Fathers' Involvement in Their Children's Schools

FINDINGS

Involvement of Resident Parents

Types of Activities in Which Fathers and Mothers Participate

Figure 1 shows the percentage of children in kindergarten through 12th grade whose fathers and mothers participated in each of the four activities inquired about in the NHES:96. Several observations can be made from this figure. First, in two-parent families, the most common activity in which parents participate is attending a general school meeting, such as a back-to-school night. Although single fathers also appear more likely to attend a general school meeting than they are to participate in the other activities, the proportion who do so is not significantly /13 different from the proportion who attend conferences or school or class events. The least common activity that parents participate in, regardless of family type, is volunteering at their children's schools, the most time-intensive of the four activities.

Second, fathers in two-parent families are substantially less likely than mothers in either type of family or fathers in single-parent families to participate in each of the activities. For example, 55 percent of fathers in two-parent families attended at least one general school meeting compared to 77 percent of mothers in two-parent families, 69 percent of mothers in single-parent families, and 68 percent of fathers in single-parent families. This pattern fits the notion that parents in two-parent families divide their labor to more efficiently allocate their resources, in this case, their time. According to economic theorists, efficiency in a family is increased by specialization in both the allocation of time and human capital (Becker, 1981). The finding that single fathers are more involved than fathers in two-parent families is consistent with existing research. A study based on the National Survey of Families and Households found that with the exception of the time spent sharing meals, single fathers spend more time with their children than biological fathers in two-parent families (Cooksey and Fondell, 1996).

Third, fathers who head single-parent families have school involvement patterns that are very similar to that of mothers who head single-parent families. The pattern of participation for both fathers and mothers in single-parent families is more similar to the participation of mothers in two-parent families than it is to fathers in two-parent families. The one activity that is substantially lower in single-parent families than it is for mothers in two-parent families is that of volunteering at the school. This result is consistent again with the roles that parents fill in two-parent and in single-parent families. In two parent families, mothers generally assume primary responsibility for the children. In single-parent families, the single parent must fill that role regardless of whether that parent is the mother or the father. However, because involvement requires a certain amount of time, single parents generally experience greater constraints on their ability to participate than do mothers in two-parent families, who can share with their husbands some of the other demands on their time. It is precisely the activity that requires the most time, volunteering, in which the greatest difference between single parents and mothers in two-parent families is seen.

Fourth, in two-parent families, there are two activities for which fathers' involvement approaches that of mothers: attendance at general school meetings and attendance at school or class events, such as sports events. For these activities, the ratios of the proportion of fathers who have participated to the proportion of mothers who have participated are 71 percent (55/77) and 79 percent (53/67), respectively, whereas the ratios of the proportion of fathers to mothers who have attended a parent-teacher conference or volunteered at their children's schools are 57 percent (39/68) and 37 percent (15/41), respectively. Fathers may find it easier to attend general meetings and school events, such as sports events, because such activities are more likely than the others to occur during nonschool and nonwork hours, thus increasing the ability of fathers to attend without missing work.

Levels of Involvement in School Activities by Fathers and Mothers

Thus far, participation in specific activities has been examined. It is also of interest to know the extent to which parents participate in multiple activities. Figure 2 tells a similar story to figure 1: Mothers in two-parent families are the most likely to show high levels of involvement in their children's schools (56 percent), while fathers in two-parent families are the least likely to show high levels of involvement (27 percent). Fathers who head single-parent families show similar levels of involvement to mothers who head single-parent families (46 percent versus 49 percent). This pattern adds further support to the notion that there is a division of labor in two-parent families, with mothers taking more responsibility for participating in school activities, whereas in single-parent families the lone parent assumes that responsibility. It also supports research that finds single fathers and mothers are more similar in their parenting behavior than are mothers and fathers in two-parent families.

Another observation that can be made from figure 2 is that most parents participate in at least some of the activities in their children's schools. Although in two-parent families nearly half the children have fathers who participate in none or only one of the four activities, 79 percent of the children have mothers who participate in two or more activities in their schools. In single-parent families, 72 percent of children living with their fathers and 74 percent living with their mothers have a parent who participated in two or more activities in their schools.

Changes in Involvement with Grade Level

Research on parental involvement in schools has shown that involvement in schools decreases as children grow older (Zill and Nord, 1994). As noted earlier, part of the decrease is due to schools offering parents fewer opportunities for involvement as children grow older./14 As can be seen in figure 3, mothers and fathers in both two-parent and single-parent families tend to decrease their involvement as children move from elementary to middle to high school. Fathers in single-parent families behave very much like mothers in single-parent families with respect to involvement in school activities. At each grade level, their involvement patterns remain at similar levels. In two-parent families, however, mothers' involvement starts out higher but decreases more than fathers' involvement so that by the high school years, though mothers still show higher levels of involvement, the differences are not as large. This pattern can also be seen by comparing the ratio of the proportion of fathers in two-parent families with high levels of involvement to the proportion of mothers in such families with high levels of involvement at each school level. In elementary school, this ratio is 43 percent (30/69). It increases to 49 percent (25/51) by middle school and to 59 (23/39) percent by high school.

The convergence of mothers' and fathers' involvement in two-parent families occurs because mothers show a steeper decline than fathers in two of the activities: attendance at a school or class event and volunteering. Figure 4 shows how mothers' and fathers' participation in these two activities change with grade level./15 Mothers' participation in school or class events declines steadily as children move from elementary to middle to high school. Fathers' participation, on the other hand, remains steady and even increases somewhat between grade school and middle school. The pattern is somewhat different with respect to volunteering. The proportion of mothers in two-parent families who volunteer in their children's school declines sharply between grade school and middle school and then levels off. Among fathers in two-parent families, the proportion who volunteer also declines between elementary and middle school, but then increases again between middle and high school. Approximately 12 percent of fathers in two-parent families with children in grades 6 through 8 volunteered in their children's schools. This figure increased to 17 percent of fathers with children in grades 9 through 12.

Familial Resources and Parental Involvement in Schools

Parental involvement in children's education involves an investment of time and other resources. In order to be able to invest these resources, it is necessary to have them available. Much existing research on parental involvement in schools has shown the association between involvement in schools and social and demographic characteristics of the family such as the number of parents present, the education level of the parents, and maternal employment. As noted previously, these factors are markers for different types of capital available to the family: financial (or physical), human, and social.

Financial and human capital. As others have found, parental involvement in school is associated with the amount of financial capital that families have available to them. In general, families with more financial resources show greater levels of involvement. For example, the proportion of fathers and mothers in two-parent families who are highly involved in their children's schools increases as household income rises (figure 5). The same is true of single fathers and single mothers, though it appears that single mothers earning the highest incomes may be somewhat less involved. This pattern could be due to demands placed on the mothers' time by their work, which interferes with their ability to be involved in their children's schools. However, the difference between the top two income categories for single mothers is not significant, so not too much emphasis should be placed on the apparent decrease in high involvement.

Parental involvement in schools is higher for children in families living above the poverty threshold and not receiving federal assistance/16 compared to those that experience economic difficulties (figure 6). This is true in both two-parent and single-mother families, though the differences are larger in two-parent families. This pattern of results is probably due, in part, to the fact that there is a wider disparity in the household incomes of families experiencing and not experiencing economic difficulties in two-parent families than in single-mother families (Baugher and Lamison-White, 1996). To the extent that income and income-related factors are linked to involvement in school, the greater disparity in two-parent families could account for the more marked difference in high involvement by these economic indicators. Single fathers are also more likely to be highly involved if they do not receive federal assistance, but there is no significant difference in the proportion who are highly involved by whether their household incomes are above or below the poverty threshold.

Home ownership is also associated with high parental involvement in two-parent families and in mother-only families, but not in single-father families. Home ownership is not only a measure of economic well-being, but it also is an indicator of stability in family life. Families who own their own homes tend to move less often than families who rent (Hansen, 1995). This stability, in turn, allows for the establishment of more ties to individuals and institutions in the community, reflecting a greater level of social capital (Coleman, 1988).

Parental involvement in schools increases with the amount of human capital in the home, as measured by parents' education (figure 7). This is true of parents in both single-parent families and in two-parent families. Income and education, however, are highly correlated. People with greater levels of education have higher incomes than those with less. To determine the unique influence of education and income on parental involvement, they both need to be included along with other potentially confounding factors in multivariate models.

It is also interesting to note that figures 5 through 7 show that mothers and fathers in single-parent families are more similar to each other in their levels of high involvement than mothers and fathers in two-parent families. Thus, even controlling for the financial circumstances of the families and the education levels of the parents, mothers and fathers in single-parent families are more similar in their tendency to be highly involved in their children's schools than are mothers and fathers in two-parent families.

Social capital. Numerous studies have shown that parental involvement in schools promotes school success (Henderson and Berla, 1994; Henderson, 1987). It seems likely that it is not attendance at school activities, per se, that leads directly to improved school outcomes, but rather that such attendance is a marker for other important factors that contribute to children's school success (Zill and Nord, 1994). For example, parents who are involved may be more familiar with the school and with their children's teachers. This familiarity may lead to better parent-teacher relations and more personal attention for their children. It may also enable the parents to intervene earlier if problems in their children's behavior or academic work should arise. Attendance at school functions also shows children that their parents believe school is important. However, it is also likely that parents who are highly involved at school also hold certain beliefs and attitudes and exhibit behaviors at home that foster the academic success of their children. Thus, at least part of the positive benefit of involvement in the schools may be due to the types of parents that become involved, including not only their demographic characteristics such as income and education, but also their own behaviors outside the school building and their attitudes and expectations regarding education.

The NHES:96 asked a series of questions about the types of activities that families did with their children in kindergarten through grade 5 and those in grades 6 through 12 in the past week and in the past month. In addition, parents of children in grades 6 through 12 were asked about their educational expectations for their children, whether they discussed future school courses with their children, and about their own involvement in community activities and attendance at religious services. Although the NHES did not ask which parent in the household did each of the activities with the child, these items can still be used to characterize the households of children whose fathers and mothers exhibit high or low levels of involvement in their schools.

These data reveal that parents who are highly involved in their children's schools are generally more likely to be involved at home, as well. Single fathers, especially those with children in elementary school, are an exception, and they will be discussed separately. Children in kindergarten through grade 5 who live in two-parent families in which their fathers or mothers are highly involved in their schools are more likely to participate in educational activities with their parents (e.g., were told a story by their parents or in the past week or visited a museum or historical site with their parents in the past month) than are children whose parents have low levels of involvement in their schools (figure 8a)./17 The same patterns are observed for children living in mother-only families (figure 8b). Single mothers who are active at school tend to be active at home, as well.

It is not only elementary school children who appear to have a richer home life if their parents are highly involved in their schools. Children in grades 6 through 12 whose fathers or mothers have high levels of involvement in their schools are also more likely than children whose parents have low levels of involvement to have played a game or sport or to have worked on a project with their parents in the past week (figure 9a and figure 9b). They are also more likely than children whose parents have low levels of involvement in their schools to have discussed how to manage their time with their parents in the past week or to have talked about future courses in the past month. Parents who are highly involved in their children's schools are also more likely than those who are less involved to believe that their children will attend school after high school and that the youth will graduate from a 4-year college. These statements are true of children living in two-parent families and those living in mother-only families.

The story is somewhat different for single fathers. Single fathers who are highly involved in their children's schools are more likely than those with low levels of involvement to participate in some activities with their elementary school children, such as having visited a museum or historical site and having attended a sports event in the past month. However, for many of the other activities there is no significant difference between single fathers with high and with low levels of involvement in their children's schools (figure 8b). What is striking about the single fathers with children in grades 1 through 5 is how highly involved at home the fathers are who have low levels of involvement in their children's schools. The proportion of elementary school children living in single-father families who have been told a story or worked on a project in the past week with their fathers is virtually the same for those whose fathers have high levels and those whose fathers have low levels of involvement in their schools.

Among single fathers of children in grades 6 through 12, the contrasts between those with high and those with low involvement in their children's schools are greater, more like those seen in two-parent and in mother-only families. Children in grades 6 through 12 living in father-only families are more likely to have played a game or sport with their fathers in the past week and to have talked with their father about future high school courses in the past month if the fathers have high levels as opposed to low levels of involvement in their children's schools. Similarly, fathers with high levels of involvement are more likely than those with low levels of involvement to expect that their children will attend school after high school and that they will graduate from a 4-year college.

These figures yield another interesting observation upon close inspection: With a few exceptions, children whose parents are highly involved in their schools are almost equally likely to have shared in any given activity with their families outside of school, regardless of whether it is their fathers or mothers who are highly involved or whether they live in two-parent or in single-parent families. For example, 74 percent of children in kindergarten through 5th grade who live in two-parent families were told a story in the past week if their fathers were highly involved in their schools, compared to 72 percent if their mothers were highly involved and to 71 percent of children living in father-only families whose fathers were highly involved and 72 percent of children in mother-only families whose mothers were highly involved (figures 8a and 8b). When differences occur, they tend to be differences between two-parent and single-parent families rather than between mothers and fathers. For example, 61 percent of children in grades 1 through 5 who live in two-parent families and whose fathers were highly involved in their schools went to the library with a parent in the past month, as did 59 percent of children in such families whose mothers were highly involved (figure 8a). In contrast, 47 percent of children in father-only families and 52 percent of children in mother-only families whose parents were highly involved had visited the library with their parents in the past month. These patterns suggest that regardless of family type, families who are involved in their children's schools tend to share other activities with their children as well. However, the constraints of being the only parent in the household may limit the ability of single parents to be as involved as they might wish.

There is another form of social capital that highly involved mothers and fathers offer their children: greater connections to the larger community. For children living in two-parent families or in mother-only families, their fathers or mothers are more likely to belong to an organization such as a community group, church or synagogue, union, or professional organization and to participate in an ongoing service activity if the parents have high levels rather than low levels of involvement in their schools (figure 10). For children living in father-only families, their fathers are significantly more likely to participate in an ongoing service activity, but are not significantly more likely to belong to an organization, if their fathers have high levels of involvement in their schools as opposed to low levels of involvement. Similarly, for children living in two-parent families or in mother-only families, their parents are more likely to attend religious services on a weekly basis if the mothers or fathers have high levels of involvement in their schools. The differences are not significant for children living in father-only families, though children whose fathers never attend religious services are significantly less likely to have fathers who are involved in their schools than are children whose fathers attend religious services at least occasionally.

School Resources and Parental Involvement

As noted previously, existing research has found that school factors exert a strong influence on parental involvement in their children's education (Eccles and Harold, 1996; Epstein and Dauber, 1991). Important school factors include whether the school is a public or private school, the size of the school,the school environment or climate, school policies and practices, and teacher attitudes and practices (Epstein and Dauber, 1991; Coleman and Hoffer, 1987). Although not all of these factors can be examined using NHES:96 data, several can. As can be seen in figure 11, fathers and mothers in both two-parent and in single-parent families are more likely to be highly involved if their children attend private as opposed to public schools. There are no significant differences in parental involvement between attending a public school that is assigned and one that is chosen. Private schools often make parental involvement a requirement, and thus, part of the higher involvement may be due to school policies. However, Coleman and others have argued that private schools, particularly private, religiously affiliated schools, have greater amounts of social capital due to the greater sense of community present in these schools (Bryk, Lee, and Holland, 1993; Coleman and Hoffer, 1987). Parents may be more willing to become involved when they know and are friendly with other parents in the school.

One reason for expecting that school size may be an important influence on parental involvement is that it may be easier to establish ties with administrators, teachers, and other parents when schools are not very large. If that is the case, parents may feel more comfortable and more welcomed by smaller schools. Figure 11 shows that high involvement in schools does tend to decrease as school size increases. However, the decreases are not uniform and are often not significant. High involvement by fathers and mothers in two-parent families is greater in smaller (less than 300 or 300-599 students) as opposed to larger schools (600-999 or 1,000 or more students) (figure 11). In mother-only families, mothers are significantly less likely to be involved if the school is very large (1,000 or more students) as opposed to smaller (less than 1,000 students). There are no significant differences in single mothers being highly involved in schools by the size of the school in schools smaller than 1,000 students. None of the differences in father-only families by school size are significant. It should be noted, however, that broad categories of size were used to classify schools. There may be threshold sizes, which differ for mothers and fathers and by grade of the children, after which high involvement decreases.

The notion that social capital within schools encourages parental involvement gains support by examining the association between fathers' and mothers' involvement in schools and several measures of the school environment. The NHES:96 asked parents of children enrolled in grades 1 through 12 about how strongly they agreed with the following statements:

Child's teachers maintain good discipline in the classroom.
In child's school, most students and teachers respect each other.
The principal and assistant principal maintain good discipline at child's school.
Child's school welcomes my family's involvement with the school.
Child's school makes it easy to be involved there.

Respondents are more likely to strongly agree with these statements if the parents have high involvement in their children's schools than if they have low involvement (figure 12). For example, among two-parent families in which fathers have high involvement, nearly half strongly agree that the children's teachers maintain good discipline in the classroom compared to about a third of respondents in families in which fathers have low involvement. It is not possible to establish the causal linkage in the outcomes from a cross-sectional survey such as the NHES. It may be that greater school efforts to promote involvement lead to greater parent involvement. Conversely, it may be that highly involved parents have the skills to establish cooperative relationships with their children's schools, and thus view the schools more positively and see them as more welcoming.

Influences on Parent Involvement in School

From the foregoing discussion, it is apparent that high levels of parent involvement in schools are correlated with other variables such as education, economic status, family environment, and school characteristics. Many of these same factors are also correlated. For example, parents with more education are more likely to have higher incomes. To obtain a sense of the relative importance of these different factors in contributing to high parental involvement, it is necessary to examine them together in a single model. Because it is of interest to examine the association between social capital in the home and high parental involvement in schools, separate models were estimated for children in grades 1 through 5 and in grades 6 through 12./18 As noted earlier, the social capital items that are available in the NHES:96 differ by the grade level of the children.

Logistic regression models were used to examine the influence of selected child, family, and school characteristics on high father and mother involvement in schools./19 To simplify the discussion, the results are presented separately for two-parent and single-parent families. Not all factors in the models are discussed. This section describes some of the major results of the models./20

Two-parent families. Because parents divide their time so as to best meet the needs of their own households, it is instructive to examine the models estimating high father and high mother involvement at the same time.

Involvement of the other parent. The levels of involvement of both parents in their children's schools are closely associated. Fathers are more likely to be highly involved as mothers' involvement increases and vice versa, though mothers' involvement exerts a somewhat stronger influence on fathers' involvement than the other way around. According to the results shown in table 2, there is an 84 percent increase in the adjusted odds that fathers of children in the 1st through 5th grade are highly involved in their children's schools with each unit increase in mothers' involvement, that is, with each additional activity that the mothers participate in. Similarly, with each unit increase in fathers' involvement there is a 65 percent increase in the adjusted odds that mothers of elementary school children are highly involved in their schools. The results also indicate that the association between mothers' and fathers' involvement grows stronger as children move from elementary school into the higher grades. Among children in the 6th through 12th grades, the adjusted odds that their fathers are highly involved in their schools increases by 175 percent with each additional activity that mothers participate in (table 3). The close association between the involvement levels of the two parents is probably indicative of shared values concerning the importance of education. That is, when parents share strong educational values, they tend to work in concert to support their children's educations by means of direct involvement. The stronger association between the involvement levels of parents of 6th through 12th graders may be due to a selection process whereby parents who value education the most are the most likely to remain involved in their children's schools as their children grow older.
Presence of a stepparent. Stepmothers are significantly less likely than biological or adoptive mothers to be highly involved in their children's schools, regardless of the children's grade level. The adjusted odds that the mothers are highly involved in their 1st through 5th graders' schools are 56 percent less if the mothers are stepmothers (table 2). Among children in grades 6 through 12, the adjusted odds are 57 percent less if the mothers are stepmothers (table 3 ). Fathers, on the other hand, are more likely to be highly involved in their children's schools if the mother in the household is a stepmother. The presence of stepmothers increases the adjusted odds that children's fathers are highly involved in their schools by 194 percent among children in grades 1 through 5 and by 197 percent among children in grades 6 through 12 relative to if the mothers are their biological or adoptive mothers. Thus, in families with stepmothers, fathers appear to assume a greater share of child-related responsibilities than they do when the children's biological or adoptive mothers are present.

Stepfathers also tend to be less involved in children's schools than biological or adoptive fathers, though the differences are only significant among children in grades 6 through 12. The adjusted odds that fathers are highly involved in their 6th through 12th graders' schools are 39 percent less if the fathers are stepfathers relative to if they are biological or adoptive fathers (table 3). Mothers of children in grades 6 through 12, like fathers, are significantly more likely to be highly involved in school if the father in the household is the children's stepfather. Thus, at least among older children, mothers assume even more of the child-related responsibilities when stepfathers are present. However, the increasing involvement of mothers of older children and of fathers, regardless of the children's ages, does not make up for the lower involvement of stepparents. A study that used NHES:93 data found that parental involvement in stepfamilies is, on average, lower than in families with two biological or adoptive parents (Zill and Nord, 1994).
From the data in the NHES:96, it is not possible to determine why stepparents tend to be less involved in children's schools. It is possible that stepparents, or the biological parents themselves, believe it is the biological parents' responsibility. It is also possible that children, particularly older children, discourage the involvement of their stepparents. These results are consistent with other research. A study based on the National Surveys of Families and Households found that biological fathers reported spending more time with their children when the mother in the household was a stepmother instead of the biological mother and that stepfathers reported spending less time with children than biological fathers (Cooksey and Fondell, 1996).

Household income and parents' education. Although the tabulations presented in figure 5 and figure 7 showed that household income and parents' education are both associated with high involvement of fathers and mothers in their children's schools, after controlling for other factors in the models the importance of household income is reduced. It remains a significant influence on high involvement in schools only among mothers of children in the 1st through 5th grades. Income has no influence on involvement in schools among mothers of older children or fathers of children in any grade level. On the other hand, education remains an important influence on high parental involvement in schools even after controlling for income and the other factors in the models./21 As fathers' and mothers' education increases, the adjusted odds that they are highly involved in their children's schools also increase. Among children in grades 1 through 5, there is a 19 percent increase in the adjusted odds that their fathers are highly involved in their schools with each unit increase in fathers' education./22 These results parallel those found in other studies, that is, that parental education is a more important influence on parental involvement than is income (Zill and Nord, 1994).

The NHES:96 data also reveal that it is fathers' education that influences fathers' involvement and mothers' education that influences mothers' involvement. Many studies use the education level of the most educated parent in the household or the mother's education when examining parental involvement in schools. Because the education of mothers and fathers are highly correlated, substituting one for the other is probably a good proxy.

Maternal employment. Mothers who work 35 or more hours per week are significantly less likely to be highly involved in their children's schools, regardless of the grade level, than are mothers who work part time. Among children in grades 1 through 5, there is a 51 percent reduction in the adjusted odds that their mothers are highly involved in their schools if their mothers work full time relative to if they work part time. However, these analyses also reveal that fathers whose wives work full time increase their involvement in their children's schools. Among children in grades 1 through 5, the adjusted odds that their fathers are highly involved increases by 40 percent if the mothers work full rather than part time. Among children in grades 6 through 12, fathers also appear more likely to be highly involved if the mothers work full time instead of part time, but the difference is only significant at the 0.10 level. These results suggest that families in which mothers work full time establish a different division of labor, with fathers sharing more of the child-related responsibilities, than families in which mothers work part time or not at all.
Children's age and sex. Whether children in grades 1 through 5 are boys or girls has no significant impact on the adjusted odds that either their fathers or their mothers are highly involved in their schools. However, among children in grades 6 through 12, the adjusted odds that children have highly involved fathers increase by 22 percent if the children are boys relative to if they are girls. This outcome suggests that fathers are more likely to increase their involvement if they have sons. Mothers of 6th through 12th graders, on the other hand, are somewhat less likely to be highly involved if the children are boys, though the association is only significant at the 0.10 level.
In a similar vein, fathers are less likely to be highly involved with middle school children (6th through 8th grade) than with high school children (9th through 12th grade). Recall from figure 4 that the proportion of children whose fathers volunteered at their schools increased between middle and high school. Mothers, on the other hand, are more likely to be highly involved with their 6th through 8th graders than with their 9th through 12th graders--a finding that also matches the results obtained earlier that showed mothers' involvement in their children's schools decreases steadily as the grade level of their children increases. This pattern of associations suggests a division of labor within two-parent families, with mothers focusing on younger children and on girls and with fathers spending somewhat more time with older children and with boys.
Family social capital. Even after controlling for the other factors in the models, the results confirm that parental involvement in schools and parental involvement at home are closely linked. Three measures of social capital were included in the models for children in grades 1 through 5. These were the number of in-home activities the children have shared with their parents in the past week, the number of out-of-home activities that the children have shared with their parents in the past month, and an indicator of whether the children have been told a story in the past week or have talked with their parents in the past month about family history. Among mothers of children in grades 1 through 5, all three measures of social capital are associated with high mother involvement in schools, even after controlling for the level of involvement of fathers and the other factors in the model. The adjusted odds that mothers are highly involved in their children's elementary schools increase by 16 percent with each additional activity participated in at home in the past week and increase by 38 percent with each additional outing that the family has gone on with the children in the past month.
Among fathers of children in grades 1 through 5, the number of activities in the home that their families have shared with the children in the past week and the number of outings that the families have shared with the children in the past month are significantly associated with high father involvement in the children's schools. With each additional activity the families have shared with the children in the past week, the adjusted odds that fathers are highly involved in their children's elementary schools increase by 17 percent. Similarly, with each additional type of outing the families have gone on with their children, the adjusted odds that fathers are highly involved in their children's schools increase by 25 percent.
More questions were available to measure social capital in the families of older children. Seven factors were included: parents' belief that the children will graduate from a 4-year college; parents' confidence that someone in the household can help the children with their homework; whether a parent has discussed future courses with the children in the past month; how often a parent helps the children with their homework; the number of activities shared with children in the past week; how often a parent attends religious services; and whether a parent regularly participates in an ongoing community service activity. Among children in grades 6 through 12, several of these items are significant influences on the odds that mothers and fathers are highly involved in their children's schools, though the specific social capital measures that are associated with parental involvement differ somewhat for mothers and fathers. Among mothers, expecting that the children will graduate from a 4-year college, the frequency with which parents help with homework, the frequency with which parents attend religious services, and having a parent who regularly participates in an ongoing community service activity are all significantly associated with high mother involvement in schools after controlling for the involvement of fathers and the other factors in the model (table 3)./23 The adjusted odds that mothers are highly involved in their children's schools are 43 percent higher among mothers who expect their children will graduate from a 4-year college compared to those who do not. Similarly, the adjusted odds that mothers are highly involved in their children's schools are doubled if a parent regularly participates in an ongoing service activity. However, it is possible that the ongoing service activity is involvement in their children's schools./24
Among fathers, the number of activities the families have participated in with the children, the frequency with which a parent helps with homework, and whether a parent regularly participates in an ongoing community service activity are associated with high father involvement. With each unit increase in the number of activities participated in with the children, the adjusted odds that fathers are highly involved in their 6th through 12th graders' schools increase by 16 percent. Similarly, with each unit increase in helping with homework, the odds that fathers are highly involved in their children's schools increase by 15 percent./25 There is a 28 percent increase in the odds that fathers are highly involved in their 6th through 12th graders' schools if a parent in the household regularly participates in an ongoing service activity.
School characteristics. The measure of school climate/26 is more consistently related to the odds of mothers and fathers being highly involved in their children's schools than any of the other school characteristic factors. Although school climate has no significant influence on the adjusted odds that fathers of children in grades 1 through 5 will be highly involved in their children's schools, it is significantly associated with high father involvement among children in grades 6 through 12 and with high mother involvement at all ages. With each unit increase in the scale, the adjusted odds that fathers of children in grades 6 through 12 are highly involved in their children's schools increase by 4 percent. Although 4 percent may not seem large, consider that the adjusted odds that fathers are highly involved in their children's schools are 20 percent higher if respondents strongly agree (a value of 4 on each of the 5 items) compared to if they agree (a value of 3 on each of the 5 items) with each statement.
Whether a school is public or private is also associated with parental involvement. The adjusted odds that fathers of children in the 1st through 5th grades are highly involved in their children's schools are 48 percent greater if their children attend private as opposed to public schools that they were assigned to. Whether a school is public or private has no effect on the odds that mothers of elementary school children are highly involved in their schools. This is probably due to the fact that the majority of mothers of elementary school children are highly involved in their children's schools, as shown earlier in the report. However, the odds that mothers are highly involved in the schools of their 6th through 12 graders increase by 93 percent if the children attend private rather than public schools that they were assigned to. There is evidence that mothers are more involved if their 6th through 12th graders are enrolled in a public school of their choice compared to if they are enrolled in a public school that they were assigned to. The odds that mothers of children in the 6th through 12th grades are highly involved in their schools are 27 percent higher if the children attend a public school of their choice as opposed to one that they were assigned to.
Children get mostly A's. Because some researchers have noted that parent involvement is higher if parents have a high assessment of their children's abilities (Eccles and Harold 1996), additional models were estimated that included parents' reports of their children's usual grades (data not shown). The result of these models suggest that fathers are more likely to be highly involved in their 6th through 12th graders' schools if their children are doing well academically. The adjusted odds that fathers are highly involved in their 6th through 12th graders' schools are 30 percent higher when parents report that their children get mostly A's than when they do not. Because the NHES is a cross-sectional survey, it is not possible to determine the direction of causation. It is equally possible that children are more likely to do well in school because their fathers are involved in their schools. It is quite likely that causation runs in both directions, with fathers more likely to be involved when their children are doing well and children doing better when their fathers are involved. There is no association between children getting mostly A's and mothers' involvement among younger or older children or fathers' involvement among children in elementary school.

Single-parent families. Fewer factors are important influences on high father or mother involvement in single-parent families after controlling for the other factors in the models.

Household income and parents' education. Once education and the other factors are controlled, household income has no influence on whether single fathers or mothers are highly involved in their children's schools. Parents' education, on the other hand, remains a significant influence on the likelihood that mothers and fathers are highly involved in their children's schools, except among fathers of children in grades 6 through 12. The adjusted odds that fathers of 1st through 5th graders are involved in their children's schools increase by 67 percent with each unit increase in the education measure (table 4). Similarly, the adjusted odds that mothers are highly involved in their children's elementary schools increase by 42 percent and the adjusted odds that they are involved in their 6th through 12th graders' schools increase by 15 percent with each unit increase in education (table 4 and table 5).
Children's age and sex. Unlike fathers in two-parent families, single fathers are more likely to be highly involved in their 6th through 12th graders' schools if the children are in the 6th through 8th grade rather than high school. However, when information on whether schools offered general school meetings or parent-teacher conferences is added to the model (not shown), this association becomes insignificant. There is no association between the children's grade level and mothers' involvement in their schools. There is also no association between the sex of the children and the involvement of either single mothers or single fathers in their schools.
Family social capital. Several of the social capital measures are important influences on the adjusted odds that single mothers are highly involved in their children's schools. Fewer of them influence the adjusted odds that single fathers are highly involved. Among children in grades 1 through 5, the odds that their mothers are highly involved in their schools increase by 30 percent with each additional activity they have shared with their mothers in the past week, by 60 percent with each additional outing they have gone on with their mothers in the past month, and by 73 percent if their mothers have told them a story in the past week or have discussed their family history in the past month. None of the social capital measures are significant influences on high father involvement among children in grades 1 through 5.
Among children in grades 6 through 12, the adjusted odds that mothers are highly involved in their schools are greater as the number of activities the mothers have participated in with their children in the past week increases, as the frequency with which they attend religious services increases, and if they regularly participate in community service activities (table 5). The adjusted odds that single fathers are highly involved in their 6th through 12th graders' schools are significantly greater if they have discussed future high school courses or plans after high school with their children in the past month. Fathers also are more likely to be highly involved in their 6th through 12th graders' schools if they expect that their children will graduate from a 4-year college and as the number of activities they have participated in with their children in the past week increases. However, these factors are only significant at the 0.10 level.
School characteristics. As in two-parent families, school climate is an important influence on the involvement of single mothers and single fathers in their children's schools, especially when their children are in elementary school. The odds that single fathers and single mothers are highly involved in their 1st through 5th graders' schools increase by 35 percent and 15 percent, respectively, with each unit increase in the school climate scale, after controlling for measures of social capital in the family, whether the schools are public or private, and other factors in the model (table 4).
Among children in grades 6 through 12, school climate only influences whether single mothers are highly involved in their children's schools. It has no influence on single fathers. The odds that single mothers are highly involved in their children's schools increase by 15 percent with each unit increase in the school climate scale.
There is some evidence that single fathers are more likely to be highly involved in their 6th through 12th graders' schools if the schools are private or are public schools of their choice as opposed to public schools that they were assigned to. However, the relationships between these factors and high father involvement are only significant at the 0.10 level. There is also weak evidence that single mothers are more likely to be highly involved in their 6th through 12th graders' schools if the schools are private (significant at the 0.10 level). However, it appears that single mothers may be somewhat less likely to be highly involved in their children's schools if the children attend public schools of their choice.
Children get mostly A's. Children making mostly A's only has a significant influence on the adjusted odds that single mothers are highly involved in their 6th through 12th graders' schools (data not shown). It has no association with single fathers' involvement at any grade level or with single mothers' involvement in elementary school. As noted earlier, the causal relationship between children doing well in school and high parental involvement is unclear, though it is likely that the two influence each other.

Parental Involvement and Student Outcomes

Much of the research on parental involvement in schools has focused on its influence on students' academic success (Henderson, 1987). Academic success can be measured in a variety of ways. This report uses a measure of academic success and a measure of academic difficulties: getting mostly A's /27 and having ever repeated a grade. Other facets of children's school lives are also important to their social and emotional development. Among these are the extent to which they enjoy school and their involvement in extracurricular activities. Children who enjoy school are more likely to perform better academically and to remain in school (Csikszentmihalyi, 1990). Participation in extracurricular activities reduces risky behaviors in adolescence, such as dropping out of school, becoming a teen parent, using drugs, or engaging in delinquent conduct (Zill, Nord, and Loomis, 1995). It also provides more opportunities to establish connections to other young people and adults. Children's behavior in school is another important school outcome. Parents of children in the 6th through 12th grade were asked if their child had ever been suspended or expelled.

School-aged children in both two-parent and single-parent families are more likely to get mostly A's, to enjoy school, and to participate in extracurricular activities and are less likely to have ever repeated a grade and to have ever been suspended or expelled if their fathers or mothers have high as opposed to low levels of involvement in their schools (figure 13). However, the differences in the proportion of children in father-only families who have repeated a grade is not significantly different for children whose fathers have high and low levels of involvement.

Of course, in two-parent families in which fathers show high levels of involvement, it is likely that mothers also have high levels of involvement. Thus, the association between fathers' involvement and children's outcomes may be due in part to high levels of involvement by the mother. Figure 14 shows these same child outcomes by whether neither parent, only the mother, only the father, or both have high involvement. This figure reveals that it makes little difference whether it is only the mother or only the father who has high involvement; as long as one of them is highly involved, children have better outcomes than if neither have high involvement; as long as one of them is highly involved, children have better outcomes than if neither have high involvement. Moreover, children have the most favorable outcomes if both of their parents exhibit high involvement. Although the advantage is relatively small, the differences between having both parents highly involved in the children's schools and having only the mothers highly involved are evident for participation in extracurricular activities, getting A's, enjoying school, and having ever repeated a grade. The question of whether mothers and fathers make independent contributions to these outcomes is explored in multivariate models below.

In order to understand the contribution of fathers' involvement in their children's schools to student outcomes, it is important to control for other factors that also influence how students do in school. Logistic regression models were estimated to examine the influence of mothers' and fathers' involvement on the five student outcomes after controlling for related child and family characteristics. In the tables presented below, only the adjusted odds of the outcomes by mothers' and fathers' involvement are shown. The adjusted odds ratios for all the factors contained in the models are contained in Appendix B. To simplify the discussion, results are presented separately for children living in two-parent and in single-parent families.

Two-parent families. Table 6 presents the adjusted odds that children in two-parent families get mostly A's, enjoy school, participate in extracurricular activities, have ever repeated a grade, and have ever been suspended or expelled from school as categorized by their fathers' and mothers' involvement in their schools, after controlling for a variety of related factors. Among the factors that were also included in the models are the children's race and ethnicity, sex, and grade level, mothers' and fathers' education, household income, family type, and maternal employment. In addition, models 3 and 5 also control for different measures of social capital within the families ./28

Get mostly A's. Children are more likely to get mostly A's if their fathers are involved in their schools. Among children in grades 6 through 12, the adjusted odds that children get mostly A's increase by 46 percent if fathers are highly involved in their schools and by 21 percent if fathers are moderately involved in their schools compared to if the fathers have low levels of involvement (model 4). Even after controlling for measures of social capital in the family, the odds that children get mostly A's are 43 percent higher if their fathers are highly involved in their schools compared to if they are not very involved (model 5). Mothers' involvement in their children's schools also influences the odds that the children get mostly A's, but mainly among children in grades 6 through 12. Once measures of social capital are entered into the models, mothers' involvement is no longer a significant influence. These results indicate that fathers' involvement in their children's schools exerts a distinct and independent influence on children making good grades and that the association is not due to the fact that mothers tend to be involved when fathers are involved. The results also suggest that for this particular outcome, fathers' involvement is more important than mothers'.
Enjoy school. Children, at least according to their parents' reports, enjoy school more when their fathers and mothers are involved. The adjusted odds that children in the 1st through 12th grade enjoy school are 30 percent higher if the fathers are moderately involved and 55 percent higher if they are highly involved relative to if they have low involvement in the schools (model 1). A similar increase in the adjusted odds occurs when mothers are involved in their children's schools (model 1). The association between fathers' and mothers' involvement and children's enjoying school is apparent at all grade levels.
Participate in extracurricular activities. Children are more likely to participate in extracurricular activities when their mothers and fathers are involved in their schools. Because it is not possible to determine the direction of causation, another interpretation is that parents are involved because their children are participating in sports teams, orchestras, or other extracurricular activities that draw the parents to the schools as spectators, coaches, or advisors. Among children in grades 6 through 12, the adjusted odds that they participate in extracurricular activities are 48 percent higher if their fathers are moderately involved in their schools and are 88 percent higher if their fathers are highly involved in their schools (model 4). The odds that they participate in extracurricular activities are 146 percent greater if their mothers are highly involved in their schools relative to if their mothers have low involvement (model 4). Adding information on social capital in the home to the models reduces somewhat the influence of parental involvement on the odds that children participate in extracurricular activities (model 5).
Ever repeated a grade. The involvement of mothers and fathers, particularly mothers, is also important in reducing the likelihood that children in elementary school have ever repeated a grade. Among children in grades 1 through 5, the adjusted odds that they have repeated a grade are 35 percent lower if fathers are moderately involved and 34 percent lower (significant at the 0.10 level) if they are highly involved. The odds that children have ever repeated a grade are 50 percent lower if mothers are moderately involved and 45 percent lower if they are highly involved in their children's schools relative to if the mothers have low involvement. Among children in grades 6 through 12, the odds that they have ever repeated a grade are lower if their fathers are moderately or highly involved in their schools or their mothers are highly involved. These associations, however, are only significant at the 0.10 level.
Ever suspended or expelled. Mothers' involvement in school, but not fathers' involvement, reduces the likelihood that 6th through 12th graders have ever been suspended or expelled from school. The adjusted odds that children have ever been suspended or expelled are 24 percent lower if their mothers are moderately involved in their schools (significant at the 0.10 level) and are 43 percent lower if their mothers are highly involved in their schools relative to if their mothers have low levels of involvement. The lack of association between fathers' involvement and the likelihood that children have ever been suspended or expelled may be due to the fact that some fathers become involved because their children are having behavioral problems.

Single-parent families. Table 7 shows the adjusted odds that children get mostly A's, enjoy school, participate in extracurricular activities, have ever repeated a grade, and have ever been suspended or expelled from school by the involvement of their mothers or fathers, after controlling for a variety of child and family characteristics. The other factors that were included in the models are the child's race/ethnicity, sex, parents' education, and household income. The single-mother families also include information on maternal employment. Because most fathers are employed full time, information on fathers' employment was not included in the models. Models 3 and 5 include measures of social capital in the families./29

Get mostly A's. Children in the 6th through 12th grade who live in single-parent families are more likely to get mostly A's if their parents are involved in their schools. In single-father families, the adjusted odds that children get mostly A's are twice as high for children whose fathers are highly involved in their schools compared to children whose fathers show low levels of involvement. Much of the association is due to the fact that such fathers are also involved at home as well. Once the social capital measures are added to the models, the influence of fathers' involvement in schools on children getting mostly A's is no longer significant. The importance of mothers' involvement in schools to children getting mostly A's only becomes significant after measures of social capital are added to the model. The adjusted odds that children in the 6th through 12th grade living in single-mother families get mostly A's are 70 percent greater if their mothers are highly involved rather than having only low levels of involvement in their schools. For children in grades 1 through 5, parents' involvement is not associated with making mostly A's after controlling for the other factors in the models.
Enjoy school. Single fathers' involvement in their children's schools is not associated with whether the children enjoy school. However, there is an association between single mothers' involvement and children's enjoyment of school. The association, however, is not consistent. Among children in grades 1 through 5, it appears that children whose mothers are moderately involved have a reduced likelihood that they enjoy school after measures of social capital are entered into the model. Among children in grades 6 through 12, children are more than twice as likely to enjoy school if their mothers are highly involved, but the relationship is no longer significant once measures of social capital are added to the model.
Participate in extracurricular activities. Single fathers' involvement has a weak association with whether their 6th through 12th graders participate in extracurricular activities (significant at the 0.10 level) and no influence on whether their 1st through 5th graders participate in extracurricular activities. The involvement of single mothers in their children's schools increases the likelihood that their 6th through 12th graders participate in extracurricular activities. Children in grades 1 through 5 are also more likely to participate in extracurricular activities if their mothers are highly involved in their schools (model 2).
Ever repeated a grade. The involvement of single mothers, but not single fathers, reduces the odds that children have ever repeated a grade among children at all grade levels. The adjusted odds for children in the 1st through 5th grade are 58 percent lower and the adjusted odds for children in grades 6 through 12 are 42 percent lower if their mothers are highly involved in their schools compared to if they have low levels of involvement.
Ever suspended or expelled. The involvement of both single fathers and single mothers reduces the adjusted odds that their 6th through 12th graders have ever been suspended or expelled from school, though the influence of mothers' involvement is no longer significant once the social capital measures are added to the models. If single fathers are highly involved in their children's schools, the odds that children have ever been suspended or expelled are 72 percent lower than if the fathers have low levels of involvement.

Introduction

Findings (next part)