Introduction

Douglas J. Besharov

Welfare provides economic assistance to single (and sometimes married) parents who need financial help to provide adequate care for children. But welfare, especially under the 1996 welfare law, is meant to provide temporary assistance only. Each additional child a low-income parent has makes the goal of self-sufficiency that much more difficult to attain. As the Alan Guttmacher Institute (1994) notes, "Closely spaced births early in a woman's life . . . increase her welfare dependency" (p. 60). Thus, a major focus of contemporary welfare policy is to discourage mothers from having additional children while on welfare.

How common is it for is it for mothers on welfare to have additional children? Using data from the 1995 National Survey of Family Growth, researchers at the National Center for Health Statistics (NCHS) estimate that about 10 percent of women who received Aid to Families with Dependent Children (AFDC) in the 12 months prior to the survey had a second- or higher-order birth (about 508,000 out of nearly 5 million; Matthew Bramlett, NCHS, personal communication, July 31, 2000). About half of the mothers told the survey that they "intended" to become pregnant.

That figure may exaggerate the number of subsequent births to welfare mothers because it is based on welfare receipt any time during the 12 months prior to the survey. For example, it includes some mothers who received welfare but left the rolls prior to the birth of the child. A recent analysis of California data by Michael Wiseman, a visiting fellow at the National Opinion Research Center, also supports the idea that the number of subsequent births to welfare mothers has been exaggerated. According to Wiseman, in 1995 there were about 76,000 second- or higher-order births to California mothers receiving welfare, representing about 6.7 percent of the state's welfare caseload (see "Counting the Capped, this volume). This figure also may be slightly exaggerated, because Wiseman counts all births in which a family member was receiving welfare nine months earlier, data that could include first births to a teenager whose family is on welfare.

Teenage births, about 79 percent of which are nonmarital, are a particular concern (Martin, Smith, Mathews, and Ventura 1999). According to the U.S. Congressional Budget Office, between 1979 and 1985, 50 percent of unwed teen mothers were on welfare within one year, and 77 percent were on welfare within five years (Adams and Williams 1990, p. 52). Of course, any teen birth puts the mother and her baby at risk of becoming welfare dependent, but the numbers are substantially lower for married teens; only 25 percent of married teen mothers went on welfare within five years.

Many subsequent births come quickly after a teen mother's first out-of-wedlock birth-making it even more difficult for them to continue their schooling or begin building work experience and skills. According to data compiled by the NCHS, in 1998, about 22 percent of all births to teenagers were second or higher-order births (Martin et al. 1999).

Writing for the Alan Guttmacher Institute, Cynthia Dailard, a senior public policy associate at the institute, describes the social consequences of such repeat childbearing:

Although those figures reflect the serious nature of teen parenthood, it is extremely important to remember that not all unwed teen mothers develop long-term welfare dependency. Many stay in school (or return), get jobs, marry, and lead full, productive lives (Besharov and Sullivan 1996). Moreover, not all subsequent births to teenagers involve mothers who are on welfare.

Nevertheless, the negative effect of subsequent births to teenage mothers is an important indicator of the more general problem. The data from both NCHS and Wiseman involve subsequent births to both single and married mothers (whether or not living with the child's father). According to NCHS, about 26 percent of subsequent welfare births involve married mothers (133,000 births vs. 375,000 births for single mothers; Matthew Bramlett, NCHS, personal communication, July 31, 2000). Wiseman, because he is using welfare data, compares subsequent births to mothers receiving AFDC-FG (generally single mothers) to mothers receiving AFDC-UP (two-parent families in which one parent cannot work or is unemployed). In 1995, 28 percent of California's subsequent welfare births involved AFDC-UP families (21,000 births vs. 55,000 births for AFDC-FG mothers.)

The NCHS data also shed some light on the "intendedness" of these births. Nearly half (about 45 percent) of the single mothers told the NCHS survey that their baby was "intended." An even greater number of married mothers on welfare (about 76 percent) said that their babies were intended. (Matthew Bramlett, NCHS, personal communication, July 31, 2000) These figures suggest that when it comes to preventing subsequent births, attitudes and motivation are at least as important as access to contraceptives.

In this volume, Glenn Loury, a professor of economics and director of the Institute on Race and Social Division at Boston University, assesses the research evidence on the effectiveness of five projects that sought to reduce subsequent births to mothers on welfare. The first two involved family caps on welfare benefits: the Arkansas Welfare Waiver Demonstration Project and the New Jersey Family Development Program. The second two involved enhanced family-planning services as part of broader interventions for teen parents: the New Chance Demonstration and the Teenage Parent Demonstration. The last one involved "authoritative" or "directive" counseling within a nurse home-visitor program: the Prenatal and Early Childhood Nurse Home-Visitation Program. Loury also examines the Dollar-a-Day program, which involved positive financial incentives for young mothers who did not become pregnant.

Broadly, Loury's conclusions are as follows:

• The existing research on family caps on welfare benefits has too many limitations to draw any conclusions, one way or the other, about their impact.
• Enhanced family-planning services, at least within job training and welfare-to-work settings, do not seem to reduce subsequent pregnancies.
• Authoritative or directive counseling approaches, at least within a nurse home-visitor program closely patterned after the original Elmira project, seem to reduce subsequent pregnancies and should be tried under other carefully evaluated circumstances.

In addition to Loury's paper, this volume contains a detailed assessment of the New Jersey study by Peter Rossi, professor emeritus at the University of Massachusetts (Amherst); it also offers three comments from people closely associated with that project and other analysts: Michael Camasso at the School of Social Work and Center for Urban Policy Research, Rutgers University (and his colleagues); Michael Laracy, a senior program associate at the Annie E. Casey Foundation; and David Murray, director of research at the Statistical Assessment Service. Some of their comments support Loury's and Rossi's assessments, and some do not. This volume also includes an analysis of the New Jersey study from the Congressional Research Service. All of these materials are appended to this report so that readers can make their own judgements.

Finally, this volume contains a paper by Michael Wiseman that presents the California estimates described above. Wiseman uses a longitudinal database to estimate the number of subsequent births to California welfare mothers in 1995, including an estimate of those who would have been subject to the family cap had that policy been in effect at that time. He finds that a significant number of children would have been subject to the family cap had it been in effect in 1995, and he provides estimates for the rate at which the number of children affected grows over time. His estimates highlight the importance of the issue.

Under traditional welfare policies, the size of the family's cash grant usually depends on the number of children. If the mother has another child while on welfare, the grant is increased accordingly.¹ Prior to the passage of the 1996 welfare reform law, a number of states were granted waivers that allowed a "family cap," a policy that eliminated or reduced the additional benefit for children conceived while the mother was receiving welfare. The 1996 law allows states to adopt family caps without Washington's approval. As of this writing, about 21 states have adopted a family cap.

Many supporters of family caps believe that the traditional rule provides a financial incentive to have children while on welfare, or at least does not provide a disincentive to having more children. They believe that family caps could send a signal to mothers that they should not have additional children until they can support them.

Many opponents of family caps fear that they will increase the material hardship faced by families on welfare. Others fear that family caps will increase the number of abortions, or they consider them a violation of a mother's right to have children. In 1995, a group of women concerned about the impending welfare reform legislation expressed the latter view as follows:

The possible financial impact of a cap on welfare families is no small concern, as Wiseman's research documents. He finds that in 1995, about 82 percent of the subsequent births to California mothers on welfare-about 60,000 welfare mothers, or about 5.4 percent of the state's welfare caseload²-involved conceptions while the mother was receiving welfare and thus potentially were subject to the family cap.³ In any one year, those percentages may seem small, but as Wiseman explains, their impact grows over time:

Two states have released evaluations that attempt to determine the impact of their family caps by means of experimental designs, considered by most social scientists the "gold standard" of evaluation. In both Arkansas and New Jersey, welfare mothers were randomly assigned to a treatment group (subject to the family cap) and a control group (not subject to the family cap) (Camasso, Harvey, & Jagannathan 1996; Turturro, Benda, & Turney 1997). If done properly, random assignment would ensure that members of the treatment group and the control group were comparable, so that any difference in subsequent outcomes could be attributed to program participation.

In both states, there was no significant difference in birth rates between the two groups within the time frame examined (two years in New Jersey and three years in Arkansas). Ordinarily, this finding would suggest that family caps have no impact on subsequent births. However, Loury notes, "the implementation problems in these experiments were so severe that one may question whether family-cap programs actually have been fairly tested." In particular, many members of both the experimental and control groups did not have an accurate understanding of the policy that applied to them. Loury also points out that the surveys on which the evaluations were based suffered from low response rates and that the respondents were not representative of the larger AFDC caseload. Thus, he concludes that "the problems of implementation and sample attrition in both sites were so severe as to preclude drawing strong conclusions."

Because of these problems, the team evaluating the New Jersey family cap performed a subsequent analysis using administrative data to compare birth outcomes for the experimental and control groups and also to AFDC recipients in the 22-month period before the intervention went into effect to the outcomes in the subsequent 38-month period (Camasso, Harvey, Jagannathan, and Killingsworth, 1998a, 1998b). The New Jersey researchers concluded that the family cap led to a decline in births and a rise in abortions. Even with this additional analysis, however, Rossi concludes that the [New Jersey family cap] "may have had the effects the Rutgers research group claim, or it may not have had those effects. We simply do not know from this research, because the deficiencies noted above are serious enough to cast strong doubts on the validity of the findings." Camasso and his colleagues defend their approach and contend that the two different approaches they used led to the same results. "To accept Rossi's conjectures," they say, "one must be willing to believe that this confluence of findings is a mere artifact of design flaws and differences in estimation methods."

Thus, these two evaluations-the only state-specific studies of family caps-are too flawed to be used as the basis for any sort of judgment about the effectiveness of family caps. They demonstrate that findings from even randomized experiments can be severely undermined by poor implementation and inadequate data (Rossi, Freeman, and Lipsey 1999). 

Loury also reviews a 1999 study that used national data to assess the impact of various policy changes (contained in welfare reform waivers) that might affect the fertility of unmarried women, such as the family cap, minor parent provision, time limit, work requirement, AFDC-UP, child support, expanded income disregard and asset limit, and school attendance and performance requirement. The study found that some waivers increased nonmarital childbearing, whereas others reduced it, and the researchers concluded that the "family cap waiver is shown to be a useful policy tool in lowering non-marital fertility." (Horvath-Rose and Peters 2000, p. 27)

Loury cautions, however, that the nonexperimental nature of the study and the difficulty in defining meaningful welfare reform variables raises considerable uncertainty about the findings. He concludes that the findings may be more suggestive of the impact of welfare reform, writ large, than of any specific waiver.

Loury also reviews the findings from an evaluation of the Dollar-a-Day program, which was targeted to teen mothers with an infant younger than five months (Stevens-Simon, Dolgan, Kelly, and Singer 1997). Its objective was to reduce the subsequent childbearing of the young mothers by providing a financial incentive ($1 for each nonpregnant day) and a supportive, peer-group environment. The teen mothers were randomly assigned to one of four groups: a monetary incentive and peer-support group; a monetary incentive-only group; a peer support-only group; or a control group receiving neither. After 24 months, 39 percent of the adolescent mothers had experienced a second pregnancy, with no statistically significant differences across the four test groups.

For those who think that lack of information about family planning and access to contraceptives is the cause of the problem, the next two projects Loury reviews by provide disappointing results.

• The Teenage Parent Demonstration, evaluated by Mathematica Policy Research (MPR), was a randomized experiment that required teen parents in three cities (Chicago, Illinois, and Newark and Camden, New Jersey) to participate in an educational or training activity and then seek employment. Case management and a rich array of services, such as child care, transportation assistance, and counseling, were provided. Although the main objectives of the demonstration were to promote school attendance and employment, it also sought to reduce high rates of teenage pregnancies and births (Kisker, Rangarajan, and Boller 1998).
• The New Chance Demonstration project, evaluated by the Manpower Demonstration Research Corporation (MDRC), also used a random-assignment research design to test the impact of a comprehensive, multiyear program in 16 sites nationwide that provided education, training, parenting, child care, and other services to young mothers who had children as teenagers and who also were high school dropouts. The program had multiple objectives, including helping participants increase their educational attainment, build labor market skills, postpone additional childbearing, and improve the well-being of their children (Quint, Bos, and Polit 1997).

Although both projects were welfare-oriented interventions, they provided the opportunity to test the impact of additional family-planning education and services on young, unwed mothers. Recognizing the obstacles to self-sufficiency posed by the birth of additional children, both projects provided extensive family-planning services. For example, in Project New Chance, the mothers received, on average, about six hours of family-planning services. In some sites, they received considerably more.

Both projects, unfortunately, showed disappointing results with regard to subsequent births-even in sites where the mothers received many hours of family-planning services. In Project New Chance, the proportion of teenagers who had a repeat birth within 42 months was nearly identical for the experimental and control groups (54.7 percent versus 55.3 percent).⁴ In the Teenage Parent Demonstrations, about three-fourths of both the experimental and control groups experienced a subsequent birth within six to seven years after entering the program.⁵

In fact, Rebecca Maynard, at the time the trustee professor of education, policy, and communications at the University of Pennsylvania, reported that in the Teenage Parent Demonstrations, more than one-fourth of the teenagers on welfare were pregnant again within one year, and fully half were pregnant within two years. Three-fourths of the pregnancies resulted in births (Maynard 1994). Loury concludes that "these outcomes imply that the information about contraception and birth control provided by the workshops was not sufficient to reduce fertility."

These and similar disappointing results have fostered a growing feeling that "nothing works" to prevent subsequent births. Worse, there is a tendency to blame the young mothers for the programs' failures, on the grounds that their social situation or personal dysfunction prevents programs from helping them. For example, writing about the disappointing lack of impact from the nine-year, $326 million, Comprehensive Child Development Program, columnist Robert Samuelson wrote that

But it is equally possible that the programs took the wrong approach to intervention. After all, we do not blame the patient for not recovering when the physician prescribes the wrong medicine.

Therein lies the significance of the final two studies Loury assesses, those on "nurse home visitation." David Olds, originally from the University of Rochester and now at the University of Denver, has conducted scientifically rigorous, random-assignment experiments in Elmira, New York; Memphis, Tennessee; and Denver, Colorado. The Prenatal and Early Childhood Nurse Home Visitation Program is a highly structured, voluntary program that uses an authoritative or directive approach to counseling low-income, first-time mothers. Nurse home visits begin during pregnancy and continue for two years after the child is born. The nurse home visitors provide a comprehensive set of services that focus on a mother's personal health and development (including services to prevent unintended pregnancy and find employment) and the quality of caregiving for the child. As opposed to interventions that are nonjudgmental, the nurse home-visitation projects deliver clear behavioral messages by public health nurses. For example, as The Washington Post reported, the traditional way of providing family planning would be to say, "If you want to avoid a second baby, here's a condom and how to use it." The authoritative approach is to say, "You shouldn't have another baby and here are ways to prevent it." (Vobejda 1998, p. A1).

A randomized experiment in Elmira, New York, found that the nurse home visitors achieved a 25 percent drop in smoking by the end of the pregnancy, a 75 percent reduction in premature births among pregnant women who had smoked, and large birthweight increases for babies born to young teen mothers (nearly 400 grams for mothers ages 14 to 16). In addition, 15-year follow-up findings indicate almost a 31 percent reduction in subsequent childbearing for low-income, unmarried mothers (1.1 vs. 1.6 subsequent births). Verified cases of child abuse and neglect were 79 percent lower, drug and alcohol problems 44 percent lower, arrests among the mothers 69 percent lower, and welfare use 33 percent lower compared with the control group. As a result, the program led to large savings in government spending. (Replications of this study are showing similarly impressive results in Memphis, Tennessee,⁶ and Denver, Colorado.)

Why the reduction in subsequent births in the nurse visitation projects but not with traditional family-planning services? Perhaps because, in addition to providing information about birth control and access to contraceptives, it is also necessary to change the attitudes and behavior of young mothers. Nicholas Zill, a vice president at Westat, Inc., for example, reports that women on welfare desire larger families than women not on welfare. His analysis of the 1988 National Survey of Family Growth found that mothers on welfare said that their ideal number of children was 3.0, compared with 2.7 for all other mothers (including poor mothers not on welfare) and 2.5 for nonpoor mothers (Zill 1998). According to Loury, "The success of home visitation seems to be due to the fact that the nurses got the message across that becoming pregnant again is not desirable." Among the factors he cites are the more intensive information about family planning provided; a greater number of unambiguous, normative message that becoming pregnant again is not desirable; and the greater likelihood that clients will "hear and respond positively to authority of the nurse." Echoing Loury, Maynard wrote that

Thus, programs like Project New Chance and the Teenage Parent Demonstration-and most other family-planning programs, for that matter-may take the wrong approach. They tend to be nonjudgmental and value neutral.⁷ "Moreover," in the words of an MDRC report, "program staff are not always comfortable or skilled in dealing with family-planning issues" (Quint et al. 1997, p. 164). According to officials from MDRC, the message New Chance mothers received was, "Think about having another child."

The home-visitation program, however, provides clear behavioral messages delivered by public health nurses: "Try not to get pregnant again. Having fewer children is in your best interest and that of your family." As Olds describes, the nurses seek to "promote change by providing a vision of the future." Adds Lorraine Klerman, professor and chair of the Department of Maternal and Child Health, School of Pubic Health at the University of Alabama at Birmingham, "The home-visiting program's approach to its program objectives was through behavior modification, using various techniques, including role modeling, to change inappropriate behaviors" (Klerman 1998, p. 3).

As Loury explains, "The home-visitation programs reviewed here differ from the financial-incentive and case-management efforts in three areas-the service provider, the population served, and the type of family-planning service offered." Loury points out that the service provider for financial-incentive and case-management programs was a caseworker, whereas with home-visitation programs, the service provider was "a nurse explicitly trained in details of contraceptive practice, in techniques to help clients establish and achieve realistic goals, and in ways of enlisting the support of family members, friends, and the mothers' partners." He also notes that the target populations were different. The home-visitation programs included only women without a previous birth, whereas the other programs generally targeted women who had previously given birth. In addition, the nurse home-visitation programs involved many differences in service delivery, with frequent contacts in a participant's own home. Moreover, family planning received more emphasis in those programs. He adds that the home-visitation programs "provided a greater number of unambiguous, normative messages that becoming pregnant again is not desirable."

Home visitation is certainly not a silver bullet that will eradicate subsequent births. In the Olds research, the rate was reduced by 31 percent, but most of the mothers still had another child. Moreover, many questions about the research findings remain unresolved. For example, the high quality of the program (compared with what might be expected in day-to-day practice) limits the generalizability of the results. (Olds has been critical of attempts to replicate his program, especially those that have used paraprofessionals instead of trained nurses.) It also would be reassuring to have an independent evaluation of the program.

Loury's assessment of the research does not suggest that programs seeking to discourage subsequent births should cease providing birth-control information. Information and access to contraceptives are necessary underpinnings of any effort. But the larger point that emerges from Loury's assessment is that programs with a clear, authoritative message against subsequent births-rather than neutral messages-increase their effectiveness.

Loury notes that "evidence is starting to demonstrate that such 'authoritative governance' may in reality be more important than the incentives with which economists are typically concerned." That notion is supported by his conclusions concerning the programs he examines in this volume. Indeed, Loury finds that "nurse home visits appear to have been more successful in reducing subsequent pregnancies" than financial incentives and case-management programs. His paper summarizes his ideas and suggests various policy implications that arise from his conclusions.

The handful of studies described in this volume should not be overinterpreted. The research of the financial incentives contained in family-cap provisions give no real indication of their effect, if any. Nevertheless, the impact of a family cap undoubtedly depends on both the intractability of the behavior and the size of the sanction. The caps in New Jersey and Arkansas were about 15 to 25 percent of the welfare grant and constituted less than 10 percent of the family's total cash and noncash benefits. Depending on the circumstances and what is counted, the difference was as little as 5 percent.

Because of the weaknesses of the evaluations, we do not know whether sanctions of this size are sufficient to change behavior, and it seems plausible that they are too small to counter the other forces in the women's lives. It also seems equally plausible that a much greater penalty would affect the childbearing behaviors of at least some welfare mothers. That approach would raise the issue of whether the possible decline in subsequent births would be beneficial enough to justify the extreme hardship that such a severe sanction would probably cause.

A broader implication of family caps, apparently not examined in current research, is the degree to which it might mobilize public and private agencies to work with unwed mothers to have them at least delay having additional children. Up to now, agencies have not done so to any appreciable degree, as Richard Nathan and his colleagues at the Nelson A. Rockefeller Institute of Government at the State University of New York at Albany, point out:

As for traditional family-planning services, the lesson is not that they have no impact. Rather, giving young mothers more education about contraception and easier access to contraceptives may simply have no additional effect. That is, once young mothers have basic knowledge about effective birth control practice, motivation and other intangible factors may be what is important.

This broader view of what it takes to reduce subsequent births is the tentative lesson from the Elmira and Memphis nurse home-visitor program, which seems to have modified the attitudes of its clients. But the experiments were conducted in a somewhat rarified atmosphere under the same principal investigator. Before more certainty can be attached to the findings, the program should be faithfully and independently replicated and evaluated in a number of different settings. Fidelity to the original model is important because a number of projects only loosely patterned after the Elmira program have had less positive results (Center for the Future of Children, 1999); independence is important because the scientific method requires that others be able to duplicate the original researcher's results. Finally, the Elmira success appears to have been concentrated among the most disadvantaged clients. It remains to be seen whether the program can be operationally targeted to such groups.

Adams, G.C. and Williams, R.C. 1990. Sources of Support for Adolescent Mothers. Washington, DC: Congressional Budget Office.

Alan Guttmacher Institute. 1994. Sex and America's Teenagers. New York: Alan Guttmacher Institute.

Besharov, D., and Sullivan, T. 1996. Welfare reform and marriage. The Public Interest 125:81-94.

Camasso, M.J.; Harvey, C.; and Jagannathan, R. 1996. An Interim Report on the Impact of New Jersey's Family Development Program. New Brunswick, NJ: Rutgers University.

Camasso, M.J.; Harvey, C.; Jagannathan, R.; and Killingsworth, M. 1998a. A Final Report on the Impact of New Jersey's Family Development Program: Experimental-Control Group Analysis New Brunswick, NJ: Rutgers University.

Camasso, M.J.; Harvey, C.; Jagannathan, R.; and Killingsworth, M. 1998b. A Final Report on the Impact of New Jersey's Family Development Program. Results from a Pre-Post Analysis of AFDC Case Heads from 1990 to 1996. New Brunswick, NJ: Rutgers University, 1998.

Dailard, C. 2000, June. Reviving interest in policies and programs to help teens prevent repeat births. The Guttmacher Report on Public Policy. p. 1.

Center for the Future of Children. 1999, Spring/Summer. Home Visiting: Recent Program Evaluations/The Future of Children (9):1.

Horvath-Rose, A., and Peters, E. 2000. Welfare Waivers and Non-Marital Childbearing. Joint Center for Poverty Research Working Paper No. 128.

Kisker, E.E.; Rangarajan, A.; and Boller, K. 1998. Moving Into Adulthood: Were the Impacts of Mandatory Programs for Welfare-Dependent Teenage Parents Sustained after the Programs Ended? Princeton, NJ: Mathematica Policy Research.

Kitzman, H.; Olds, D.L.; Sidora, K.; Henderson, C.R.; Hanks, C.; Cole, R.; Luckey, D.W.; Bondy, J.; Cole, K.; and Glazner, J. 2000. Enduring effects of nurse home visitation on maternal life course: A 3-year follow-up of a randomized trial. Journal of the American Medical Association 283(15): 1983-1989.

Klerman, L. 1998, March. Can intervention programs prevent subsequent births to teenage mothers? Paper presented at Preventing Second Births to Teenage Mothers: Demonstration Findings. College Park, MD: University of Maryland Welfare Reform Academy.

Loury, G. C. 1998, June 29. Uneconomical. The New Republic. p. 14.

Martin, J.A.; Smith, B.L.; Mathews, T.J.; and Ventura, S.J. 1999. Births and Deaths: Preliminary Data for 1998. Hyattsville, MD: National Center for Health Statistics.

Maynard, R. 1994, April. Life prospects of teenage parents. Paper presented at The Causes and Costs of Teen Motherhood. Washington, DC: American Enterprise Institute.

Nathan, R. 1998, June. Remarks. Presentation at Pregnancy Prevention and Welfare Reform. College Park, MD: University of Maryland Welfare Reform Academy.

Nathan, R.P.; Gentry, P.; and Lawrence, C. 1999, April 2. Is there a link between welfare reform and teen pregnancy? Rockefeller Reports.

NOW Legal Defense and Education Fund. 1995, April 19. Open letter from the Committee of One Hundred [On-line]. Available: http:/csf/colorado.edu/lists/matfem/95/0387.html.

Quint, J.C.; Bos, J.M.; and Polit, D.F. 1997. New Chance: Final Report on a Comprehensive Program for Young Mothers in Poverty and Their Children. New York: Manpower Demonstration Research Corporation.

Rossi, P.H.; Freeman, H.E.; and Lipsey, M.W. 1999, Evaluation: A Systematic Approach Thousand Oaks, CA: Sage Publications.

Samuelson, R.J. 1998, February 18. 'Investing' in our children. The Washington Post. p. A21.

Stevens-Simon, C.; Dolgan, J.; Kelly, L.; and Singer, D. 1997. The Effect of monetary incentives and peer support groups on repeat adolescent pregnancies. Journal of the American Medical Association 277: 977-982.

Turturro, C.; Benda, B.; and Turney, H. Arkansas Welfare Waiver Demonstration Project: Final Report. Little Rock: University of Arkansas, 1997.

U.S. Congressional Budget Office. 1990. Sources of Support for Adolescent Mothers Washington, DC: Author.

Vobejda, B. 1998, May 13. Strictly speaking, success. The Washington Post. p. A1.

Zill, N. 1998. Family planning services among welfare recipients. In Besharov, D.J.; Stewart, F.H.; Gardiner, K.N.; and Parker, M.L., eds. Family Planning Services for Special Populations. Menlo Park, CA: Kaiser Family Foundation: 3-16.