March 27,1996
Washington, D.C.
Prepared for the
NICHD Family and Child Well-Being Research Network
by
Child Trends, Inc.
Angela Dungee Greene
Carol Emig
Gesine Hearn
May 1, 1996
- ABSTRACT
- PREFACE
- INTRODUCTION
- The Need for Research on Fatherhood and Male Fertility
- A Public-Private Partnership to Develop Better Information on Fathers
- TOWN MEETING ON FATHERING AND MALE FERTILITY
- The Importance of Data on Fatherhood and Male Fertility to the Development of Public Policies That Support Children and Families
- Welfare Reform
- Child Support
- Issues Related to Male Fertility
- Developing a Conceptual Framework for Studying Fatherhood and Male Fertility
- Expanding the Concept of Fatherhood
- Multiple Family Forms
- Fathering in a Larger Social Context
- The Investment Paradigm
- A Focus on Child Well-Being
- A Framework for Studying Fragile Families
- The Convergence of Practice and Research
- Survey Design and Data Collection
- The Problem of Undercounting Males
- Men as Survey Respondents
- Ways to Improve Data Collection and Survey Design
- Survey Content
- Low-Income Fathers' Financial Contributions to Their Children
- Father Involvement After Divorce
- Parenting Behaviors of Young Fathers
- Understanding the Role of Fathers
- The Fiscal and Political Challenges to Improving Data on Fathers
EFFORTS TO IMPROVE FEDERAL DATA ON FATHERS
March 1996
The family structures in which American children are raised and cared for have changed dramatically during the past decades. We observe great variety in family situations and structures and, due primarily to increases in divorce and nonmarital childbearing, more than half of all children are likely to spend some portion of their childhood in a single-parent family. The great diversity in family structures implies a great diversity in the roles of mothers and fathers. However, while there has been considerable research on the role of mothers in the family, relatively little is known about the roles of contemporary fathers and the importance of fathering within the contemporary family. When they are engaged with their children, fathers can play a key role in their development of children; they can provide emotional and psychological support for mothers and children; and they can contribute monetary and physical support to the family. But beyond identifying these broad functions, the available research literature provides few additional insights into the nature of fatherhood in America today.
Thus, fatherhood is not only a socially and politically important issue, it is a role that has undergone substantial and rapid change. Moreover, scientific knowledge about this issue is deficient, and social programs frequently ignore the contribution of fathers to their children and families. More information and stronger theoretical concepts are needed to assess the importance of the roles fathers can play in their children's lives. Existing large-scale data bases fail to provide reliable and detailed information on men's family life. Recently, President Clinton asked for more government leadership on the topic of fatherhood. The Fatherhood Initiative, as directed by the Domestic Policy Council and Vice President Gore's National Performance Review, strives to foster and support more involvement of fathers in the lives of their children and more information about fathers.
The current effort to improve research and data regarding fathers is comprised of four interlocking sets of activities. The Fatherhood Initiative itself, as noted, is a broad range of activities coordinated by the White House in which all federal agencies are actively participating. The second set of activities arises out of a newly-created organization called the Federal Interagency Forum on Child and Family Statistics, comprised of the major agencies within the federal government charged with the responsibility for assembling information on families and children. The Forum is very interested in fatherhood and has charged its Data Collection Committee to suggest ways to improve information about fathers. The third set of activities arises out of the Family and Child Well-being Research Network, a consortium of seven scholars funded by NICHD for five years. The purpose of the Network is to create a public/private research partnership to improve knowledge about a wide spectrum of issues involving families and children. The Network has chosen fatherhood as a major emphasis area. The fourth set of activities has been initiated outside the federal government through the National Center on Fathers and Families. The Center is involved in program development, policy research and dissemination activities around father involvement. Core support for the center is provided by the Annie E.Casey Foundation with additional programmatic support provided by the Ford Foundation.
The activities of these different efforts are being coordinated to ensure that concrete recommendations will be made for changes on federal data collection efforts and research on fathering. In an effort to address these measurement issues, several projects are planned for the next year:
- A town meeting on male data on March 27, 1996, to hear from and discuss different perspectives on the issues of becoming a father, being a father, and parenting. The Federal Interagency Forum on Child and Family Statistics is inviting speakers to give short testimony on methodological, theoretical, and political problems concerning male data and also to submit a written document. The roundtable discussion with Forum members and speakers will be open to a wide audience. Representatives from the major federal statistical agencies, universities, and private foundations will participate in this meeting. Written testimony will also be invited from persons not attending.
- A workshop on qualitative research on men and fathers will be held June 11-12. Small-scale studies on parenting using clinical, ethnographic, and anthropological approaches will be presented and assessed for information regarding fatherhood. Ways to integrate approaches and findings from small-scale studies into large-scale surveys will also be discussed. Relevant findings of this workshop will be presented at the Network conference in October. Federal Interagency Forum members, the Family Research Network and the National Center on Fathers and Families are being asked to participate.
- A conference on fatherhood will be held on October 10 - 11, sponsored by the NICHD Family and Child Well-Being Research Network. At the conference, analytic studies using existing sources of information on fatherhood from both federal statistics and other surveys will be presented and critically reviewed. The Federal Interagency Forum and National Center for Fathers and Families will again be invited to participate. There will also be a half-day meeting on methodological issues concerning men and fathers in large-scale surveys the day after this conference.
- A working conference on measurement and data collection issues concerning male data in large-scale surveys is planned for early 1997. This conference will profit from the Network conference, the small-scale studies workshop, and the roundtable on male data. The Federal Interagency Forum and NICHD will co-sponsor and organize this conference with the National Center for Fathers and Families and the Network participating. The outcome will be recommendations for changes in the federal statistical system that would improve the collection of information on fathering.
Two additional activities are still in the planning stage. The first is a methodological workshop bringing together researchers who have worked to collect better data on fertility. Second, the Department of Health and Human Services is planning a symposium that will focus on the possibility of including father related information in administrative data.
A critical characteristic of these efforts is that they rely on a private/public partnership that includes the federal government, university-based scholars, private nonprofit organizations, and charitable foundations. If you would like to be involved in these efforts or to be kept apprised of these efforts, please let planners know by contacting Gesine Hearn by phone at (301)496-1174 or by e-mail at hearng@hd01.nichd.nih.gov.
Interest in fathers, and specifically in their contribution to children's well-being, has grown in recent years. President Clinton has asked federal agencies to assume greater leadership in promoting father involvement in the lives of children and, with this goal in mind, has directed the federal government to examine existing federal data and data collection efforts. As a result, the major federal statistical agencies are reviewing their current approaches to gathering information on fathers and exploring new ways of conceptualizing, measuring, and collecting data about fatherhood and male fertility.
On March 27, 1996, the Federal Interagency Forum on Child and Family Statistics sponsored a Town Meeting on Fathering and Male Fertility, the first in a series of related federal meetings and activities to develop better data on fathers. The Forum, consisting of the major federal statistical agencies, fosters coordination, collaboration, and integration of federal collection of data on child and family issues and federal reporting of child and family conditions. At the Town Meeting, invited speakers presented ideas for improving the federal statistical system's capacity to gather data on fathers to an expert panel that included the leadership of the federal agencies belonging to the Forum. (Panelists and invited presenters are listed in Figure 2 of this report.)
In the course of the next year, other meetings and activities sponsored by the federal government will build on the information gathered at this Town Meeting, culminating in a March 1997 conference to present an agenda for improving federal data on fathers. In addition, two private sector efforts are coordinating closely with the federal initiative to improve data on fathers -- work by the National Center on Fathers and Families (NCOFF) at the University of Pennsylvania and work by the Family and Child Well-Being Research Network, an interdisciplinary consortium of seven scholars supported by the National Institute of Child Health and Human Development that seeks to advance data-based research on children and families and bring it to the attention of the public policy community.
Special thanks for planning and organizing the Town Meeting are given to Linda Mellgren and Anne Benson (Office of the Assistant Secretary for Planning and Evaluation at the Department of Health and Human Services); Jeffery Evans, Christine Bachrach, and Judy Whalen (National Institute of Child Health and Human Development); Kristin Moore (Child Trends, Inc.); Gesine Hearn (NICHD Family and Child Well-Being Research Network; and Freya Sonenstein (The Urban Institute). Support for the Town Meeting was provided by NICHD, the Family and Child Well-Being Research Network, and the Annie E. Casey Foundation.
This report, prepared by Angela Dungee Greene and Carol Emig (Child Trends, Inc.) and Gesine Hearn (NICHD Family and Child Well-Being Research Network), is a summary of the presentations made at the Town Meeting and of submitted written testimony. Subsequent meetings and conferences will add to and refine the continents and suggestions offered here.
The composition of the American family and the traditional roles and responsibilities of family members have changed dramatically in recent decades. In particular, the proportion of female-headed households has increased as more marriages dissolve and more women bear children outside of marriage. Moreover, in an increasing number of households, mothers and fathers -- regardless of marital status--are employed, so that the roles of breadwinner and primary caregiver are much less likely to be determined by gender than in the past.
These changes have led policymakers, scholars, and child and family advocates to seek greater information on the role of fathers in family life and in the lives of children. While there has been considerable research on the role that mothers play in the life and functioning of a family and in the well-being of children, information on fathers is not as extensive. Society has traditionally assigned to the father the role of primary economic provider, but as noted above, this is a role he increasingly shares with his children's mother -- or in some cases fails to fulfill. Fathers may also play other key roles in encouraging children's development and in providing emotional and psychological support to a spouse and to children. But the extent to which fathers do this, and whether and how their contribution is unique, need further exploration since relatively little data exist on the role of fathers in their children's lives and within the family.
Improved information on fathers would potentially meet a number of important societal needs. It could contribute to the development of public policies related to welfare, child custody, and child support; it could provide important insights and guidance to parents and to professionals concerned with child development; and it could assist employers interested in developing workplace policies that help their employees balance their family and work responsibilities.
In response to increased interest in issues related to fatherhood, the federal government, major foundations, and leading researchers and research centers have launched a series of interrelated activities intended to generate more and better information about fathers and to support greater involvement by fathers in their children's lives
At the behest of the President, the White House has launched the Fatherhood Initiative, directed by the Domestic Policy Council and the Vice President's National Performance Review. The Initiative, involving all federal agencies, promotes greater involvement by fathers in the lives of their children and supports the need for improved data on fatherhood and male fertility. In particular, federal agencies are charged by the President to incorporate fatherhood issues wherever appropriate in government-initiated research on children and families.
The Federal Interagency Forum on Child and Family Statistics, which includes six federal agencies with major responsibility for gathering information on families and children, has taken the lead in efforts to improve the quality and quantity of data on fathers gathered by the federal government. With the support of federal agencies, private foundations, and nongovernment researchers, the Forum is sponsoring or participating in a series of activities related to data collection on fatherhood and male fertility. These activities are described in Figure 1.
In the private sector, several foundations with strong commitments to improving child well-being and strengthening family life have invested significant resources in programmatic efforts to strengthen fathers' engagement with their children and in research to understand better the roles that fathers play in family life and the factors that affect their ability and willingness to fulfill their parenting responsibilities. In the research community, several efforts are underway that complement and advance federal efforts to develop an appropriate research agenda to improve understanding of issues related to fatherhood and male fertility, particularly as they affect existing or proposed public policies related to welfare, child support, child custody, and other child and family issues.
This report summarizes the major findings from the Town Meeting on Fathering and Male Fertility. This meeting was the first of the series of conferences described above and the starting point for a year-long process to improve information collected on men and fatherhood. The process will assess the nature, scope, and quality of federal data on fathers, and develop an action agenda for improving the federal government's capacity to conceptualize, gather, and measure information from and about men and fathers. The Town Meeting was neither a definitive assessment of the federal data system nor a complete agenda for improving data on fathers. Rather, it is the first step in examining the many conceptual, methodological, and practical issues related to improving information on fathers. Subsequent meetings will add to and build upon the insights and suggestions offered at the Town Meeting.
The Town Meeting provided an opportunity for the leadership of the major federal agencies responsible for gathering data on children and families to hear from a cross-section of researchers, state officials, Congressional staff, and advocates on the importance of gathering better data on fatherhood and male fertility. Invited presenters offered oral testimony and responded to questions; written testimony was also welcomed. (Federal agency representatives and invited presenters are listed in Figure 2.) The town meeting provided a structured forum for government officials and those who use federal data to begin to explore the strengths and limitations of the existing statistical system, ways to improve that system, and strategies for implementing these improvements in the face of budget constraints.
Presentations and discussions at the Town Meeting fell generally within five broad categories:
- The relevance of data on fathering and male fertility to the development of public policies that have significant effects on the well-being of children and the strength of families;
- The conceptual framework that should guide the collection of data on men;
- The limitations of existing data on men and ways to improve that data;
- Issues for inclusion in future surveys of men; and
- The fiscal and political challenges to improving data on fathers.
The increase in mother-only households and the absence of fathers in the lives of many children have important implications for public policies intended to support positive child development and to strengthen families. In particular, concern over the growth in mother-only households and the precarious economic status of many of these families has spurred a sustained national movement to change fundamentally the nation's chief welfare program, Aid to Families with Dependent Children.
Offner began his testimony by calling men the forgotten party in welfare reform. Mothers, as the recipients of income support and the targets of job training and employment programs, are the chief focus of both existing and emerging welfare policies. In Offner's view, many governors and state program administrators would welcome opportunities, through welfare reform, to reconnect men with their children. In particular, the emphasis on work in virtually all welfare reform proposals offers the possibility of reinvolving men. As an example, Offner suggested that a state welfare program might permit and encourage mothers and fathers to decide together which one will fulfill a work requirement and which one will provide child care. But developing policies that involve fathers in key decisions about their children's financial support and general well-being requires that policymakers understand more than they currently do about the relationship between welfare mothers and the fathers of their children. For example, how many of these men have sustained relationships with the mothers and/or their children? How many currently work or have skills that make them employable? What is their income? Existing state-level data do not provide sufficient information on fathers to allow states to develop programs and policies that they can be reasonably sure will be cost effective. Offner contended that better information on fathers from the federal government would enable governors and program administrators to focus on fathers as well as mothers in welfare and employment programs.
Several presenters pointed out that better information on fathers could also help inform the design of child support policies that optimize absent parents' willingness to pay support. Sorensen noted that, unlike welfare reform, reforming child support policy to strengthen child support enforcement has widespread support. Yet very little is known about noncustodial fathers' ability or willingness to pay additional child support. Instead, Sorensen pointed out that policy reforms tend to be based on the stereotypical image of noncustodial fathers as "deadbeat dads;" but this image does not recognize the diversity among noncustodial fathers' ability to pay child support. For example, the following questions cannot be answered from existing data: How many noncustodial fathers have a limited ability to pay child support? How many of them have children receiving public assistance? What impedes these fathers from being able to pay meaningful levels of child support? Campbell also noted that no explanatory data exist at present to determine whether retention by the state of all or part of an absent parent's child support payment to offset the cost of a family's's AFDC grant affects his willingness to pay child support.
Sonenstein noted a strong policy interest in three questions related to male fertility. First, a better understanding of men's views about childbearing and childrearing is essential to promoting responsible parenting. Second, given the importance of condom use in preventing HIV infection and other sexually transmitted diseases and in preventing pregnancy, she noted the need for more and better information on male fertility behavior, particularly the factors that facilitate or discourage condom use. Third, there is growing policy interest in the nature of relationships between adult males and teenage females, since half of teen mothers report that the fathers of their children are men over the age of 20. Yet no data are available that adequately describe the characteristics of these older males or the nature of their relationships with teenage females.
Conceptual frameworks facilitate research by posing testable hypotheses and suggesting methods of inquiry. In exploring the development of a conceptual framework for the study of fathers, presenters at the Town Meeting offered some broad conceptualizations of fathers and families, but also considered a number of highly focused issues, such as the implications of different family forms for father involvement. In general, there was considerable agreement or complementarity among presenters. In particular, there was consensus on the importance of expanding the existing conceptual view of men and their roles as fathers and potential fathers.
There was general consensus among presenters that the study of fathers and fathering lacks an adequate conceptualization of what it means to be a father. Several presenters were critical of the tendency in existing survey research to study fathers largely with measures designed for mothers. They pointed out that questions developed for women about family structure and parent-child interaction do not fully capture the ways men interact with their children and the material support they provide to children and their mothers. For example, by applying the traditional mother template to fathers, researchers may focus on the time fathers spend engaged in activities like child care and housework, to the exclusion of other activities that often are performed by fathers, such as providing financial support, running errands, and cutting the grass. In contrast, surveys of nonresident fathers tend to overlook questions related to time spent with children, and instead focus largely on their financial contributions to their children. As Horn noted, a broader conceptualization of what it means to be a father would ask both resident and nonresident fathers about their interactions with their children and their economic and other contributions to the household. Ingram, in submitted written testimony, also advocated a conceptual framework that captures both economic and emotional contributions by fathers to their children.
Horn recommended a broad conceptualization of fathers that captures the diverse contributions of both resident and nonresident fathers, and suggested using a model of father involvement developed by Michael Lamb, James Levine, and Joseph Pleck as a potential framework for questionnaire development. This model has three components: 1) engagement, which refers to how much time the father spends in direct interaction with the child in the form of caretaking or play; 2) accessibility, which refers to how available the father is to the child, but is different from direct interaction; and 3) responsibility for the care of the child, which addresses the child's material needs.
Several presenters urged that greater attention be paid to differences among fathers in their attitudes and behaviors toward their children and the reasons for these differences. Braver's study of divorced fathers showed that they differed significantly from never-married fathers in terms of their child support efforts, their level of involvement with their children, and the dynamics of their relationship with their children. The two groups also differed markedly in their basic demographics and economic status.
Fathering may have a unique meaning and intrinsic value for fathers that affects their behaviors and interactions with their children. Mincy observed that when a low-income man becomes engaged in the life of his child, he is more inclined to improve his personal circumstances by participating in employment training, literacy training, or substance abuse programs. Braver noted that divorced fathers who believe that they continue to share childrearing decisions with the mother are more likely to pay child support and remain actively involved with their children than are their "parentally disenfranchised" counterparts. Flor, in submitted written testimony, similarly noted that fathers who are given equitable access to their children after divorce are more likely to pay child support on time and in full. These interactions have not been fully explored.
As Levine and others pointed out, men are connected to their children in different ways, through different family formations. Accordingly, there needs to be -- and is beginning to be--a shift away from the two-parent, middle-income, white family as the sole point of reference and toward a greater examination of multiple family configurations, particularly in the context of different ethnic groups. Campbell noted that family structures vary tremendously and there may be different parent-child interactions in each. She cited, for example, never-married couples, and couples who may have married after their child's birth. Similarly, Horn asserted that researchers tend to study resident fathers, divorced fathers, and never-married fathers in relative isolation, and fail to observe the movement into and out of these categories across the lifespan. A lifespan approach, he noted, would more effectively capture transitions over time.
Fathers perform their roles in a large social context consisting of many interrelated components. Several presenters highlighted the importance of accounting for these varied contextual factors in the study of fathers. For instance, Sweet described numerous interrelated spheres or aspects of family life, including spousal or partner relationships, kin networks, parenting, financial matters, and distribution of household chores and tasks. This multi-dimensional family life, in turn, intersects with other important social institutions and broader networks, such as the workplace or job market, religious organizations, voluntary associations, and friendships. Parenting behaviors are best understood as they relate to the many aspects of family life and to the broader society in which families function. Sweet indicated that this holistic approach is the theoretical underpinning of the National Survey of Families and Households, which measures important aspects of fathers' roles and related contextual factors.
Michael proposed a paradigm that considers the investments by parents in children. Referring to the various sources of money in both noncustodial situations and intact families, such as parents' earnings, unearned property income, noncustodial parental support, and government transfer payments--he questioned whether different sources of income affect children in different ways, if income effects vary among children, and whether any effects are related to family structure. In particular, he noted that there is very little information about the allocation of resources to children in "blended" families in which some of the children are from prior marriages. In addition, Michael suggested a focus on how parents contribute to "outputs"; or characteristics of their children rather than how much time parents spend with them. For instance, because fathers interact with their children differently than mothers, they may contribute more to the development of certain characteristics in children, such as poise and self-confidence, entrepreneurship, and long-term planning.
Concern about the well-being of children is an important reason for studying fathers. Campbell applauded the Town Meeting and other recent efforts to focus on fathers, but reminded the panel that promoting child well-being also required continued work around mothers and their issues. Both fathers and mothers contribute to their children's well-being; one parent is not inherently more or less significant than the other. In fact, traditional roles of caretaker and homemaker for mothers and economic provider for fathers are now often shared by or interchanged between two parents, or are assumed by one parent in a single-parent family. Campbell also proposed a shift in emphasis away from factors associated with family dissolution and dysfunction, and toward an examination of the dynamics that promote healthy and viable families of various formations. Such an approach might yield insights into what creates successful families, marriages, and parents that could, in turn, inform efforts to assist troubled families.
Mincy testified that the traditional model of a family, in which a father, mother, and children reside together in a household formed by marriage, guides most existing data collection efforts and public policy decisions. This model, in which cohabitation, conception and childbirth all occur within the marital arrangement, has several serious shortcomings when used to study fathers, mothers, and children in low-income communities. For example, it does not address the process by which never-married families are formed, and it prevents analysts from studying poverty among low-income men who live with their parents or in someone else's household. It also makes it difficult for researchers to study actual or potential relationships between unmarried parents and between these parents and their children.
Men in the traditional model of families are generally employed at wages that have enabled them to marry, establish households, and support children. However, in low-income communities and among individuals who are only marginally connected to the labor force, fertility is rarely contingent on marriage and employment. For example, some adolescent males father children, and some adult men father children by young girls, possibly because their poverty or near-poverty precludes marriage to adult women. Thus, cohabitation, conception, and childbirth frequently occur outside of marriage, and the result is a female-headed family.
To understand this set of relationships and to design programs and policies that encourage and facilitate childrearing within marriage, Mincy proposed a framework that focuses on "fragile families," consisting of children born outside of marriage to two disadvantaged parents. His model uses these families as the unit of analysis and examines what can be done to engage both parents in promoting the child's well-being. Such a framework would allow researchers and policymakers to understand the relationship between unmarried parents and between a child and a father, regardless of where the father lives. It would allow for distinctions between couples formed as a result of abusive or predatory relationships and couples formed as a result of relationships that have some potential to be strengthened. In the latter case, intervention at the point of cohabitation or conception might strengthen a relationship and increase the likelihood of a viable marriage.
Many of the points summarized above conform with a conceptual framework developed by the National Center on Fathers and Families (NCOFF) and presented at the Town Meeting by Gadsden. This framework, consisting of seven assumptions or core learnings" about fatherhood, represents a convergence of research and practice. It is derived from the experiences of practitioners and points to several key areas of research. NCOFF's "core learnings" are:
- Fathers care, even if their caring is not always shown in conventional ways.
- Fathers' presence matters to the economic and social well-being of children and families.
- Unemployment is a major impediment to fathers' involvement, particularly for low income fathers from an impoverished background who may have difficulty engaging in their children's lives.
- Existing approaches to public benefits, child support enforcement, and paternity establishment often create obstacles to involvement, at least in the view of fathers who need to become involved with their children.
- A growing number of mothers and fathers need additional support to develop vital skills to share the responsibilities of parenting.
- The transition from biological father to committed parent has significant developmental implications for young fathers, related to role transitions and expectations.
- The behaviors of both young fathers and young mothers are significantly influenced by intergenerational beliefs and practices within families of origin.
There was considerable consensus among presenters on the shortcomings of existing research on fathers. As noted above, much of the research on children, families, and fertility has focused on female respondents, often relying on women to provide information about their male partners or the fathers of their children. Nevertheless, there is an increasing number of data sources that provide information on fathers as attempts have been made to include men when large household surveys are conducted and when administrative data are gathered. Among these large surveys are the Current Population Survey (CPS) and related Supplements on Child Support and Alimony conducted by the Bureau of the Census; the Panel Study of Income Dynamics (PSID), based on a probability sample of households first interviewed in 1968 and sponsored principally by the National Science Foundation since 1983; the National Longitudinal Survey of Youth (NLSY) which was initiated in 1979 by the Bureau of Labor Statistics; the Survey of Income and Program Participation (SIPP), an annual household survey conducted by the U.S. Bureau of the Census since 1984; the National Survey of Families and Households (NSFH), sponsored by the National Institute of Child Health and Human Development and conducted by the University of Wisconsin in 1987 with a follow-up in 1992; and the 1981 and 1988 National Health Interview Survey (NHIS) Child Health Supplements, designed and funded by the National Center for Health Statistics. In submitted written testimony, the Fatherhood Research Subcommittee of the Department of Health and Human Services indicated that the 1995 National Survey of Family Growth (NSFG) will have improved data on fathers including demographic characteristics, fertility behavior, child support, and child care. The NSFG is sponsored by the National Center for Health Statistics, and the 1995 data will be available January 1, 1997.
In addition, data on fathers gathered from birth certificates is potentially available for 4 million births each year. This information, collected by the National Center for Health Statistics, includes the father's age, race/ethnic origin, and educational attainment. In 1993, father's age was missing on 16 percent of all birth certificates and on more than half of nonmarital birth certificates. In fact, state laws preclude the collection of birth certificate data on fathers for nonmarital births unless there is a paternity acknowledgment. Recent legislative and regulatory changes by the Office of Child Support Enforcement to increase voluntary paternity affidavits at birth should improve data on non-marital births. However, federal budget cutbacks have curtailed collection of all educational attainment statistics of the father on birth certificate data collected since 1994 (Fatherhood Research Subcommittee, DHHS).
Although these data sources provide important information on fathers and male fertility, they still have significant limitations that future surveys should address. Presenters described some of these limitations, provided examples. and offered suggestions for improvements.
In virtually all existing surveys and collections of administrative data, certain sub- populations of men are undercounted. Administrative data are limited to fathers who have established legal paternity or who are divorced. Household surveys often miss low-income and minority males. Men in institutions -- primarily in prisons -- are rarely included in household surveys.
Sorensen highlighted sample design weaknesses in three surveys that include male respondents: the NLSY, SIPP, and NSFH. These surveys undercount men, especially black men, who are not resident fathers. She reported that even the decennial census estimates that about 14 percent of black males ages 30 to 34 were undercounted in 1990. Moreover, with the exception of the NLSY, these surveys do not identify and interview institutionalized males, yet NLSY data indicate that 40 percent of these young men ages 26 to 32 are noncustodial fathers. Sorensen emphasized the need to design new strategies to identify and interview underrepresented noncustodial fathers in households and in institutions.
Cave pointed out that surveying nonresident fathers directly may not completely remedy the problem of underrepresentation. In fact, many men fail to acknowledge fatherhood, and measures restricted to those men who voluntarily acknowledge fatherhood are subject to self-selection bias. Cave also asserted that high mobility for poor nonresident fathers leads to survey nonresponse biases, especially when the sampling frame is households rather than individuals. Mincy added that unemployment, poverty, and attendant factors affect some men's ability to form families. Consequently, when children are born outside of marriage, the mother usually establishes a household with her child and the father lives elsewhere. These nonresident fathers may reside with their parents, or with other relatives or friends, and may move among them. As a result, they are often missed in surveys.
Findings from the NSFH and other surveys indicate that men are less satisfactory respondents than women. They have lower response rates in all sample surveys, and the quality of information provided by men, especially on issues related to children's behavior or development, is inferior to that provided by women. For instance, Sweet reported that when mothers and fathers are asked factual information about their children, there are significant differences in the reports, with fathers more likely to report that they don't know the answer. Overall, mothers' responses are considered to be more accurate.
Michael, in subsequent written comments, acknowledged that the response rate in sample surveys is indeed lower for men, but did not consider men to be less adequate reporters. He noted that many surveys place considerable emphasis on information about which fathers have traditionally been better informed, asking detailed questions, for example, about family income by source, value of assets such as savings and retirements, insurance provisions, and details of expenditures. His general assertion was that there are some matters of fact on which each parent is probably the better informant, and there is not consensus about what factual items are best reported by each. More importantly, there are many other items of high priority in surveys that are judgmental, attitudinal, or expectational, and on these items it is important to have both parents' views. Too often, only the mothers' views are asked and analyzed.
According to Sonenstein, national fertility studies have excluded males based on two rationales. First, men are often thought to be more difficult respondents than females to interview and there are doubts about the reliability of the information they provide. Second, when resources limit data collection to one respondent, reports from mothers and potential mothers are often considered sufficient. Sonenstein noted, however, that findings from the National Survey of Adolescent Males (NSAM) challenge these assumptions. The NSAM's response rate was well within the acceptable range. The Institute for Survey Research obtained interviews with 73.9 percent of eligible males ages 15 to 19 living in households, and 89 percent of the original sample consented to a second interview two and a half years later. In addition, Sonenstein cited evidence of high response reliability for the adolescent males, as well as findings that substantiate the importance of ascertaining and incorporating the male perspective on fertility issues.
Sorensen discussed the sequence and wording of survey items in the context of underreporting of fatherhood by male survey respondents. She noted that when men are interviewed they do not necessarily identify themselves as nonresident fathers, and they are consequently underreported in self-reported data. However, the extent of underreporting appears to depend on both the sequence and wording of the questions. With respect to sequence, Sorensen recommended the approach used by the NLSY to minimize underreporting. This survey asks men first about their fertility and then about where each of their children lives. She also urged that surveys of men adopt questions similar to those posed in the fertility supplement of the SIPP. This supplement asks women an extensive series of questions including how many children have been born to the woman, the age of each child, and where each child resides. Sorensen recommended that surveys exploring male fertility and fathering first ask this set of questions to help identify absent fathers and then proceed to items regarding financial support of and contact with each child. She and others suggested that survey items for males include their fertility and marital history, income, household status, current financial support of their children, and whether those children are receiving public assistance.
Mincy recommended longitudinal designs that follow male fertility and fathering with the same vigor as those that follow female fertility and mothering, and suggested sampling frames consisting of individuals rather than families or households to facilitate tracking males over time. He also stressed the need to develop more aggressive methods to counteract the problem of undercounting, and the need to calculate and report poverty rates for men based on their own income, regardless of the income of others in the household in which they reside.
Michael suggested the use of phone surveys or return mail surveys in longitudinal designs for appending modules that ask questions about interactions between and among the father, mother, and child. Manser noted that one benefit of longitudinal surveys is that they initially involve respondents when they are young and then follow them for several years. For instance, the new cohort of the NLSY will include 12- to 17-year- old males and females and track respondents into adulthood, thereby increasing the likelihood of finding and interviewing males who become absent fathers. Sweet recommended more research on the way particular constructs are measured, including the assessment of existing measures and the development of new ones. He suggested the use of focus groups, cognitive interviewing and survey measurement experiments to determine the most appropriate wording and sequence of items and thereby facilitate survey construction.
Braver referred to findings from his extensive research on divorced couples in Arizona and noted the potential merits of developing a nationally representative database of interviews with matched mothers, fathers, and their children where feasible. Sweet emphasized the importance of conducting a large-scale comprehensive family survey like the NSFH but with improvements in the survey design and content of the original NSFH. He noted the need for a new cross-sectional survey before the end of this decade and recommended the selection of a new cross-sectional sample over reinterviews with the full NSFH sample. In his view, it is essential to have a large sample size with oversampling of strategic subpopulations, including one-parent and step-parent families, low-income families, and minority families, especially Southeast Asians, Native Americans, and families with disabled or chronically ill parents or children. Sweet also stressed the importance of collecting identical information from both men and women, since few aspects of family life are uniquely male or uniquely female.
Levine suggested that researchers literally "turn-the-tables" in the display of data to reveal the missing paternal dimensions in our understanding of child care. He pointed out that data on the primary care arrangements of children are currently collected and displayed in relation to the mother's employment status. Using "working mothers" as the reference point obscures data about the father's employment status that may be critical to understanding family patterns of child care need, preference, and use. The collection and display of data on the primary care arrangements of "working fathers" would yield a new way of understanding the father's role. How, for example, do child care arrangements vary with a father's employment or unemployment? How do they vary with his status as custodial or noncustodial parent, or with his status as a husband or an unmarried domestic partner?
Gadsden suggested the need for complementary sources of information, such as field studies, to inform the development of national and longitudinal surveys. Programs that involve fathers present the opportunity to collect important information. However, the experimental and control group evaluation design is difficult to implement because fathers' programs are often small and relatively new. Program evaluation efforts are further complicated by problems associated with securing voluntary father follow-up information from program participants (Fatherhood Research Subcommittee, DHHS).
As Gadsden noted, programs that involve fathers require effective evaluation and monitoring mechanisms to enhance the usefulness of the information and facilitate its dissemination to other data collection efforts. The integration of multiple methodologies is critical to an examination of fathering and male fertility issues. Sweet recommended systematic efforts to link qualitative research findings with survey research efforts so that the two approaches to data collection can strengthen each other.
Several presenters provided specific recommendations regarding the content of future surveys. Recommendations addressed issues related to low-income fathers' financial support of their children, father involvement after divorce, parenting behaviors of young fathers, and improved understanding of a father's role.
Sorensen indicated that surveys fail to ask males whether their children are receiving public assistance; consequently, there is little evidence to accept or reject the claim that fathers of children on AFDC can afford to pay meaningful levels of child support. In fact, there is very little information in general about these fathers. Primus urged that data be collected that would help answer whether children are better off when paternity is established, and if so, how. In addition, he recommended items aimed at investigating whether government policies are neutral with respect to the living arrangements of children. As prerequisites to this inquiry, Mincy suggested gathering information on state variations in policies related to housing assistance, paternity establishment, minimum child support orders, and other policies that affect household composition. Offner recommended that federal statistical agencies provide more comprehensive state-level data to inform state legislative and programmatic efforts to involve fathers,
Cave based some of his recommendations for survey content on findings from the Parents Fair Share Demonstration, a project to improve the employment status and child support compliance of noncustodial parents of children on AFDC. He recommended that data on absent fathers include measures of formal child support as well as informal support, such as clothes, food, toys, books, diapers, medical and dental care, child care and monetary contributions directly from the father to the mother. Cave also recommended supplementing measures of formal earnings with measures of informal earnings, such as unrecorded transactions that are common among employers of low-wage workers. To capture more complete earnings data, he suggested asking fathers first about consumption and then about income, and then reconciling information from these two areas of inquiry. In addition, Cave noted the need to apply hardship measures -- items that gauge an individual's ability or inability to pay for food, shelter, and medical care -- to absent fathers as well as to custodial parents and children. This information would help to distinguish between those fathers who choose not to pay child support and those who are unable to pay. Since child support enforcement may cause conflict between parents and may have unintended consequences for the custodial mother, Cave also recommended measures of violence between parents. Finally, he noted that surveys that ask mothers about absent fathers should include items on fathers' basic demographic characteristics, such as age and education, his degree of contact and involvement with the child, and whether the mother limits or completely prevents the father's involvement with his children.
Koon, referring to findings from Missouri's JOBS program and the state's earlier participation in the Parents Fair Share Demonstration, pointed to gaps in knowledge about noncustodial fathers of children on AFDC. He urged that future surveys ask how many children noncustodial fathers have in multiple households. Koon also recommended questions similar to those posed by Cave regarding contact of noncustodial fathers with their children and families and informal support, especially unrecorded monetary support that may supplement low AFDC benefits. He suggested items related to fathers' employment patterns, including types of jobs, wages, length of time spent on specific jobs, reasons for job changes, total time employed during the year, and their participation in the underground economy. This information would provide important guidance to policymakers and program administrators designing employment and job training programs. For example, he speculated that fathers may fail to participate in voluntary job training or job search assistance programs because they were already employed in the underground economy, but he has no data to answer this question and others definitively. Additional items of interest to state administrators would explore how these fathers live and what types of social services, if any, they use. Koon also suggested inquiry into the factors that contribute to family breakup or father absence, whether existing policies encourage fathers to leave home so that mothers can qualify for state benefits (as some participants in JOBS and the Parents Fair Share Demonstration have indicated), and specifically whether the current welfare system contributes to family breakups or to the failure of families to form in the first place. Finally, Koon recommended that the Census continue to administer the migration item, which is currently in danger of elimination. He noted that migration patterns of noncustodial fathers across state lines are important to follow.
Several presenters noted the importance of including additional survey items pertaining to child support and custody arrangements. For instance, Campbell pointed out that women are asked why they do not receive support, but fathers are not asked why they fail to pay. Previously mentioned items targeting the different forms of formal and informal support may reveal additional information about child support compliance. Campbell also recommended greater investigation of the relationship between joint custody or visitation and the payment of child support. Does the noncustodial father's involvement with his children prior to the award of child support increase the likelihood that he will continue to pay support? Or do joint custody arrangements lead to involvement and thereby increase the likelihood of child support compliance? And how does the presence of a stepfather in the child's home affect the noncustodial father's child support compliance? Does it lead fathers to feel that they are no longer obligated to pay support?
In written testimony, Flohr, a member of a state commission examining divorce laws, noted the need for more information on the effects of various child custody arrangements on child well-being. She was also interested in the number of fathers wishing to assume primary custody for their children, and the number who are awarded such custody.
Braver recommended more questions about factors related to divorce in general and to fathers' reasons for lack of involvement with their children after divorce. He stressed that fathers and mothers often express strikingly different views, so that when fathers are asked direct questions, a new picture of existing circumstances emerges. In submitted written testimony, Flor indicated the need for more accurate data on family income. He pointed out that child support is nontaxable income for the recipient but taxable income for the provider, so there may be underreporting of the custodial parent's income and overreporting of the noncustodial parent's income, resulting in exaggerated differences in reports of their financial circumstances.
Gadsden highlighted the influence of intergenerational factors on the parenting behaviors of young fathers. She referred to questions about the quality and quantity of social support in families of origin, the intergenerational transfer of parenting beliefs and practices, the goals and expectations of families, and the difficulties presented by families. She recommended items on the father's childrearing practices and what it means to the father to be involved in the life of his child. What are the possibilities for these young fathers to make a contribution to their children's educational and emotional development? What do they think is necessary? Gadsden noted that many young fathers want to improve their children' s life prospects, but their own disadvantaged childhoods provide them with little insight into how to do so.
Several presenters emphasized the importance of asking men questions about their children, their interactions with their children, and the significance of their role as a father. Michael noted that the role of father may provide a man with self dignity, self identity, or a sense of accomplishment or pride. Gadsden mentioned that since men may conceptualize their roles broadly or narrowly, items pertaining to fathers should clearly delineate the roles of surrogate fathers, stepfathers, biological fathers and so forth. Similarly, Horn pointed out the lack of clarity associated with survey items that refer to father figures and suggested a closer examination of the concept of a father figure in children's lives. In sum, the data need to explain better how men perceive and define their roles in the lives of children.
McMillen cautioned the research community that, in the current fiscal and political climate, funding requests for new data collection efforts will be received with skepticism. He noted that budgetary constraints would cast any proposal to gather new data as a zero sum game and asked what existing data researchers and the federal statistical agencies are willing to give up in exchange for new data. He also pointed out that support was strong in Congress for the Family Privacy Act, which would severely hamper data collection by requiring written parental consent before minor children could participate in government-funded surveys.
McMillen advised the research community to frame proposals for new data in terms of those public policies that are most relevant to this Congress. He noted that many members of Congress trace the origins of many social problems to the breakdown of the traditional family. He urged researchers to recognize that framework in their discussions with members of Congress and to educate them about changes that have occurred in family life in recent decades.
Panel members Wallman, Riche, Maddox, and Primus strongly endorsed McMillen's comments and urged the research community to work with their agencies to place the questions to be answered by new data in the most compelling policy context, to help the agencies set priorities for data collection, and to provide political support when the agencies' funding requests are considered by Congress . Wallman noted the need to apply these new questions to existing data so as to identify specific gaps in knowledge that should be the focus of new data collection efforts. Riche added the need for clearer guidance on what level of data -- local, state, or national -- would be most helpful to the development of effective public policies. Griffith remarked that new data on fathers could inform educational policies -- particularly those related to early intervention to improve school readiness -- as well as policies related to welfare and family income. Primus, Maddox, and Raub all endorsed a focus on child well-being as the most compelling rationale for gathering additional data on fathers.
