Differential effects of parental involvement on cognitive and behavioral outcomes by socioeconomic status.

Parental involvement is a much investigated area of research. Unfortunately, there are many conflicting findings. In some cases, researchers claim parental involvement positively affects academic achievement, while others claim parental involvement negatively affects academic achievement. Finally, a third grouping of studies find parental involvement has no significant relationship to academic achievement. A parallel body of research investigating the forces affecting dropping out has also been developed. Unfortunately, little research has actually attempted to link the various dimensions of parental involvement to dropping out or other at-risk behavioral outcomes, such as truancy (one exception being Astone and McLanahan 1991). Using the concepts of cultural and social capital, I provide a theoretical framework for why there would exist differential effects of parental involvement across cognitive (e.g., science achievement) and behavioral (e.g., truancy and dropping out) outcomes. I further investigate the variable relationships between involvement and each outcome by socioeconomic status. Findings indicate that parental involvement is generally a salient factor in explaining behavioral but not cognitive outcomes, with greatest support for parent-child discussion and involvement in Parent-Teacher organizations. Findings also indicate that specific dimensions of involvement have greater effects for upper SES students. These findings can be considered preliminary support for the contention of Bourdieu (1977) and Lareau (1989) that the greater levels of cultural capital possessed by the members of the upper class magnifies parental involvement's effect for advantaged students.

1. Introduction

When reviewing parent involvement studies, one is struck by the high degree of inconsistency between the studies and their relative conclusions (e.g., Astone and McLanahan, 1991; Brookover et al., 1979; Epstein, 1992; Ho Sui-chi and Willms, 1996; Keith, 1991; Milne et al., 1986). Many of these inconsistencies can be traced to each study using different theoretical conceptualizations of parent involvement without clearly outlining the concept(s). In terms of inconsistent concepts, the clearest example is comparing Lareau' s and Coleman' s concepts of cultural and social capital and how they relate to parental involvement.

Lareau (1989) states that cultural capital (e.g., networks with other parents, an understanding of the schooling process and teacher jargon, and contact with school personnel) is a significant determining factor in a child's school performance, especially for upper-middle class children. She contends that cultural capital is then translated, or put into effect, via parents' increased levels of parental involvement with the child. Coleman and his colleagues (e.g. Coleman 1987, 1991; Coleman and Hoffer, 1987) conclude that social capital is a crucial determinant of a child's academic success. Coleman contends that parents becoming involved in their child's schooling (e.g. parent involvement with school activities, forming networks with parents of their child's friends, parent-child discussion about school) creates extra sources of social constraint to influence the child's behavior. For example, parents talking to their children and/or becoming involved in the school conveys a message to the child of the importance of schooling and education. In a similar vein, parent involvement in parent-teacher organization (PTO) activities and/or the development of inter-generational closure serves as extra sources of social constraint to stem potential non-normative behavior.

This inconsistency in the linkage between cultural capital, social capital, and parent involvement leads to somewhat contradictory theoretical and empirical findings. While Lareau contends that parent involvement and networking with other parents is the vehicle through which cultural capital is conveyed to one's child, Coleman (1988) contends that social capital in itself is a contributing factor to a child's success. Thus, researchers using Lareau's conceptualization attempt to include cultural capital measures such as parent-school involvement, and specifically parent-teacher contact, while those using Coleman's framework attempt to include social capital elements such as parent-school involvement, parent-child discussion, and inter-generational closure.

Another aspect of much previous research on parent involvement that is particularly troubling is the clear inconsistencies surrounding involvement's effect on student outcomes. Epstein's (1987, 1991) research raises questions about the presumed positive relationship between involvement and achievement, concluding in one study that while gains are higher for some achievement tests, it is not the case for mathematics (Epstein, 1991). Epstein contends that gains in achievement might only occur in subject areas where parents feel less apprehensive about helping (see Griffore and Bubolz, 1986 for a similar argument), but she does not empirically examine this hypothesis. This negative finding between helping with homework and academic achievement is further supported by Horn and West (1992) and a similar relationship is revealed by Milne et al. (1986). Finally, Keith (1991) uses a large national database and also finds inconsistent results for the relationship between parent involvement and student outcomes.

In short, there has been much inconsistency in findings linking parent involvement to academic achievement. I propose that this inconsistency is likely related to one of three possible causes: 1) the use of perception measures by teachers rather than direct reports by students and/or parents; 2) a failure to fully conceptualize parent involvement into its constituent parts, including parent-child, parent-parent, and parent-school components; and 3) not fully assessing the extent to which parent involvement differentially affects academic achievement by social class. These three shortcomings can be alleviated, but a fourth explanation may be applicable to this inconsistency: parent involvement has little direct effect on achievement because it is essentially a cognitive outcome and parent involvement predominantly affects behavioral outcomes.