A number of studies have found that parents who are involved in their adolescents' lives are able to influence their academic success, although certain forms of involvement appear to matter more than others. McNeal (1999) examined the relationship between parent involvement and science achievement in a national sample of students. The measures of parental involvement included parents' PTO involvement, the frequency of parent-child discussions, parental monitoring, and parents' use of educational support strategies. Greater frequency of parent-child discussions were related to higher levels of science achievement. However, he found that adolescents whose parents were more involved in PTO activities actually had lower levels of science achievement than adolescents who were less involved in these activities. Further, McNeal (1999) found that the relationship between parental PTO involvement and parental monitoring with adolescents' achievement was stronger for families with higher levels of SES than for those with lower levels of SES. Higher levels of parental monitoring were moderately related to lower levels of science achievement. The results were found after controlling for race, gender, SES, prior achievement, GPA, hours worked, hours of homework, having repeated a grade in school, and family structure. Gutman and Eccles (1999), in a path analysis of data on a sample of 617 Black and White adolescents from the Maryland Adolescent Development in Context longitudinal study, found that adolescents from families with higher levels of family income were less likely to have parents who were financially stressed. Adolescents with parents' who reported feeling financially strained, as indicated by their reports of worrying about money and of having difficulty making ends meet, were less likely to be involved in their adolescents' school and were more likely to have negative relationships with their adolescents. Parental involvement was measured by parent and adolescent reports of parental attendance at open house events, parent teacher association meetings, and volunteering in classroom. Parent-adolescent relationships were measured by levels of harsh discipline and conflict between parents and adolescents reported by both parents and adolescents. Adolescents' achievement, as indicated by school records of their GPA, was positively influenced by parental involvement and was negatively influenced by negative parent-adolescent relationships (Gutman and Eccles, 1999). Catsambis (1998), in an analysis of NELS data, found that adolescents whose parents had high educational aspirations for them in the eighth grade had higher levels math, reading, and science test scores, after controlling for parental work status, family structure, prior achievement in the relevant subject, race, SES, and family size.