【正文】 o do not hold any assets. This may positively affect parenting behaviors and investment and thus affect children’s educational attainment.. Assets and children’s educationIn the last decade, as more attention has been given to wealth as an aspect of household economic status, some empirical studies have examined the impact of household wealth on children’s education. These studies have focused on investigating direct effects of parental assets on children’s wellbeing. For example, Mayer (1997) found that investment ine and inherited ine explained more variance in children’s educational test scores and achievement than did total family ine. Studies also reported that children were more likely to graduate from high school if they lived in households where parents were homeowners (Aaronson, 2000。Green amp。 White, 1997。 Kane, 1994).Several recent studies have examined possible independent and distinct effects of assets from ine on children’s education. For example, Conley (2001) indicated that parental net worth had a strong effect on the postsecondary schooling of children–net of ine and other measures of socioeconomic background. Williams (2004) found that, controlling for many other factors, parental wealth (net worth, account ownership and stock/IRA ownership) was positively associated with educational achievement of children. Williams also found that the effects occurred even among very inepoor families. Similarly, Zhan and Sherraden (2003) found that lowine single mothers’ assets (home ownership and savings) were positively associated with children’s educational attainment. Furthermore, this study found that ine was associated with educational achievement when assets were not in the model。 however, the relationship between ine and children’s education disappeared when assets were included.. Parental expectations and children’s educationParents with higher expectations for their children are more likely to set higher standards for heir children’s schooling and social functioning than parents with lower expectations. They are also more likely to transmit the values of doing well in school and of getting along well with eachers and peers. Consequently, as conceptualized in some literature (Reynolds amp。 Lee, 1991。Reynolds amp。 Walberg, 1992), parent expectations may be viewed as a dimension of the home environment that directly as well as indirectly influences children’s behaviors and achievement.Empirical literature provides consistent support for the positive relationship between parent expectations or aspirations toward their children and their children’s educational achievement ( Axinn et al., 1997。 Furstenberg amp。 Hughes, 1995。 Hanson, McLanahan, amp。 Thomson, 1997。Reynolds amp。 Gill, 1994。 Smith, Beaulieu, amp。 Seraphine, 1995). For example, the study by Smith et al. (1995) found that parental expectation of their children’s college attendance was a strong and positive predictor of actual subsequent college attendance of their children across urban, suburban, towns, and rural areas. Furstenberg and Hughes (1995) found similar results among AfricanAmerican teenage mothers and their children.In most of these studies, parent expectations and aspirations are viewed as independent traits. The influence of parental assets on their expectations and the mediating effects of expectations on the link between parent resources (especially parental assets) and children’s wellbeing have rarely been examined.. Parental involvement and children’s educationStudies have suggested that parental involvement of children’s activities may signal the route through which a parent’s skills and motivations are transferred to children and should be positively associated with children’s cognitive and other development (Becker, 1993。 Hill amp。 O’Neill, 1994).Findings from empirical studies, however, are not consistent. Recent studies indicate that while overall it appears that parent involvement in school was associated with academic outes, parent involvement in the home was not (Barnard, 2004。 also see a recent metaanalysis by Fan amp。Chen, 2001). The different operational definitions of parental involvement in these studies may partially contribute to the inconsistent findings (Fan amp。 Chen, 2001).Several studies have also noted the differential effects of parental expectations and parenting behaviors on children’s education. They indicated that parental expectations and attitudes, rather than specific behaviors such as involvement in school activities, better explained children’s academic outes (Fan amp。 Chen, 2001。 Reynolds amp。 Gill, 1994).. This studyUsing a nationally representative sample, this study examines how parental assets of a group of children affect parent expectations and involvement at school and at home, and children’s academic achievement measured 2 years later. Through longitudinal data analyses, this study will contribute to the literature by specifying the relationship between parental assets and children’s academic outes, and by investigating possible mediating effects of parental expectations and involvement on the relationships between parental assets and other parental characteristics with children’s educational performance.3. Data and methods. SampleData were extracted from the mother–child data set of the National Longitudinal Survey of Youth (NLSY79). In 1979, 12,686 individuals between 14 and 22 years of age, including an oversample of minority and economically disadvantaged white youth, prised of the original NLSY. From 1979 through 1994, respondents were interviewed annually and interviewed biennially thereafter. In 1986 and every 2 years afterwards, a number of assessments were administered to the original female participants and to their biological children. By 2000, 11,205 children were born to the 6283 NLSY female respondents (Center for Human Resources Research, 2001).The study sample included children who were between the age