexander, 1978). These arguments revolve around whether increasing economic and human resources in a nation alters the role of families, particularly the importance of social and motivational support for children, in educational stratification. If economic development influences the process of intergenerational transfer of socioeconomic status, then it may also affect the relative disadvantage of living in a nontraditional family.
The analyses in this article explore variation in the relationships between two aspects of adolescents' social backgrounds: (a) their parents' education, and (b) family structure, as well as mathematics achievement across 34 nations participating in the Third International Mathematics and Science Study (TIMSS). Using information from over 200,000 middle-school students (modal age is 13 years old), we examine the extent to which performance on the TIMSS mathematics test is related to these two indicators of adolescents' social backgrounds after taking into account their academic ability and orientation toward schooling. Hierarchical Linear Modeling (HLM) is used to determine the extent to which variations in the effects of parents' education and family structure on mathematics achievement are related to national levels of economic development.
SOCIAL BACKGROUND AND MATHEMATICS ACHIEVEMENT
In the quarter of a century between 1972 and 1997, the percentage of 6- to 12-year-olds living with only their mothers doubled from 12% to 24% in the United States (National Center for Education Statistics [NCES], 1999). …