Skip to main content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Building Self-Sufficiency Among Welfare-Dependent Teenage Parents: Lessons from the Teenage Parent Demonstration

Publication Date

By Rebecca Maynard

The Teenage Parent Demonstration (TPD) was a major, large-scale, federal demonstration initiative sponsored by the U.S. Department of Health and Human Services in response to three concerns:В  (1) rising welfare caseloads;В  (2) persistently high rates of teenage pregnancies and births; andВ  (3) the high probability that teenage parents will go onto welfare and have especially long spells of dependency.


Background and Key Elements

The demonstration began in 1986, before the passage of the Family Support Act and the implementation of the Job Opportunities and Basic Skills Training (JOBS) program.  Nonetheless, the demonstration programs paralleled those envisioned by the architects of the current JOBS provisions and requirements for custodial teenage parents.  Pregnant and parenting teens in the demonstration sites who had one child and were receiving Aid to Families with Dependent Children (AFDC) were required to:

  • Stay in school, if they were attending school at the time of program enrollment
  • Return to high school or enter an adult education program, if they had already dropped out of school
  • Enroll in postsecondary education, if they had already completed high school or received their GED
  • Enroll in an appropriate skills training program as an alternative to or after completing schooling, or
  • Seek employment as alternative to or after completing education and/or training

The cornerstone of the program intervention was case management .  Case managers' responsibilities resembled those of current JOBS case managers in many states:  conducting individual assessments to determine supportive service needs; working with the young mothers to identify appropriate major activities -- school, job training, or employment; helping resolve impediments to participating in major activities or preparatory endeavors; and monitoring program participation.  The demonstration programs also provided a rich array of services to complement the case management -- workshops, child care assistance, transportation assistance, and counseling.

This report synthesizes the results of the first phase of a major, multipart evaluation of the Teenage Parent Demonstration.  This phase of the evaluation covered an average of 30 months of follow-up of the study sample and resulted in more than a dozen major reports covering a wide range of issues related to service needs, service use patterns, service delivery strategies, and program impacts.  In 1998, another report was published covering six to seven years of post-enrollment outcomes for both the young mothers and their children (See full report and shorter, readable synthesis and lessons learned report).

Plan of this Report

Here, we first describe the demonstration programs and profile the population of welfare-dependent teenage parents in the demonstration sites.  Then, we discuss early lessons about the program's effects on out-of-home activity rates and impacts on a wide range of economic and social outcomes.  Finally, we discuss program design and implementation lessons.

Teenage Parent Demonstration Design

From late 1987 through mid-1991, the states of Illinois and New Jersey operated demonstration programs under grants from the Office of Family Assistance of the U.S. Department of Health and Human Services.  The Illinois program, Project Advance, operated in the south side of Chicago and the New Jersey program, Teen Progress, operated in the cities of Newark and Camden.  Each of these areas is characterized by high rates of unemployment, poverty, and crime.  In the three demonstration sites, all teenage mothers who had only one child and began receiving AFDC for the first time for themselves and their child were required to attend an intake session.  During the demonstration period, almost 6,000 teenage mothers joined the welfare rolls in these sites, and nearly 90 percent attended intake and enrolled in the demonstration.  Consistent with the evaluation design, about half were selected at random to participate in the demonstration programs; the remainder became part of a control group receiving regular AFDC services.

The Program Intervention

Participation in the demonstration was mandatory -- underscoring both the obligation of the young mothers to take charge of their lives and work toward self-sufficiency and the responsibility of the program to help them overcome obstacles to fulfilling this goal.  Those selected to participate were required to develop and comply with approved plans for engaging in activities aimed at promoting their eventual self-sufficiency.

Case managers helped participants decide what education or training to pursue, found open slots in appropriate programs, coaxed and pressured the young mothers to stick to their plans, and counseled them when crises arose.  If the teenage parents persistently failed to participate in planned activities, case managers initiated sanctions until the teens were in compliance.  The sanction consisted of a reduction in the AFDC grant by the amount normally allocated to cover the needs of the mother, generally $160 in New Jersey and $166 in Chicago, that remained effective until the young mothers complied with participation requirements.

All three demonstration programs required participants to attend a set of initial workshops designed to enhance their personal skills, convey information that would help them cope with their new responsibilities, and prepare them for education, training, and employment activities.  Workshop topics included child support, family planning, health and nutrition life skills, family management, motivation, parenting, employment preparation, education preparation, and HIV and drug abuse prevention.

To help participants move toward self-sufficiency, the demonstration programs promoted participation in education, job training, and/or employment, relying heavily on existing community services.  However, they also developed some in-house services, using both their own staff and staff from other agencies.  All three programs offered child care and transportation assistance to address these barriers to program participation.

Resource Levels

For the steady-state operating period of the demonstrations, average program spending per participant was about $1,400 a year; during this time, participants received AFDC for an average of eight to nine months.  These direct program expenditures were supplemented by an average of about $800 per participant in community-provided services, counting alternative educational services, but not regular high school programs.  In all sites, the major share of resource costs (40 to 50 percent) was associated with case management and support services.  Job training was the next largest component, accounting for up to a third of the project-related resources at the Chicago site and 12 to 14 percent at the others.

Target Population and Study Sample

The target population for the demonstration consisted of all teenagers who, for the first time, were parents and receiving AFDC (either as the head of their own cases or as "minor" mothers) or, in Illinois only, had no children but were in the third trimester of a pregnancy and receiving AFDC.  Of the nearly 6,000 eligible individuals who were identified in the three demonstration sites, 5297 (89 percent) completed intake (Table ES-1).  About half were then selected at random to participate in the demonstration programs (receiving enhanced services); the others served as a control group (receiving regular services).

Camden Newark Chicago TOTAL
Sample Sizes
Eligibile Teenage Mothers 1,256 1,346 3,360 5,962
Completed Intake 1,281 1,190 2,889 5,297
Attemped Follow-Up Survey and Basic Skills Retest 1,218 1,190 2,151 4,559
Completed Follow-Up Survey 1,057 971 1,839 3,867
Completed Basic Skills Retest 836 743 1,484 3,063
Response Rates (%)
Completed Intake 98 88 86 89
Completed Follow-Up Survey 87 82 85 85
Completed Basic Skills Retest 69 62 69 67

NOTE:  Follow-up surveys and basic skills retests were attemped with all sample members who completed intake in Camden and Newark, but with only a randomly selected 75 percent of those who completed intake in Chicago.

Information Sources

Data for the evaluation came from site observations, interviews with program staff, program data, state records data, and interviews with the sample members.  Sample members completed a group-administered baseline survey and took a basic skills test at intake.  The program experiences of all young mothers in the enhanced-services group were monitored through automated case-tracking systems.

Information on outcome measures was obtained through follow-up surveys and retests with those in the enhanced- and regular-services groups (Table ES-1).  Follow-up surveys were completed with 85 percent of those in the follow-up survey sample, which included the full baseline sample in Camden and Newark and a representative 75 percent of the Chicago baseline sample.  Basic skills retests were completed with 67 percent of the follow-up sample.  Data on welfare and earnings were obtained from administrative records for the full study sample, including the 11 percent of eligible teenage mothers who failed to complete intake.  Child support data were obtained from administrative data only for those who completed intake.

A substantial amount of qualitative data complemented this quantitative information.  We conducted regular program monitoring visits throughout the demonstration.  We also systematically gathered information about the experience, characteristics, and problems of the young mothers in the sample and the efforts of those in the enhanced-services group to deal with program participation requirements.  Focus groups with the young mothers (88 sample members), in-depth semi-structured interviews (70 sample members), and case conferences with program staff (focusing on 46 enhanced-services group members) provided the opportunity to collect this information.

Characteristics of the Target Population

As a group, participants were young and had substantial educational deficits when they arrived at the programs (Table ES-2).  About 30 percent of the mothers had dropped out before completing high school, and most of those still in school were behind grade level.  More than half of the young mothers had some work experience before enrolling in the demonstration, but most of this employment had been short term and at low wages.

Camden  Newark  Chicago  TOTAL
Age (Percent)
--Age 17 or younger 46.5 29.6 21.6 29.1
--18 or older 53.4 70.4 78.5 70.9
Race/Ethnicity (Percent)
--Hispanic 37.5 25.4 5.1 17.0
--Black, non-Hispanic 56.3 71.8 85.0 75.5
--White, non-Hispanic 6.2 2.8 9.9 7.5
Separated, Widowed, or Divorced (Percent) 8.0 4.7 2.7 4.3
Average Number in Household 4.9 4.8 4.6 4.7
Average Age of Youngest Child (Months) 7.8 12.0 9.3 9.6
Living with Parent (Percent) 46.2 53.4 46.1 47.7
Welfare Household During Childhood (Percent) 69.0 69.2 66.0 61.9
Attending School at Intake (Percent) 46.5 36.8 45.3 43.7
Completed High School/GED (Percent) 21.2 26.1 40.0 32.7
Non-English-Speaking (Percent) 8.6 5.2 0.2 3.2
Basic Skills
--Reading grade equivalent 7.4 7.5 8.4 8.0
--Math grade equivalent 7.6 8.1 7.8 7.8
--Ever Held a Job Percent 49.4 55.4 50.9 51.6

Sample Size 1,218 1,190 2,889 5,297

SOURCE:  Program Intake Forms.

All of these young mothers faced significant barriers to self-sufficiency simply by virtue of their living arrangements.  Many had left their parents' homes and were receiving no support from the fathers of their children.  Almost all lived in poverty, often in dangerous neighborhoods.  Moreover, they had relatively few role models in their communities to guide them toward social and economic independence.

Most were identified and brought into the program when their children were still infants.  Thus, for many, child care was perceived as a major barrier to participation in the programs.  Although the program helped participants find and pay for child care, resistance to considering care by nonrelatives was high, primarily because the young mothers felt they could never trust a stranger to care for their children.

Most of the young mothers in the focus groups agreed that future childbearing should be either postponed for a long time or avoided altogether.  Many, however, acknowledged having problems with using effective birth control consistently.

Only a handful of mothers in the sample cooperated with the child support enforcement agency, even though sizable numbers were in contact with the fathers of their children and many received support from them.  Even those who received modest informal support from the fathers, however, generally felt it was in their best interest to resist cooperation with the enforcement agency.

Program Participation

Through the efforts of committed staff, the programs succeeded in achieving participation rates that compare very favorably with those achieved in other work-oriented welfare programs.  Rates are also quite high given that these were nonselective, comprehensive coverage programs that made commitments to work with all new teenage parents on welfare.  Nearly 90 percent of the eligible teenagers completed program intake.

Of the teenage mothers who completed intake and were assigned to the enhanced-services group, 92 percent (82 percent of the full sample) participated in subsequent program activities.  More than 80 percent completed an extensive assessment and developed a self-sufficiency plan, 72 percent completed one or more program workshops, and 70 percent engaged in at least one of three major activities -- school, job training, or employment.  Many engaged in more than one of these activities; at some time during the demonstration, 47 percent attended school, 29 percent had some type of job training, and 33 percent were employed.

Participation in program activities was highest among those who had higher basic skills, were enrolled in school at intake, did not have any health problems, were black, and/or lived with mothers not employed outside the home.  Participation was lowest among school dropouts who would have been mandatory participants under the JOBS program (30 to 35 percent in any month, compared with 40 to 50 percent for high school graduates and those in school at the time of program enrollment).  Spells of inactivity were common among participants, with 80 percent having at least one spell of inactivity and more than 25 percent having multiple spells.

Role of Mandatory Participation Requirements

The mandatory participation requirement and sanction policy compelled many of the teenage parents to get involved in the program and maintain their participation.  The sanction policy was especially helpful in gaining initial cooperation with participation requirements as evidenced by the fact that nearly two-thirds of the young mothers responded only after the threat of a sanction (See Figure ES-1 below).

Program Enrollment Rates by Level of Encouragement

Overall, 62 percent of those who completed intake were warned at some time of possible sanction because they failed to fulfill requirements for ongoing program participation.  More than one third had their grants reduced one or more times for failure to comply with ongoing requirements.

Impacts on Economic, Social, and Demographic Outcomes

The prospects for these young mothers were not promising without some form of intervention, judging by the experiences of those in the regular-services group.  Two years after starting to receive welfare for themselves and their first child, 75 to 80 percent would still be on welfare; 80 to 90 percent would be living in poverty; over half would have another child; only 10 to 25 percent would have a job; fewer than 10 percent would be living with a spouse or a male partner; and only one fourth would have regular contact with the fathers of their children.  Poverty rates would exceed 90 percent for those who did manage to get jobs. (See Table ES-3)

The benefits of participation in the demonstration programs included increased rates of school attendance, job training, and employment.  The program-induces increases in employment were accompanied by earnings gains that, in combination with program sanctions, resulted in lower rates of dependence on public assistance.  There was little or no measurable change in economic welfare, however, except for those who became employed.  The hoped-for improvements in social and demographic outcomes generally have not been observed to date.  Also, we have not yet examined possible impacts on the children of these young mothers.


Outcome Measures Regular-
Group Mean
Group Mean
Estimated Impacts

Percent Change

Activity Levels and Choices (Two Years Following Intake)
School, Job Training, or Employment (Percent) 66.4 78.8 12.4** 18.7**
Percent of Months Active 27.5 35.2 7.7** 28.0 **
In School (Percent) 29.3 41.6 12.3** 42.0**
In Job Training (Percent) 22.6 26.8 4.2** 18.6**
Employed (Percent) 43.1 48.2 5.1** 11.8**
In Job Club (Percent) 6.7 24.9 18.2** 273.7**
Income, by Source (Two Years Following Intake)
Monthly Earnings $114 $137 $23** 20.0**
Monthly AFDC Benefits $261 $242 -$19** -7.3**
Percent of Months Receiving AFDC 80.9 78.3 -2.6 -3.2
Monthly Food Stamp Benefits $127 $125 -$2 -1.6
Percent of Months Receiving Food Stamps 70.2 67.7 -2.5** -3.6**
Monthly Child Support $23 $20 -$3 -13.0
Social and Demographic Status (at Follow-Up)
Percent with Income Below Poverty 86.2 84.9 -1.3 -1.5
Living with Supportive Adult (Percent) 50.0 51.5 1.5 3.0
Living with Spouse or Male Partner (Percent) 8.8 9.9 1.1 12.5
Paternity Established (Percent) 46.2 49.8 3.6** 7.8**
Receiving Regular Financial Support from Child's Father (Percent) 10.0 9.3 -0.7 -7.0
In Regular Contact with Child's Father (Percent) 26.1 27.9 1.8 6.9
Number of Repeat Pregnancies 1.00 1.01 0.01 1.0
Number of New Births 0.60 0.64 0.04* 6.6*

Sample Size 1,924 - 2,275 1,943 - 2,284 3,867 - 4,559 3,867 - 4,559

Source:  Earnings, AFDC, food stamp, and child support data are from administrative records.  All other data are from follow-up surveys conducted an average of 28 months after sample intake.  The larger sample sizes pertain to those outcome measures derived from administrative data sources.

 *  Statistically significant at the 10 percent level, two-tailed test.
** Statistically significant at the 5 percent level, two-tailed test.

Program-Induced Increases in Activity Levels

Overall levels of participation in school, job training, or employment over the two years following intake were substantially higher than they would have been in the absence of the programs and the supportive services they offered.  Only two-thirds of those receiving regular AFDC services were in school, job training, or a job during the two years after sample enrollment, while nearly 80 percent of the enhanced-services group members were active.  The net result was a 12 percentage point (19 percent) increase in participation levels. (See Table ES-3)

For all three sites, the programs were most effective in increasing school enrollment levels -- resulting in estimated increases of 12 percentage points, from 29 to 41 percent.  Program-induced increase in the likelihood of receiving job training or having a job over the two years following intake were substantially smaller (four and five percentage points, respectively).  Over the follow-up period, 43 percent of those receiving regular services and 48 percent of those in the demonstration programs had some employment; 23 percent of those receiving regular services and 27 percent of those in the demonstration programs had some type of job training. (See Table ES-3)

The impact of the programs on overall activity rates, school enrollment, and employment emerged very early after intake and persisted throughout the 24 months of enrollment.  During any month, between one-fourth and one-third of the enhanced-services group (including those still receiving AFDC and those not) were in school, job training, or employed, compared with 19 to 29 percent of those in the regular-services group.  (See Table ES-3)

The net result is that the enhanced-services group members spent a much higher proportion of their time in school, job training, or employment than did those offered only regular services.  For example, over the 24 months following intake, those in the regular-services group were active 27 percent of the time, while those in the enhanced-services group were active 35 percent of the time -- a 28 percent increase.  (See Table ES-3)

It is especially notable that the pattern of impacts was similar across all three sites.  Yet, the programs tended to promote different types of activity gains among various groups.  For example, they tended to increase school attendance most among younger mothers, those with low basic skills, and those who had not graduated from high school -- characteristics that parallel the JOBS-mandatory participation requirements.  Impacts on job training and employment were especially large for older youth and those with higher basic skills.

Impacts on all three activities were largest for Hispanics.  Compared with those in the regular- services group, Hispanics in the enhanced-services group were 55 percent more likely to engage in a major activity (74 versus 49 percent), twice as likely to attend school (42 versus 21 percent), 37 percent more likely to have job training (23 versus 17 percent), and 68 percent more likely to have a job (42 versus 25 percent).

Impacts on Income Sources and Economic Status

The demonstration programs altered the income sources for the young mothers somewhat.  As noted earlier, significantly more of the enhanced-services group received income from employment than did their regular-services group counterparts -- differences that resulted in an average of $23 a month more in income from employment, most of which was due to the higher employment rate among the enhanced-services group.  These earnings gains contributed to significantly lower participation rates in the food stamp program, but not lower average food stamp benefit amounts.  They also led to lower AFDC benefit amounts, but not lower AFDC participation rates.  Child support payments were relatively low (an average of only $23 per month) and similar between the regular- and enhanced-services groups.  (See Table ES-3)

Program-induced impacts on income sources began to emerge in all three sites during the second quarter after program enrollment, as participants finished their workshops and some moved into employment.  The impacts persisted through the two-year follow-up period, particularly in Camden.  The average program-induced reductions in AFDC benefits tended to parallel the employment impacts.  Moreover, the patter of relatively larger impacts on AFDC benefits among those with higher basic skills, Hispanics, and older youth (a five percentage point reduction) was similar to that observed for earnings.

Regardless of whether young mothers participated in the demonstration programs, the vast majority were living in poverty at the time of the follow-up survey.  Only those who got a job were able to escape poverty in large numbers -- 80 percent of those employed versus 2 percent of others.

Social and Demographic Impacts

The programs produced few significant impacts on social and demographic outcomes.  Moreover, there were no consistent patterns across sites in the few estimated changes observed.  Notable results did occur for living arrangements, child support, and pregnancies and births.  (See Table ES-3)

Although there was no significant overall impact on the likelihood that those in the enhanced- services group would live with a potentially supportive adult -- a parent, spouse, or male partner - - there was a sizable and statistically significant impact in Camden.  There, 52 percent of the enhance-services group were living in such arrangements at the time of follow-up, compared with 47 percent of those receiving regular AFDC services.

An overall four percentage point increase in the incidence of paternity establishment for the study sample as a whole resulted from the significant impacts of four and five percentage points, respectively, observed in Newark and Chicago.  There were not impacts observed in Camden.  Moreover, the impacts in Newark and Chicago were substantially larger among those enrolled in the program later in time, when the demonstration child support services were stronger.  These higher rates of paternity establishment were not, however, accompanied by significantly higher levels of child support.

A different pattern of impacts was observed for financial support form noncustodial fathers.  At the time of the follow-up survey, the enhanced-services group members in Camden were nearly 30 percent more likely than their regular-services group counterparts to receive financial support from their children's fathers and to have regular contact with them.  In contrast, negative or no impacts were observed in Newark and Chicago.

There were no significant impacts on pregnancy rates.  Young mothers in the enhanced-services group, however, were somewhat more likely than their control-group counterparts to report subsequent births -- an increase concentrated in the Chicago site and among older participants.  In all three sites, there was a tendency for the program to reduce pregnancy and birth rates among the younger participants and Hispanics -- groups that were most prevalent in Camden.

Program Design and Implementation Issues

Implementing mandatory programs for welfare-dependent teenage parents presented major challenges not previously addressed in the AFDC program.  Success depended on staff's acceptance of the notion that it was appropriate to target teenage parents for this type of intervention.  It also depended on their accepting, or at least tolerating, the idea of requiring these young mothers to go to school, job training, or work (and imposing consequences on those who failed to accept this responsibility), though complying with program requirements meant the mothers had to leave their babies in the care of another person for substantial blocks of time.

The programs had to recognize and address the special circumstances that prevented some young mothers from maintaining a full-time schedule of work or school.  For some, these circumstances were episodic.  Nonetheless, when they occurred, it was essential for the program to offer services designed to help the mothers conquer the barriers.  Staff had to provide follow-up and use project resources for those in need, including those whose underlying reason for nonparticipation or noncooperation was not immediately evident.  For example, a case manager who took the initiative to visit the home of a young mother who repeatedly failed to show up for program classes found that the participant and her partner had to sleep in shifts at night so that one of them could guard their baby's crib against rats at all times.  The case manager helped the couple find better housing and the young mother began attending program classes.

Four aspects of program implementation were especially challenging:

  1. Outreach and recruitment;
  2. Designing appropriate workshops;
  3. Case management; and
  4. Developing appropriate school, job training, and employment options.

Enrolling Teenage Mothers in the Program

In contrast to many small-scale, voluntary programs for teenage parents, the intent of the Teenage Parent Demonstration -- as with the adolescent parent provisions of the Family Support Act -- was to serve all teenage parents who met the program eligibility criteria.  To achieve this goal, the programs had to develop systems of universal identification of eligible young mothers, plus outreach and follow-up procedures to promote initial and ongoing participation.

The experiences of the demonstration underscored the desirability of early identification and referral.  It also highlighted the importance of attention to case detail and strong quality control in the process of identifying minor mothers.

A combination of manual and automated procedures for identifying eligible teenage parents was essential in this demonstration and most likely would be required in any replication.  Manual identification procedures, while time-consuming, have the advantage of providing an opportunity to motivate clients from the start.  In contrast, automated procedures tend to be less burdensome, but more prone to error (because of inconsistencies in data input), and require a longer lag between AFDC enrollment and identification.

The programs achieved high rates of initial and moderate rates of ongoing participation -- an achievement founded on the mandatory participation requirements.  Case managers were held accountable for helping the young mothers to address their barriers to participation and the young mothers had to comply or face financial penalties.  They used a variety of approaches to overcome clients' reluctance to participate -- reasoning with them, encouraging them, and speaking with clients' mothers to win support for their daughters' participation, for example.  They chided clients when they missed appointments or slacked off in attendance, and reminded them they had to choose between participating and a sanction.  These efforts often extended over long periods, with many clients going through recurrent cycles of participation and resistance.

Program policies and actions that facilitated case managers' efforts to promote participation included offering flexible schedules, providing on-site child care, promoting informality at meetings, using group meetings to break down isolation, and assigning participants to case managers immediately following intake.

Designing Workshops for Teenage Parents

Demonstration workshops served three purposes: 

  1. As a way for participants to acquire important information -- about nutrition, drugs, family planning, workplace demands, parenting, child support, and other topics;
  2. As personal development tools -- integrating participants into the program; building motivation, interpersonal skills, and program acceptance; and dispelling fears about the program; and
  3. As assessment opportunities -- enabling program staff to assess participants' behavioral and cognitive strengths and weaknesses directly.

Staff in all three programs generally agreed on the purposes of workshops, but adopted quite different approaches to integrating them into the sequence of program activities -- approaches that differed in the length of initial workshops, the types of staff used to conduct workshops, and the emphasis placed on initial versus ongoing workshops.

The extensive initial workshops offered in New Jersey provided greater opportunity for socialization and formation of peer relationships.  The extensive workshops also offered more opportunity for participants' personal interests, communications and social skills, family problems, and motivation to be clarified before decisions were made about involvement in ongoing education or training.  On the other hand, the brief, closely spaced workshops offered in Chicago held the young mothers' generally brief attention more successfully.  Moreover, the briefer workshops allowed new participants to move more quickly into substantive education or training.

Using case managers to run workshops had several advantages, the most obvious being that case mangers used the workshops as assessment opportunities for participants assigned to their caseloads.  Using regular case managers to run workshops also held down program costs but added to the strain on them and limited their opportunities to tap outside expertise.

The programs tapped a variety of sources of expertise and specialized skills for workshops.  For example, under contracts or occasionally no-costs interagency agreements, workshop leaders came from the local Planned Parenthood Association for family planning workshops, from county extension services for nutrition and life skills workshops, and a nonprofit drug rehabilitation program for an AIDS/drug abuse workshop, and several small nonprofit agencies for life skills and grooming workshops.

Case Management

Because of the complex needs and diversity of this population, strong case management was an essential feature of the programs.  The demonstration experience highlighted the importance of individualizing services for young welfare mothers and of modifying them over time as necessary.  This individualization can best be accomplished if a single staff person -- such as a case manager or continuous counselor -- becomes familiar with a teenager and has ongoing responsibility for her.  Case managers spent much of their time trying to find the right combination of supportiveness and helpfulness on the one hand, and pressure and clear expectations on the other.  These efforts sometimes extended to home visits.

Case management services were almost universally appreciated by the young mothers.  Although many were sanctioned or warned that they would be sanctioned, their feelings about the program were generally extremely positive.  Praise for the personal and caring attention of case managers and other program staff was especially high.  Case managers linked the teenagers to the services they needed, monitored their progress in the program, offered advice and guidance for personal problems, and provided much-needed support and encouragement.  For many young mothers, the case managers also served as role models or surrogate parents:

"When I go to Project Advance, they know me, they speak to me . . . If I do something that's stupid, they know it, they tell me.  When I do something good, they all praise you for it.

It is the one program that actually motivates someone to do something.  There was always something going on even if you weren't working.  You didn't have to be in the street, and you didn't have to sit in your house." (A client)

In these programs, as in other initiatives that are run as part of the state welfare system, flexibility in recruiting staff trained to work with this population was limited.  As a result, substantial staff training and skilled supervision were needed to help case managers work effectively with the broad spectrum of clients they had to serve.

Program managers had to organize their staffs and define roles for a broad range of functions:  counseling individual clients, leading group intake sessions, conducting program workshops, maintaining client case records, entering data into automated systems, collecting and recording attendance data for on-site and off-site program activities, issuing sanction warning notices and communicating with income maintenance to impose or end sanctions, developing child care resources and arranging child care, developing contacts with community service providers, and coordinating special tasks to support the research data needs.  Program managers helped case managers by providing specialized staff roles, developing service links, providing adequate supervision, monitoring and controlling caseloads, and promoting staff morale and stability.

Education and Job Training Services

To enforce mandatory participation requirements, programs had to make adequate activity options available for the young mothers.  This proved to be a major challenge.  Job training was available through many providers, including community colleges, vocational high schools, JTPA, and proprietary schools.  Access tended, however, to be restricted primarily to high school graduates and those with reasonably strong basic skills.

All three sites had ample number of educational opportunities, but the programs often failed to meet the range of needs reflected among the large portion of the population for whom education was the only immediately available option.  In addressing the varied needs, the programs used a combination of existing and new educational programs; each type of program offered advantages and disadvantages.  Existing General Educational Development (GED) and Adult Basic Education (ABE) programs were provided primarily by community colleges and adult schools operated by local school districts.  The young mothers, however, often felt uncomfortable in classes with older adults and teachers accustomed to serving a broader adult population tended to be insensitive to the problems faced by teenage parents.  In-house classes ensured that participants talked frequently with their case managers and that case managers and academic instructors could maintain close ties.  Nonetheless, all educational programs still had to deal with the limited attention spans of the young mothers and their lack of interest in noncontextual learning.

Successful educational alternatives included intensive and very personalized academic instruction built around group interaction and cooperation, support counseling, group research projects in the community, and paid work experience assignments.  One option placed participants in alternative secondary educational settings, which sometimes required a parental waiver of school district responsibility and negotiation with school district officials who were reluctant to see their regular enrollments diminish.

Looking Ahead

This demonstration offers important lessons for current efforts to redefine the nation's social welfare policies.  We have documented that states can operate large scale, mandatory work- oriented programs for the most vulnerable welfare recipients -- teenage parents.  These programs can achieve relatively high participation rates and will promote at least the necessary first steps toward self-sufficiency -- significantly increased school attendance, job training, and employment.  Both program staff and the young mothers who were required to participate in these programs felt that the programs were helpful and that the participation requirements were fair -- case managers would say the requirements were essential to the program's success.

Two important questions remain unanswered from this study.  One is whether these early education, training, and employment impacts of the program will translate into longer-term increases in self-sufficiency.  This question is being addressed through a longer term follow-up of the study sample and their children.

The second unanswered question is how much improvement in outcomes could be achieved in programs that can address some shortcomings of the demonstration programs.  For example, still higher participation rates and improved longer-term outcomes likely could be achieved if the programs had greater capacity to create stronger education and training options tailored to the needs of the young mothers; if they found more effective ways to help the young mothers control their fertility -- something most of them want to do; and if the child support system had stronger incentives for parents -- mothers and fathers -- to cooperate and for the agency to focus on this population, which has limited prospects for making significant support payments in the short run.  Architects of future welfare policies and programs should build on the operational success of this demonstration in serving very large caseloads of teenage parents and in sensitively, but effectively, using sanction to promote participation in sufficiency-promoting activities, but also attend to the areas in which the program and/or its implementation could be strengthened.

Summary of Key Findings and Lessons Learned

  • It is feasible to design and operate mandatory education, training, and employment programs that serve large numbers of teenage parents, despite the fact that the needs of this population differ substantially from those of adult AFDC recipients.
  • With active monitoring of participation, it is possible to achieve significant rates of participation.  Nearly 90 percent of the young mothers who were identified as eligible for the programs enrolled and, of those who enrolled, 92 percent participated in program activities beyond the initial intake and assessment.  Throughout the demonstration period, the programs kept between 30 and 50 percent of the young mothers actively in school, job training, or a job, while others were enrolled in workshops and other activities preparatory to entering one of these major self-sufficiency-oriented activities.
  • The number of new AFDC applicants who are teenage parents is a relatively small proportion of the applicant caseload -- 17 to 26 percent in the demonstration sites.  Therefore, intervening early and implementing programs to serve this population are not likely to require a significant share of overall program resources.  Furthermore, effectively serving teenage parents as they come onto AFDC could have large impacts on the overall welfare caseload, because of the relatively long expected durations of dependency among this population.
  • About one-third of the young mothers used agency-funded child care.  The remainder relied on unpaid care -- generally provided by relatives -- or paid for the care themselves or with the assistance of family members.  An important part of the child care supportive services component of the demonstration was the provision of adequate information to alleviate the fears and worries of these young mothers about leaving their child in the care of someone else, especially a stranger.
  • These types of programs can promote significant and sustained participation in education, training, and employment activities -- activities that are likely to affect the young mothers' long-run prospects for self-sufficiency.
  • The largest impacts on participation were on school enrollment (12 percentage points) -- a 42 percent increase over the levels of participation these young mothers would have had under the regular AFDC regulations and services.  This finding is in line with the expectation that, within state JOBS programs, educational activities should be the primary activity for most teenage parents who have not yet successfully completed high school.
  • The demonstration programs also led to significant increases in employment and participation in job training (gains of 4 to 5 percentage points -- 12 to 19 percent).
  • Programs for teenage parents can achieve these significant impacts while case managers maintain relatively large caseloads -- 50 to 80 active cases or total caseloads of 100 to 140 young mothers.  The keys to effectively managing caseloads of this size were reliance on automated case tracking systems and routine use of the mandatory participation requirements as a case management tool.

These findings are especially noteworthy for three reasons:

  1. In contrast to most other welfare demonstrations, we found consistent results in all locations where programs were operated.  Not only were the operational findings reinforcing and complementary across the sites, but also the pattern of program impacts on activities, as well as on economic, social, and demographic outcomes, was similar in all three demonstration sites.
  2. This is one of a handful of programs for teenage parents or disadvantaged youth in general that have been shown to have significant positive impacts for participants.  Moreover, none of the other programs with positive impacts except Job Corps -- a residential program for at-risk youth -- shows as strong or as consistent a pattern of results as we found for this demonstration (see for example, Mallar, Kerachsky, and Thornton 1982;  Quint and Riccio 1985;  Polit and White 1988;  Bloom et al. 1993;  and Maxfield 1990).
  3. The scales of operation, administrative structures, and funding levels of these demonstration programs were consistent with those that would be encountered in a full-scale, national implementation of the program model.  This is not the case with most programs for teenage parents insofar as others have tended to be very small scale (less than 100 participants and often many fewer than this), to serve a selective volunteer population, sometimes to devote substantially more resources per participant than was available for this demonstration, and to be operated outside of the welfare programs.  The three programs operated in this demonstration could, indeed, be replicated on a large scale and within the welfare system.

Author and Contract Information

Report prepared by:


Mathematica Policy Research, Inc.
P.O. Box 2393
Princeton, NJ 08543-2393
TEL: (609) 799-3535
FAX: (609) 799-0005

Publication date: June 1993

Project Director: Rebecca A. Maynard

Deputy Project Director: Alan M. Hershey

MPR Reference No: 7700-021

Report prepared for:

U.S. Department of Health and Human Services
Assistant Secretary for Planning and Evaluation (ASPE)
Hubert H. Humphrey Bldg, room 404E
200 Independence Ave., SW
Washington, DC 20201

Federal project officers: Reuben Snipper, ASPE and Nancye Campbell, ACF

Report was prepared for, and funded by: The Assistant Secretary for Planning and Evaluation (ASPE) and the Administration for Children and Families (ACF), U.S. Department of Health and Human Services (DHHS), under contract HHS-100-86-0045.


This report does not necessarily represent the official opinion or policy of the Department of Health and Human Services. The results of this study and the views expressed are solely those of the author.


This paper is a synthesis of the results of the first phase of a major, ten-year demonstration and evaluation effort (TPD Home Page). Specifically, it draws heavily on a series of process and implementation reports (Hershey and Nagatoshi 1989; Hershey 1991a, 1991b, and 1991c; Hershey and Silverberg 1993; Hershey and Rangarajan 1993; and Maynard 1992); a report on a special in-depth study of the target population and its program experiences (Polit 1992); a report on service needs of and use by the target population (Gleason et al. 1993); and an impact analysis report based on two to four years of follow-up data (Maynard, Nicholson, and Rangarajan 1993). (See Project Reports)

Authors of these reports as well as many other individuals contributed in significant ways to the preparation of this report. Much of the information on the demonstration programs was provided by Melba McCarty, Yvonne Johnson, and Frank Ambrose, who oversaw the demonstration programs -- Project Advance in Chicago, Teen Progress in Newark, and Teen Progress in Camden, respectively. Janet DeGraaf, Bonnie Mecanko, and Lydia Davis-Barrett of the New Jersey Department of Human Services and Denise Simon, Dan Davis, David Gruenenfelder, and Charlie Mugler of the Illinois Department of Public Aid were instrumental in the design, implementation, and oversight of the programs and were extremely helpful during the information assembly process. And, the young mothers in the evaluation sample patiently answered our many questions. Reuben Snipper, project officer for the evaluation, and Nancye Campbell and Judith Reich, project officers for the demonstration, provided helpful guidance throughout the design and implementation of the evaluation. William Prosser from the office of the Assistant Secretary for Planning and Evaluation provided guidance at various points throughout the project, but especially in the design and implementation of the special substudies on child care issues, and Howard Rolston at the Administration for Children and Families provided invaluable support of both the demonstration projects and the evaluation.

Alan Hershey, co-investigator and deputy project director, worked diligently with the programs to ensure that they were implemented in accordance with the demonstration design and that we documented the operational lessons from the demonstration. He also directed the process and implementation analysis. Denise Polit, co-investigator, was instrumental in the development of the overall evaluation plan and directed a companion in-depth study of a subsample of program participants; Myles Maxfield designed and oversaw the development of the client tracking systems; Charles Nagatoshi provided technical assistance to the programs and monitored their operations; Anu Rangarajan and Walter Nicholson played key roles in the impact analysis; Phillip Gleason, Anu Rangarajan, and Denise Polit collaborated on a report on service needs and use; Marsha Silverberg worked with Alan Hershey on the cost analysis; and Ellen Kisker, Anne Gordon, Margaret Strain, Marsha Silverberg, and Peter Schochet conducted the special child care studies. Anne Bloomenthal, Sandra Scott, West Addison, Cynthia Butchley, Lynn Leubuscher, and Dexter Chu constructed the data files and prepared the tabulations and graphs for the report. John Homrighausen and Cindy Pressler directed the survey data collection. Doreen Ambrose, Monica Capizzi, Cindy Castro, Debra Jones, and Marjorie Mitchell produced the report. Joanne Pfleiderer copyedited the report.

I gratefully acknowledge these contributions and accept sole responsibility for any remaining errors or omissions in the report.

Rebecca Maynard