This report aims at filling some of the gaps in the past research by systematically examining the implementation of maternity group home programs across a large number of sites. To do this, we selected interesting maternity group home programs in seven states and conducted site visits to more than 20 different homes.
Site Selection. In determining which maternity group home programs to focus on in this study, we found it useful to first clarify what exactly a maternity group home is. For the purposes of the study, we defined a maternity group home as a residential program providing substantial supervision and other services primarily to pregnant and/or parenting teenagers.4 This definition excludes programs that provide only limited adult supervision — such as having staff on site only during standard business hours on weekdays — as well as programs that provide housing and supervision but no other services to their residents.
After restricting the definition, we developed a set of other criteria to help us select sites for this study. Because one main goal of the current study is to assess the feasibility of conducting a more rigorous evaluation of the effectiveness of select maternity group homes, our emphasis was on sites that seemed to have the highest potential for inclusion in such an evaluation. To this end, we focused primarily on programs that met the following four broad criteria:
- Strong, well implemented intervention
- Ability to generate adequate sample sizes
- Services offered distinct from what participants would otherwise receive
- Unmet need for program services
We relied on a variety of resources to explore how well various maternity group home programs meet each of these criteria. These resources included the directory of maternity group homes compiled by the Social Policy Action Network (SPAN), a review of past studies of maternity group homes, Internet searches, reviews of program websites, and preliminary telephone conversations with staff from certain programs.
It was difficult to find programs that clearly met all four criteria. In particular, identifying programs that were both large enough to generate adequate samples for an evaluation and operating in environments with clear evidence of substantial excess demand for program services proved to be particularly challenging. When necessary, we prioritized the ability to generate adequate sample sizes above the existence of unmet need for program services.
Because of these selection criteria, the homes included in this study are not necessarily typical or representative of maternity group homes nationwide. In particular, because a rigorous evaluation would require large sample sizes, our selection process focused on relatively large maternity group home programs — those serving relatively large numbers of resident families. To achieve these numbers, most of the sites included in this study are actually either networks of homes operated by several different organizations or programs in which multiple homes are operated by a single organization. Thus, although the individual homes included in our study are not necessarily any larger on average than other homes, those in our study are more likely to be part of larger programs. These program ties may have implications for homes’ funding, structure, and operations.5 For example, being part of a larger program my provide the homes in our study with access to resources that independent homes may not have. In addition, homes that are part of larger programs may have to follow program guidelines concerning whom to serve or how to structure the home. Thus, the homes included in our study may be different from other maternity group homes in a variety of ways.
Data Collection. Once we had selected the sites, we contacted the director of each program to request program documents and plan a site visit to gather additional information. Two site visitors went to most programs and spent two or three days in the program’s coverage area. During the visits, we met with network- or program-level staff at each of these seven sites, in addition to visiting maternity group homes themselves. The text box below shows the topics discussed during conversations with staff.
|Topics Discussed During Site Visits|
At each program, we met with the program director and other staff of the network agency or managing organization. We collected any additional program documents that were available. In some cases, we also met with staff of an agency that provided the majority of funding or referrals for the maternity group home program.
We visited every maternity group home in the three programs that included fewer than four homes each (see Table I.1). In the other programs, we visited between three and five homes each. We selected which homes to visit in consultation with the network or program director, based on representativeness, variety, and location. In some cases, we also visited other, unrelated, homes in the same area. At each home, we toured the facility and met with key staff. Where possible, we also spoke with residents and observed program activities.
|State||Program Name||Number of Homes||Capacity in All Homes|
|Georgia||GCAPP Second Chance Homes||8||4||44|
|Maine||St. Andre Group Homes||4||3||16|
|Massachusetts||Teen Living Program||20a||5||167|
|Michigan||Teen Parent Supportive Housing Services Collaborative||3||3||34|
|New Mexico||Teen Parent Program||5||4||38|
|New York||Inwood House Maternity Residence||1||1||36b|
|Washington||Friends of Youth Transitional Living Program||2||2||20|
|GCAPP = Georgia Campaign for Adolescent Pregnancy Prevention
a In addition to these 20 maternity group homes, the Massachusetts program includes three homes designed specifically for more mature teens who have completed a regular maternity group home program and are deemed ready by program staff to transition to semi-independent living. However, these three homes do not fit the definition of maternity group home used in this study due to the lower levels of supervision provided.
b Inwood House officially has the capacity to serve 36 residents. However, the home has been operating below this capacity for some time and, in response, has reduced staff and converted some space for other uses. Thus, in this report, we consider the capacity to be 24 when calculating staffing ratios and costs per resident.