Dr. Chandra Muller is assistant professor of sociology at The University of Texas at Austin.

Parents and Schools Parental involvement captures the attention of a nation that makes education a top priority. Parents recognize the benefits and value of active participation in their children's education. School administrators realize that a parental involvement policy is an appealing tool for dealing with student problems and declining budgets, and most school districts have instigated such policies. State and national policy makers discuss parental involvement policies at conferences and in the media. Sociologist James Coleman, in his influential 1966 study, showed that attributes of parents made more of a difference in how students performed at school than any measurable qualities of the school. Not only did family background make a difference in achievement, he pointed out, but it affected how well students used the school's resources. Perhaps most controversial, the combined family background of students in a school affected the achievement of each student, independent of the student's own parents' attributes. If a student's classmates' parents had high levels of education, that student would be better off regardless of the level of education of the student's own parents. No one knew why family background mattered so much. Our research at The University of Texas at Austin explores how parental involvement encourages academic achievement among most students. It examines the most effective ways for parents to be involved in their child's education and probes into how parental involvement is affected by maternal employment and family structure. Why does family background make such a difference to children? Parents with more money can smooth problems. Beyond money, parents with higher levels of education often have better knowledge about how to make a system like a school work to their child's advantage. Even before a student enters school parents may provide a rich home environment to prepare their young child for learning and success. Sociologists and developmental social psychologists now realize that what parents do with and for their child makes a difference in how a child succeeds in school. Recent studies have focused on the importance of the social relationships in educating a child. Parents with money help their child succeed in school only if they maintain a relationship with their child that facilitates their productive use of the money. Likewise, parents with knowledge about how the system works are effective only if they offer guidance appropriate to the child's needs. Parental involvement is complex to evaluate because there are many ways for parents to be involved. The form involvement takes depends on the resources of the parents (like money and education), the interests and priorities of parents, and the needs of the child. It is through the relationships parents have with their child that their resources are conveyed. It is important to distinguish between the resources of parents and the relationships parents have with their child. Otherwise it would be impossible to determine whether an observed statistical relationship is a function of the parents' resource or the parent's relationship with the child. Parental action is likely to have a differential impact depending on the circumstances and needs of the child. Yet similarities exist among the consequences of involvement for students despite the complexity. Study of Parental Involvement. The research described in this article is based on analyses of the National Longitudinal Study of 1988 (NELS), a nationally representative data set of 24,599 eighth graders and their parents, teachers, and principals. Data were collected again two years later when most students were tenth graders. Both students and their parents reported on qualities of the parent-child relationship. After extensive analyses, I selected twelve measures of parental involvement from about 160 in the eighth grade data set. These measures were selected because they vary along several dimensions and because the respondents appear to have interpreted the questions in a way that allows for a straightforward analysis. In addition to extensive surveys, students were given achievement tests designed by the Educational Testing Service. I chose to measure achievement with eighth grade mathematics test scores and the gains on those tests between eighth and tenth grades. Mathematics is most easily assessed on standardized tests, and it is likely to indicate the impact of learning at school since mathematics test scores are most responsive to classroom exposure. Furthermore, mathematics ability is important for students' later post-secondary educational and occupational opportunity. The transition between eighth and tenth grades is important. For some students it represents a time when mathematics becomes an elective. In addition, high school math classes may be more competitive and focused on individual instruction. The forms of involvement most strongly associated with eighth grade test scores are different from the involvement that predicts test score gains. The initial test score is a good measure of the student's ability to take achievement tests, which in large part measures the student's knowledge base and academic ability. In contrast, test score gain is a measure of growth; given a certain test score, gain measures the increase (or occasionally decrease) made by the student during the two-year period. Thus the initial test score measures more about how the student is prepared for test taking and learning. Test score gains, in which the student's initial score is held constant, only measure preparation if it contributes to the additional learning that takes place between the eight and tenth grades. Involvement and Academic Performance. Parental involvement varies along three important dimensions having to do with the circumstances and consequences of involvement. The first two dimensions are most useful for understanding the way involvement is related to eighth grade test scores. The last dimension is more important for understanding the association with gains in test scores. The first dimension distinguishes involvement that takes place in the home, the community, or the school. Home-based involvement, such as frequency a parent talks with a child about school experiences or restricts television watching, may be influenced by the interests and values of the parent. Those activities are much less dependent on the parents' work schedules and school cooperation than are school-based activities like attendance at school events, parent-teacher organization (PTO) meetings, or a conference to discuss the student's academic progress. A second important dimension of involvement is related to the purpose of involvement. Parents may become involved in response to a particular stimulus. Meeting with a teacher, checking the child's homework, or having a serious talk about the future are examples of ways a parent may attempt to intervene in an academic problem at school. In contrast, involvement may be motivated more by a general interest or priority in education and less directed at a particular consequence. Involvement that is not intervention may take place in the home, for example, by talking about school and limiting television, or it may make use of community or school resources such as arranging after-school supervision or enrolling in extra music class. These activities indicate a combination of interests of parents that complement academics and a structuring of the out-of-school activity of adolescents. Eighth grade test scores are most responsive to involvement based in the home or community that is not directed at a specific goal such as an intervention. The amount parents talk with their child about school experiences is the best predictor of the student's eighth grade test score. When students talk frequently with parents about school experiences (at least three times a semester) the student is likely to score about 1.3 points higher on a mathematics achievement test, raising the score from the average to the top third of all eighth graders. A student who never talks with parents about school is likely to score 2.7 points below the average, or around the bottom quartile. Other forms of involvement that are positively associated with achievement include whether the student is enrolled in outside music classes (which may improve a student's test score by 1.2 points, on average) and whether the student receives adult supervision after school. My analyses indicate a significant drop in achievement only if a student is unsupervised for more than two hours per day. These forms of involvement, especially talking about school experiences, most likely reflect the involvement of parents throughout the child's life. Bettye Caldwell and others who study younger children have documented the advantages of verbal interaction between parent and child and other activities like reading to a child. It is likely that most parents who talk with their eighth grader also had higher levels of verbal interaction when the child was younger. Likewise, an eighth grader who is supervised probably had earlier supervision. These are interactions in which the parent and child have an established relationship, and, while the content of the interactions complements what happens in school, the purpose is not directed at a particular goal. In contrast, little evidence can be found of a positive consequence when a parent intervenes in a student's problem. For instance, a student who is average in every way but whose parent checks his or her homework often is likely to score among the lowest 30 percent of test takers (instead of among the middle 50 percent). Similar patterns are found when parents frequently contact the school about a child's academic behavior or when fathers talk with their child about high school programs. For example, parents who frequently contact the school are likely to have a child perform in the bottom third of the math achievement test distribution instead of the middle. These negative associations illustrate why it is impossible to establish a causal relationship between involvement and test performance. Almost certainly the cause of lower test scores among students whose parents check homework, contact the school, or discuss high school program planning is not the action of the parent. Rather, it may be attributed to an unmeasured factor like the student having a problem at school which causes both the parents' involvement and the lower achievement. Strong negative relationships persist when test score gains are examined, suggesting that they probably do not contribute to student improvement. An exception is when parents intervene by checking homework or contacting the school when the child has very low grades. If a child with D's and below reports that his or her parents is intervening at a high level then the student is likely to experience above average test score gains in high school. Involvement and Learning Gains.The third dimension of involvement has to do with the relationship of the parent to the school and the educational process. Students whose parents are active in the PTO gain slightly more in test scores. PTO may be a way for parents to engage in authoritative action, possibly by influencing school policy or by obtaining information such as course offerings. In this case parents are authoritative about the particular needs of their child, but they may not engage in actual delivery of services such as teaching. In contrast, parents who volunteer at the child's school, becoming an aide to help supplement the school's needs, have children who gain slightly less on test scores. In this case the school is the authority which parents support and contribute. Managing activities are associated with larger gains while support activities are accompanied by less growth. It is impossible to determine how much the support activities are an attempted intervention. Evidence indicates that when the parent takes a subordinate role relative to the school it does not lead to greater learning gains for the student. The managing role implies parental authority about the needs of the child. Parents may engage in these activities at home and in the community, as well. A managing parent may structure the out-of-school activities in a way that is conducive to the child's learning and engagement in school-related matters. Test score gains are largest among students whose parents limit going out with friends and provide after-school supervision. Each action contributes to an additional gain of about one question on the tenth grade achievement test, or a gain of about one-quarter of a grade level. An important component of the activities of the managing parent may also be to talk with the child about what happens in school, thereby maintaining current knowledge of the child's experiences (and needs) and showing support for the school material in a general way. The amount parents talk with their child about school is a predictor of test score gains, but it is not as important for predicting gains as it is for measuring initial test scores. Once the contribution to the eighth grade test score is taken into account, talking about school experiences is associated with an additional gain of about two-thirds of a question. Involvement and the Changing Family. Many positive forms of involvement are associated with the resources and interests of parents. Two sociological changes in the family-increased numbers of mothers with young children who are employed outside the home and climbing divorce rates-are a source of concern about the ability of parents to have the resources for adequate involvement with their children. Maternal employment and divorce (which results in single parent and stepfamilies) affect the time, money, and attention parents devote to their child. The NELS data allow us to examine how parental involvement is related to whether the mother is employed outside the home and to variation in family structure. Overwhelmingly, parents are most involved when the mother is employed outside the home part-time. These parents have the highest levels of almost every form of positive involvement. The pattern exists in intact families (those where parents are living together and are not divorced), in single parent families, and in stepfamilies. Moreover, students whose mothers are employed part-time have higher levels of achievement regardless of the parents' involvement level. Probably these parents are involved at higher levels in many unmeasured ways with their children. Differences are evident in the involvement of parents and the achievement of students if the mother is employed full-time or not employed. The largest and most important difference is the amount of after-school supervision available to students. Students whose mothers are employed full-time are unsupervised for longer periods after school, and they attain lower test scores. If students whose mothers are em-ployed full-time were supervised at levels comparable to those whose mothers are not employed, the former would attain comparable test scores. Students of single mothers are also more likely to be left unsupervised after school, and the lack of supervision has a negative impact on their test performance as well. Researchers at the University of Chicago who have studied involvement and family structure using NELS find that parents in intact families in general have much higher levels of involvement. Students in intact families have higher levels of achievement, which can be partially explained by parental involvement and income. The lower levels of achievement of students in single mother families is due entirely to income and involvement differences. It is difficult to generalize about students in stepfamilies. Those whose mothers are employed part-time are similar to students in intact families. Others have characteristics more like students in single parent families. It appears that stepfamilies may vary more than others in the extent to which they allocate family resources, including money and parental attention and interest, to the child. Policy Implications. Our findings about after-school supervision can be translated into an important policy consideration. Institutions vary in how they accommodate the needs of employed parents. Schools schedule activities to facilitate involvement differently, and employers vary to the extent they allow flexibility in work schedules. The versatility of each institution may influence the involvement of parents, especially outside the home. Given the dramatic increase in labor force participation of mothers, many institutions have not kept pace with the changing needs of parents, thereby making involvement outside the home more difficult. The most striking example is the apparent lack of available opportunities for supervised after-school activities for children of employed parents. Little evidence supports the idea that parents have less interest in a relationship with their child when the mother is employed full-time or when they are not in an intact family. The form of involvement that best reflects the interest of the parent in the child is the amount parents talk with their child about everyday school activities. Parents tend to talk more with their child when the mother is employed part-time, and quite possibly as a group these parents have more interest in being positively involved in their child's education. There is no difference, however, between this kind of relationship when children of mothers employed full-time and those not in the labor force are compared. Time spent by mothers in the labor force and variation in family structure make the most difference to parents and their child in relations that involve others, particularly when the timing of the activity is inflexible. Conclusion. Students appear to benefit from relationships with parents that are positive, ongoing, verbal, and supportive of education in a general way. In addition, students gain from a parental style of guidance and management of activity both in and out of school. Our research suggests that relationships between parents and child that are important for adolescent development might benefit if the institutions in our society were structured in a way that did not force parents to choose between involvement in important social relationships with their child and working outside the home. It appears at this time that opportunities to balance activities may be more available to families with more resources-income, education, and two natural parents. Improved availability of supervised after-school activities, flexible work time, and school policy that adapts to the needs of parents' work schedules are examples of potentially beneficial policy. Acknowledgments: This research was supported by a Spencer Foundation Post-doctoral Fellowship from the National Academy of Education and a grant from the American Educational Research Association Research Grants Program.