SHM PROJECT-12 month data collection OMB Justification statement.Section A

SUPPORTING STATEMENT

FOR OMB CLEARANCE

PART A

DHHS/ACF

SUPPORTING HEALTHY MARRIAGE (SHM)

PROJECT EVALUATION

LOW-INCOME MARRIED COUPLES DATA COLLECTION ACTIVITIES –

12-MONTH FOLLOW-UP DATA COLLECTION

November 26, 2007

TABLE OF CONTENTS

A. JUSTIFICATION

A1. Circumstances necessitating data collection 4

A1.1. Previous research on marriage education 5

A1.2. Overview of SHM evaluation 6

Major research questions 6 The SHM program model 6

Key components of the SHM evaluation 7

Overview of data collection components 9

Timeline for the SHM evaluation 10

A2. How, by whom, and for what purpose data are to be used 11

A2.1 The overall role of the SHM 12-month data collection effort 13

A2.2 The Role of Each 12-Month Data Component 14

12-month survey 14

12-month observational study 27

Qualitative data collection efforts for implementation research 31

A3. Use of information technology for data collection to reduce respondent burden 32

A4. Efforts to identify duplication 32

A5. Burden on small business 32

A6. Consequences to federal program of policy activities if the collection of

information is not conducted or is conducted less frequently 33

A7. Special Data Collection Circumstances 33

A8. Form 5 CFR 1320.8 (d) and consultations prior to OMB Submission 33

A9. Justification for Respondent Payments 36

A10. Confidentiality 38

A10.1 Confidentiality and Follow-up Data Collection Activities 39

A11. Questions of a sensitive nature 40

A12. Estimate of the hour burden of data collection to respondents 41

A13. Estimates of capital, operating, and start-up costs to respondents and 43

record keepers

A14. Estimates of costs to federal government 43

A15. Changes in burden 43

A16. Tabulation, analysis, and public plans and schedule 43

A16.1 Analysis plans for 12-month data collection efforts 43

A16.2 Storing, analyzing, and presenting qualitative data for implementation research 46

A17. Reasons for not displaying the OMB approval expiration date 47

A18. Exceptions to Certification Statement 47

A. Justification

A1. Circumstances Necessitating Data Collection

Recent declines in marriage in the United States have had disproportionate ill effects on children living in poverty. Increasing numbers of children living in poverty are born to unmarried parents. Children of poor married parents are also twice as likely as children of affluent married parents to experience their parents’ break-up (McLanahan & Sandefur, 1994). At the same time, an accumulating body of evidence points to markedly better outcomes when children are raised by married parents, and suggests that these differences are partly due to marriage’s effects on the income, relationships, and quality of parenting available to children (Amato, 2000; McLanahan & Sandefur, 1994).

For these reasons, as the federal government and state governments develop new programs and policies to inform provisions in the 1996 TANF legislation to support the formation and maintenance of two-parent families, there is great interest in preventive strategies aimed at improving the quality and duration of marital relationships. Thus, in 2001 the Administration for Children and Families (ACF) within the U.S. Department of Health and Human Services launched its Healthy Marriage Initiative.

The ACF initiative seeks to help couples who choose marriage for themselves access services that will help them develop the skills and knowledge to form and sustain healthy marriages. These services center on research-based marriage education curricula developed by experts in the field. Prior to the ACF initiative, these services were primarily available to middle- and upper-class couples, and formal evaluation of these programs was limited to a series of small-sample studies. The initiative emphasizes broadening access to marriage education services to low-income populations while the marriage education services are accompanied by other supports and referrals that families may need in order to participate in and to sustain healthy marital relationships. It also includes a rigorous evaluation agenda.

The Supporting Healthy Marriage (SHM) Demonstration and Evaluation is the first large-scale, multi-site test of marriage education programs for low-income married couples.¹ The project is being conducted by MDRC, a non-profit, non-partisan social policy research organization, under contract to the Administration for Children and Families in the Department of Health and Human Services. MDRC is also working with a team of partners on this study, including Abt Associates, Child Trends, Optimal Solutions Group, and McFarland and Associates, as well as a group of experts in the field of marriage education. This project offers a tremendous opportunity to build knowledge about how to strengthen and maintain healthy marriages and enhance child development. The study design is based upon random assignment, the strongest known method for assessing program impacts. The multi-site structure provides flexibility to assess a variety of approaches to marriage education over a long follow-up period. This well-designed study will illuminate the determinants of healthy marriages, and provide important information about the causal links between such improvements and outcomes for children, adults, and families.

A1.1 Previous Research on Marriage Education

There is evidence that interventions designed to teach couples new skills for relationship functioning can improve both the quality and stability of their marriages (Halford et al. 2003; Silliman, et al, 2002; Reardon-Anderson et al., 2005). Among experimental studies that were included in prior reviews, twelve reported findings for six or more months of follow-up (Halford et al., 2003; Silliman, et al, 2002; Reardon-Anderson et al., 2005), and nearly all of these reported at least some longer-term positive effects on couple interaction, satisfaction, or union stability. Overall however, these evaluations focused on small, middle class samples, often had non-experimental research designs, and examined a limited range of outcomes. In addition, effects tended to fade over time, indicating a need to develop and test more intensive or extended models. Most of the relationship skills training programs tested in early trials were relatively low-intensity interventions. Higher-intensity approaches – for example, a combination of either more hours of program services or for a longer period of time combined with information about other services addressing stressors, such as financial, mental health, alcohol and substance abuse, and employment issues– may be particularly critical for low-income populations given their high rates of exposure to a variety of personal and financial challenges that can strain couples’ relationships.

Findings from several recent experimental studies address some of the limitations of earlier programs and are of particular interest. These interventions often lasted for several months and focused on couples making the transition to parenthood or on couples with young children. They have examined impacts on not only marital outcomes but parenting and children’s well-being as well. The Becoming a Family project and Bringing Baby Home program (Bringing Baby Home, 2006; Jordan et al., 2001) both focused on preventing declines in marital satisfaction and enhancing other outcomes during the transition to parenthood. These studies have found a range of positive effects, including marital satisfaction (but not stability) for five and a half years (Cowan and Cowan 1992); infants’ language and emotional development at age one, as well as parenting, co-parenting, father-infant attachment, and couple relationship quality (Shapiro and Gottman, 2005; Bringing Baby Home, 2006).

Three studies focused on parents with slightly older children have also shown positive effects. The Schoolchildren and their Families study found that couples in a group intervention aimed at marital relationships showed not only greater marital satisfaction, but improved parenting, improved test scores, and reduced behavior problems for children as long as 5 years later (Cowan and Cowan, 2006). An evaluation of the Incredible Years, a parent education curriculum focused on skills for parents of children with behavior problems, found that the intervention was more effective when it addressed marital difficulties, depression, and social isolation rather than simply parenting issues (Webster-Stratton and Taylor, 2001). Most recently, the Supporting Fathers’ Involvement Study – implemented in community-based Family Support Centers in California with a group of primarily Hispanic fathers – found improvements in: relationship satisfaction of both partners, parenting stress of both partners, self-reported anxiety of both partners, fathers' involvement with the day-to-day tasks of childrearing, and parents' descriptions of their children's aggression (Cowan et al., 2006).

Thus, although there is evidence that relationship education programs can strengthen marital quality and longevity, as well as other family outcomes, this conclusion is based on studies that have either been focused on middle class samples or were conducted in relatively “hothouse” conditions by the same academics who developed the interventions. Thus, SHM will make a substantial contribution to our knowledge base by conducting an independent evaluation of an intensive marriage education program, designed for low-income couples and operated at scale by real-world community organizations.

A1.2 Overview of the SHM Evaluation

The SHM evaluation, which began in September 2003, builds on the evidence described in the previous section. It is the first large-scale, multi-year, multi-site rigorous test of marriage education programs for low-income married couples including case management to encourage couples to attend the marriage education activities and to refer them to supplemental services in the community as needed. It is designed to inform program operators and policymakers of the most effective ways to help low-income couples strengthen and maintain healthy marriages. Below, we list the major research questions addressed in the evaluation.

Major Research Questions in the SHM Evaluation

1. How effective is marriage education plus case management for low-income married couples and what outcomes does it affect? Marriage education has shown some positive effects on middle class couples in improving relationship quality. Can similar positive effects be found for low-income married couples? Can marriage education increase marital stability and improve child well-being as well?

2. Who benefits the most and least from marriage education with case management services? Low-income married couples are a diverse group. An important question for SHM is whether marriage education works better for some groups than for others. Do they have different effects for couples about to have their first child or those whose oldest child is about to become a teenager? Do couples whose marriages are troubled benefit more or less than other couples?

3. Why do some marriage education programs work better than others? Because SHM deals with a relatively new type of social intervention, implementation research holds the promise of being able to identify best practices. The project will describe each site’s goals and service models; the start-up challenges sites faced; and early lessons on designing marriage skills programs, securing program funding, building interagency partnerships, identifying and recruiting couples, and encouraging participation.

The SHM Program Model

The goal of the SHM programs operated in these sites is to help interested married couples understand how to form and maintain better relationships, to become better parents for their children, and to have healthy long-lasting marriages. To accomplish this, every SHM program will include three components to be delivered over a period of about twelve months: a core marriage education curriculum, extended marriage education activities that continue after the core sessions have been completed, and family support / case management to encourage couples to attend the marriage education activities and to refer them to supplemental services available in the community. All SHM components and services will be voluntary for both members of the couple, and will be free of charge.

The core marriage education curricula will cover a recommended set of topics and will be delivered primarily in a group setting with both partners of the couples present. This component will cover a broad range of topics and last at least 24 hours over several months. The SHM team has worked with program sites to identify curricula that meet both the needs of their target populations and the SHM curricula criteria discussed below. The team will also offer technical assistance to help programs adapt the chosen curriculum to the unique needs of the couples enrolled in their programs. Couples will be most motivated to attend and apply what they are learning if programs tailor their activities to be rewarding, fun, and relevant to program participants.

The second component, extended marriage education activities, will typically begin after the core curriculum is completed. Providers can design this element to include a wide range of engaging activities that reinforce and integrate the skills and concepts learned in the marriage education sessions, as well as to provide new information. Booster sessions, social events, peer mentoring, and marriage coaching are some examples of extended marriage education activities.

The third component of the SHM programs addresses the numerous stressors that destabilize marriage among low-income individuals. Family support is an integral part of the program. Family support coordinators will not only be responsible for helping couples remove stressors and barriers, but will also maintain ongoing contact with couples, providing motivation and facilitating conversations about reinforcement activities jointly created by the family support coordinators and marriage education group facilitators.  Although the SHM programs will not provide any direct intervention or treatment outside the marriage education classes, the family support coordinators will refer clients for necessary services, whether for treatment, employment, housing, or social service benefits.

Key Components of the SHM Evaluation

The SHM evaluation will consist of a start-up pilot phase, an embedded implementation research study, and an impact study examining the effectiveness of marriage education programs targeting low-income married couples with children. We provide a brief description of each phase of the study below.

The pilot phase of the study is currently underway; MDRC and its subcontractors are working with eight pilot sites around the country to implement and test SHM programs. The SHM project includes the following sites:

• University of Central Florida, Orlando, FL

• Catholic Charities, Wichita, KS

• University Behavioral Associates, Bronx, NY

• Public Strategies, Oklahoma City, OK

• Center for Human Services, Shoreline, WA

• Community Prevention Partnership, Reading, PA

• Becoming Parents Program, Seattle, WA

• Texas Health and Human Services, El Paso, TX

During this start-up phase of the project, the research team is working intensively with each site to develop, refine, and pilot test its proposed demonstration project and to put in place the random assignment and data collection protocols necessary to implement the full-scale evaluation. In each site, after the program to be tested was designed in detail, MDRC’s team has supported staff training and the start-up process of getting programs running, and is conducting an assessment of program activities. Beginning in the pilot phase, participants are being randomly assigned to research groups to allow the research team to monitor the random assignment procedures. This will help ensure that the model is being operated as planned and that the flow of clients through the program will be consistent with both program and evaluation requirements. As a result of the pilot experience, projects will refine their program models to reflect any lessons learned and will then begin to assign couples to the actual research sample. MDRC will continually monitor the demonstration programs and examine any research design challenges to develop recommendations to strengthen the programs.

The implementation study is intended to assess the operations of the SHM program and how well the program model is put into place in each of the SHM sites. The implementation study will describe the SHM services, discuss the operating environment and any local circumstances relevant to the programs’ successes, provide insights into the variety of populations served by each SHM program, and identify successful strategies used to recruit, engage, and retain couples in the program. It will also explore the challenges that staff faced in implementing the SHM program models and identify best practices used to overcome these challenges. Finally, the study will be used to better understand the SHM curricula and whether or not participants found the content particularly useful and/or relevant to their daily lives. To collect these data, the SHM team will conduct field visits to each of the SHM programs operating in the eight sites across the United States. Data collection activities will consist of 1) nonparticipant observations of facilitators and couples participating in program activities such as marriage education workshops, one-on-one sessions between family support coordinators and couples assigned to the program group; 2) open-ended interviews and small group discussions with SHM program staff; and, 3) focus groups with some couples in the program group to hear about their experience with the program and to inform the design of future couple-reported surveys. In addition, the SHM team will conduct follow-up surveys (described in more detail below) to learn about participation in SHM services and other similar services for study participants assigned to the program and control groups.

The impact study will examine the effect of the SHM programs on marital quality and stability, parenting behaviors, child well-being, and economic outcomes. This analysis will include impacts across all sites in the national SHM project, as well as site-specific impacts and impacts on different subgroups of the sample. The baseline data, follow-up surveys, observational study, direct child assessments, adolescent survey and couples’ program administrative records will all contribute to the impact study.

Overview of Data Collection Components for SHM Evaluation

The purpose of the current submission is to request OMB approval of the couple-reported survey and the observational study protocol that are being proposed for the 12-month follow-up effort with study participants, and the plans for the qualitative data collection efforts that are part of the implementation research for this project. Justification for the remaining data collection activities will be provided in later submissions.

The data collection components of the implementation and impact studies for the full-scale, multi-year SHM evaluation consist of the following:

• Baseline data. Throughout the pilot and full-scale phases of the SHM evaluation, the research team will collect baseline information from couples in the program and control groups (OMB No. 0970-0299) to help describe the population being served, to assess the validity of random assignment, and to define key subgroups for later analyses. All eight pilot sites have begun collecting baseline data.

• Control services survey. As part of each site’s six-month pilot, the research team will conduct a brief survey (OMB Control No: 0970-0330) to examine the differential between services received by the control group and the program group. The survey will be fielded as SHM sites end their pilot studies, beginning in late 2007.

• Interviewer-administered surveys at 12- and 36-month follow-ups. Follow-up surveys at 12 and 36 months after random assignment will be used to evaluate program impacts. These surveys will include all sample members in the program and control groups, and will measure several outcomes, including marital relationship outcomes (e.g., communication, conflict resolution, time spent together as a couple, commitment to the couple relationship, intimacy, fidelity, marital stability, and satisfaction), parental psychological well-being and health, parental employment and economic outcomes, material and financial hardship, social support and networks, co-parenting relationship, parenting, family functioning and routines, father involvement, and child well-being and adjustment. With the current submission, ACF is seeking OMB approval for the 12-month follow-up survey. A separate submission requesting OMB approval of the 36 month survey will be provided at a later time. The 12-month follow-up is expected to begin in April 2008, and the 36-month data collection will follow two years later.

• 12-month videotaped observations of couple and parent-child interactions. As part of the current submission, ACF is also requesting OMB clearance to conduct an observational study of couple interactions and parent-child interactions at the 12-month follow-up. As described later, the sample for the observational study will consist of a randomly selected subgroup of couples from the survey sample in each site. This videotaped data will then be reviewed and coded by the research team to assess the impact of SHM on the quality and nature of couple and parent-child interactions. The observational study is expected to begin fielding in April 2008.

• Qualitative interviews and small group discussions with program staff and study participants. As part of the implementation research for the full-scale evaluation of SHM services, the research team proposes conducting field visits to program sites to conduct one-on-one interviews, small group discussions with program staff, naturalistic observations of program activities, and focus groups with some couples in the program group. ACF seeks OMB approval for these qualitative data collection efforts with the current submission.

• Direct child assessments. At the 36-month follow-up, the SHM team proposes to conduct direct assessments of children’s emotional and behavioral adjustment, cognitive development, and academic skills. One child per couple will be selected to participate in direct assessments of children’s well-being. These children will range in ages from infancy to adolescence (up to 15 years old at random assignment). We will seek OMB clearance for these data collection activities at the appropriate time.

• Adolescent survey. At the 36-month follow-up, the research plan calls for a short survey with pre-adolescent- and adolescent-aged children (between the ages of 9 and 15 at the time of follow-up). This brief interview will be conducted in connection with the direct child assessments of one child per couple. The adolescent survey will aim to tap, among other outcomes, adolescents’ relationships with parents; time use; sibling, peer and romantic relationships; academic functioning; and emotional and behavioral adjustment. We will seek OMB clearance for this data collection activity at the appropriate time.

Timeline for the SHM Evaluation. For the full-scale phase of the SHM evaluation, random assignment began in March 2007 in the Oklahoma site, and is expected to end in the last site in March 2010. The implementation research will begin after clearance is received from OMB and will continue throughout the demonstration phase. The summary of the findings from the implementation research will be included in an interim implementation report for the evaluation, planned for 2010. The baseline data collection is currently being fielded and is expected to be in the field until March 2010 as well. The 12-month follow-up data collection effort is expected to begin in April 2008 and will be ongoing until March 2011. The 36-month follow-up data collection activities are scheduled to begin in mid-2010 and will be fielded until mid-2013. A draft of the 12-month impact report is planned for late 2011, and a final report is planned for 2013.

A2. How, By Whom, and For What Purpose Are Data to be Used

This document requests OMB clearance for activities related to the 12-month data collection instruments and qualitative research activities that are part of the implementation research. This section of the submission provides more detail on the 12-month data collection instruments and qualitative research activities, as well as the proposed sample for each of these data collection efforts, procedures for collecting these data and the roles that the resulting data will play in the full SHM evaluation.

Proposed sample participating in the 12-month data collection efforts

We propose administering the 12-month survey instrument to all study participants in the full-scale SHM study in each program demonstration site. All of the respondents to the 12-month survey will be adults assigned to one of the research groups. We also propose collecting videotaped observational data at the 12-month follow-up in all eight SHM sites with a random subset of couples (including husbands and wives) who were randomly assigned to the program and control groups with at least one child aged 0 to under 15 years old at random assignment.² As part of the observational study, the couples will be asked to engage in a series of discussion-based interactions alone, and couples or single parents (for couples who are separated at the time of follow-up) will be asked to engage in a series of interactions with one of their children while being videotaped. In most of the SHM demonstration sites, the Focal Child will be randomly selected from all the couple’s or single parent’s eligible child who range in age from 0 to up to 15 years old at random assignment. In the two SHM demonstration sites testing curricula which include a parenting component targeting newborn infants, the child who is 3 months old or younger in the household at random assignment will be purposefully selected to participate in the 12-month survey. The Focal Child will also be targeted for the parenting and child well-being measures of the follow-up surveys and the direct child assessments and adolescent survey collected at the 36-month follow-up. The survey and observational data collected at the 12-month follow-up, together with the follow-up surveys and direct child assessments that will be administered at the 36-month follow-up, will be the primary sources of information for key SHM study outcomes.

Full-scale SHM evaluation eligibility criteria

For the full-scale SHM study, each of the SHM sites will recruit 800 couples over a period of up to 24 months. In each site, 400 of these couples will be randomly assigned to the program group and 400 will be assigned to the control group. Figure 1 shows the intake and baseline data collection procedures for enrolling in the SHM study.

Couples will be considered eligible to participate in the SHM study if they meet the following criteria: 1) The couple is married and has children or is expecting a child; 2) Both spouses are adults (18 or older); 3) Both spouses volunteer to participate; 4) The couple is not experiencing serious family violence issues; and 5) The couple understands a language in which the SHM program is being offered (most sites will offer SHM services in English and Spanish).

SHM sites will employ recruitment strategies that target couples in economically disadvantaged areas and will seek to engage couples with incomes below 200 percent of the federal poverty threshold. Sites are encouraged to meet this objective by working with organizations that already predominantly serve this population; or to locate programs in geographic areas that include large numbers of lower-income families, as well as to tell couples that the program is designed for low-income couples.

Figure 1.

A2.1 The Overall Role of the SHM 12-month Data Collection Effort

Data collected 12 months post-random assignment will serve a variety of purposes in the SHM evaluation. Below, we detail several ways in which the data will be used by MDRC and its partners.

1. To estimate intervention impacts on outcomes that are primary targets of the intervention.

The survey and videotaped observational data will assess key outcomes and provide important sources of information about couple relationships and family functioning (12 months after random assignment is conducted) that cannot be garnered by administrative data or other secondary data sources. These data sources will also identify mediating mechanisms that might account for short-term and longer-term effects of the intervention.

2. To describe the population being served.

Information collected about the control group in the 12-month survey will be used to augment descriptive information collected at baseline on the demographic, social, marital, and economic characteristics of the populations involved in SHM marriage education programs. In addition, the observational data collected about control group families will provide descriptive information about the quality of the families’ communication skills and interactions (couple, parent-child, and co-parenting) in the absence of the program.

3. To validate measures.

Because some of the measures of relationship quality, expectations, and values and ideals have been developed specifically for the SHM evaluation, and have not been tested and validated with low-income couples, the survey and observational data will help us to understand how different items relate to each other, as well as to validate survey items with data from the observational study.

4. To obtain contact information for subsequent waves of follow-up data.

Each study participant will be asked to complete a contact information sheet to help ensure that each individual (and couple) can be tracked throughout the follow-up period and located for subsequent data collection activities at the 12-month follow-up. About 6 and 24 months after random assignment, study participants will also be contacted by the survey firm and will be asked to update their contact information. Having access to high-quality contact information is critical for achieving high response rates during each of the follow-up waves of data collection.

The Role of Each 12-Month Data Component

The SHM 12-month data collection consists of two key components: an interviewer-administered survey of both partners in couples randomly assigned at study entry, and videotaped observations of couple, co-parenting, and parent-child interactions. Both of these components gathers unique information about participants. In the following section, we describe each of the 12-month data collection components, the information it collects, and the role that the data play in the SHM evaluation.

One of the couple’s children (the Focal Child) will also be the focus of the parenting, non-residential parental involvement, and child well-being measures of the survey, and will be asked to participate in the co-parenting and parent-child interactions as part of the observational study at the 12-month follow-up (see description of these research activities below).  In most of the SHM demonstration sites, the Focal Child will be randomly selected from all the couple’s or single parent’s eligible child who range in age from 0 to up to 15 years old at random assignment. In the two SHM demonstration sites testing curricula which include a parenting component targeting newborn infants, the child who is 3 months old or younger in the household at random assignment will be purposefully selected to participate in the 12-month observational study, the parenting measures and child well-being measures of the follow-up surveys. The Focal Child will also be targeted for the direct child assessments and/or adolescent survey collected at the 36-month follow-up.  

12-MONTH SURVEY

We propose administering the 12-month follow-up survey to both adults in all couples randomly assigned to program and control groups in all sites of the SHM study. The survey will be conducted using a mixed-mode methodology that consists of a combination of computer-assisted telephone interviews (CATI) and computer-assisted in-person interviews (CAPI). In a mixed-mode approach, the survey firm generally attempts to survey each respondent by telephone. Field interviewers then attempt to interview respondents who cannot be contacted by telephone in person. Study participants can refuse to complete the survey, or answer any of the questions on the survey, and will not be penalized in any way.

The approximate administration time for this survey is 50 minutes per person. In the design of this survey, the research team has employed several strategies to minimize the burden on respondents, such as using screening questions so that respondents will skip over sections of the survey that are inapplicable to them. The 12-month survey is attached as Appendix A. This draft of the survey has yet to be formally pre-tested.

In selecting outcome domains and their measures for inclusion on the survey instrument, the SHM research team prioritized measures based on their relevance to the SHM intervention and study population. Outcomes selected are either related to the direct objectives of the SHM intervention, such as marital relationship and child well-being, or mediators of those relationships (such as service receipt, parenting outcomes, and employment and economic well-being) (see Figure 2). In addition, whenever possible, measures were selected that have been tested and validated in prior research with low-income and racially and ethnically diverse populations. We also considered how the information gathered on the 12-month survey would complement or supplement the other data sources being proposed for the SHM evaluation in an effort to minimize the duplication. Finally, we drew heavily upon items that were used in the SHM baseline instrument (OMB No. 0970-0299), so that we could identify changes in responses over time when these could lend insights into the impacts observed, and the Building Strong Families (BSF) 15-month survey (OMB No. 0970-0304) to maximize outcomes that can be examined in common across projects evaluating the effectiveness of marriage education programs.

Figure 2.

Conceptual Framework for Supporting Healthy Marriage Project

“Healthy Marriage”

We propose including the following topics in the survey instrument:

Household information. We propose asking all respondents to describe their household structure so that we can estimate the impacts of the SHM intervention on residential instability (via marital stability) and poverty status. These measures include the number of children and adults in the household, the respondent’s current living arrangements, and whether the respondent experienced residential instability, such as the number of times the respondent has moved since random assignment. The extent to which SHM services affect these aspects of household composition and residential instability hinges on whether the program (1) encourages couples to stay together or separate and/or (2) affects couples’ financial circumstances over the follow-up period. For example, it could be that families who are receiving SHM services achieve more stable financial circumstances, and in turn more stable housing and family arrangements than those in the control group. In particular, residential instability in children’s lives places them at risk for less optimal developmental outcomes (Moore & Vandivere, 2005; Stoneman, Brody, Churchill, and Winn, 1999). Finally, information collected about children living in the household at follow-up will be used to verify that the Focal Child who was randomly selected from children who were living with the couple at random assignment, is still present.

Marital status and stability. A key question is how the SHM intervention affected divorce and marriage rates, as well as the duration and number of spells of separation. For this reason, marital status and marital history are among the most important outcomes to measure carefully as part of the SHM evaluation. We will need to gather information about the number of times the couple separated or lived apart over the follow-up period, how long these spells apart lasted, and the reasons why the couple separated. We propose including measures of whether the marital relationship has been in trouble and the couple has thought about separation or divorce as additional indicators of marital stability and quality. We also propose including measures about whether the respondent is involved with a new partner and the status of that relationship. Lastly, we propose including two items that assess how many times the respondent had been married prior to random assignment, as prior research indicates that individuals who have previously married and divorced are at a higher risk of seeing their remarriage dissolve compared to individuals in first marriages (Coleman, Ganong and Fine 2000). The measures in this section of the survey are taken from several large-scale evaluations and national longitudinal datasets such as the Building Strong Families Evaluation (OMB No: 0970-0304), the Hard-to-Employ Evaluation and Demonstration Project (OMB No: 233-01-0012), and the Fragile Families and Child Well-Being Study.

Marital interactions and relationship appraisals. We propose asking respondents who are still married at the time of follow-up questions aimed at assessing the marital interactions and their own appraisals of the quality of the relationship with their spouse. A subset of these questions will also be administered to respondents who are separated, but still have contact with their former spouses, at the time of follow-up. The items asked of separated couples who still have some contact with each other is limited to the constructs of communication, problem solving and conflict resolution, time spent together with the children, and commitment to children, which are most likely to be affected by the SHM intervention and likely to have implications for their children’s well-being.  Though it is conceivable that other dimensions of the quality of the relationship could be affected post-separation as well, it could be somewhat awkward or frustrating for many separated individuals to answer such questions if the post-separation relationship is somewhat contentious or interaction is limited. We have therefore chosen to limit the scope of the questions asked of separated couples. These measures are listed separately by construct of interest in Appendix B.

Assessing interactions and quality of the relationships is of central importance to the SHM impact analyses. Not only is it important for the well-being of parents themselves (McLanahan & Sandefur, 1994; Cowan & Cowan, 2006), but it also may affect their children’s well-being (Amato, 2000; McLanahan & Sandefur, 1994). Research has shown that the quality of the parents’ relationship has important direct effects on children (Cummings & Davies, 1994; Emery, 1982; Grych & Fincham, 2001). In addition, relationship quality is highly correlated with the likelihood that the couple will stay together (McLanahan & Sandefur, 1994).

Prior research suggests that any measure of relationship quality and interactions should cover multiple domains (Moore et al., 2007), yet there is little empirical evidence about which dimensions are the most critical elements of a high quality couple relationship. In the current study, the dimensions of marital interactions and quality that are included in the survey were selected based on: (1) an extensive review of published research, other surveys, and perspectives of leading researchers completed by Child Trends (Moore et al., 2007; Moore et al., 2005); (2) the aspects of relationship quality that are particularly targeted by the SHM program model; and (3) dimensions of marital quality that may be of particular cultural relevance for the low-income and racially and ethnically diverse sample in the SHM evaluation. Accordingly, the survey taps multiple dimensions of relationship quality, each of which finds support in prior research (see Appendix A). Furthermore, under each of these constructs, we propose including measures that not only tap respondents’ own feelings and behaviors, but also include measures assessing the respondents’ perceptions of their spouses’ behaviors and feelings about the couple relationship. This strategy is used because prior research suggests that individuals’ perceptual biases, especially the ways in which spouses selectively attend to the positive or negative aspects of the marriage and their partners’ behaviors, can have important implications for global assessments of marital satisfaction and stability of marital relationships (Carrere, Buehlman, Gottman, Coan & Ruckstuhl, 2000; Fincham et al., 1990; Baucom et al., 1989).

To the extent possible, measures of relationship quality have been drawn from scales that have been shown to have strong psychometric properties. Many of the measures were taken directly or adapted from scales that appear in national studies or studies that included low-income and racially and ethnically diverse respondents, such as the BSF 15-month follow-up survey, the Fragile Families and Child Well-Being Study, the National Study of Families and Households, the National Evaluation of Early Head Start, and The Early Childhood Longitudinal Study – Birth Cohort. Ideally, we would rely exclusively on well-validated existing scales in our measures of relationship quality. However, while some scales, such as domestic violence measures, have been tested and validated with low-income and racially and ethnically diverse populations, few such measures exist for many of the remaining dimensions of relationship quality that we propose examining in the SHM survey instruments. Therefore, given that some of our measures are breaking new ground, the research team has conducted a series of one-on-one interviews with low-income and racially and ethnically diverse couples drawn from Oklahoma, Texas and Washington, DC. These tests (each conducted with nine or fewer participants) have been used to refine the survey measures and to understand the extent to which the selected measures are culturally relevant and tap the intended constructs of interest.³ After we have collected the survey data, we will conduct psychometric analyses to see which relationship quality measures are correlated with each other and can be used together in scales.

• Communication. We define communication to be a set of specific skills that have been the focus of a great deal of basic research and relationship intervention; namely, the degree to which one clearly expresses one’s ideas and needs to one’s partner and, in turn, one listens to and demonstrates understanding of one’s partner. SHM program services focus on improving communication between spouses as a way to deter conflicts that lead to unhealthy relationships, and the survey will allow us to understand if the intervention improved these communication patterns between spouses. Communication items will be asked of both intact and separated respondents who are still in contact. These measures are important because previous research has shown that couples’ communication is predictive of relationship dissolution, divorce or separation (Silliman et al., 2002). Items have been drawn from the SHM baseline instruments, the BSF 15-month survey, and the Family Adaptability and Cohesion Evaluation Scales II (Cowan & Cowan, 1992), and were developed by the SHM team through cognitive testing.

• Disagreement and Conflict Resolution. We propose including a series of questions about the frequency with which couples argue, the degree of hostility associated with couples’ conflict, and the couples’ conflict resolution processes and successes in resolving conflicts and disagreements. Prior findings suggest that poorly managed conflict is highly predictive of relationship dissolution (Stanley, 2003), has been shown to be harmful to children (Cummings et al., 1991), and is associated with poorer physical and mental health among the individuals involved (Fincham, 2003). Furthermore, we hypothesize that the SHM programs could have direct improvements for the ways in which couples resolve disagreements and reduce the extent to which disagreements involve escalating negative exchanges (Bradbury, 2002; Gottman, 1994) by exposing couples to less destructive strategies for managing conflict and improved communication skills. Finally, because couples are able to handle conflict in more productive ways, SHM programs may also reduce the frequency with which couples argue. Measures regarding these topics were drawn from a variety of sources, including the SHM baseline instrument, BSF 15-month survey, the National Evaluation of Early Head Start, the National Survey of Families and Households, the Early Childhood Longitudinal Study – Birth Cohort (ECLS-B), the Couple Communication Questionnaire, the Gottman Sound Relationship House Questionnaires, and developed by the SHM team through cognitive testing.

• Violence. We propose including a series of measures assessing aspects of psychological violence (the extent to which partners’ actions intend to cause psychological suffering or pain to the other partner) and physical abuse and violence in couple relationships for a number of reasons. The prior literature suggests that domestic violence is associated with poor mental and physical outcomes for the victim, a higher likelihood of the dissolution of the relationship, and poor outcomes for children, especially if they are exposed to the violence (Lawrence, 2002). Second, prior ethnographic research, as well as cognitive interviews conducted by the SHM team, suggests that physical violence, at least in relatively mild forms such as pushing and shoving, seem to be fairly common. This qualitative evidence suggests that it will be important to assess types of violence that range from pushing, shoving and yelling to more serious and dangerous forms of physical violence and abuse, as well as the frequency of such acts. Third, some of the proposed measures of physical violence will be asked of both spouses about the actions of his or her spouse. Several studies have found that men and women initiate violence against an intimate partner at approximately the same rate. For example, in the National Family Violence Survey, both men and women reported that violence was initiated by each partner at least 40% of the time (Stith, Smith, Penn, Ward & Tritt, 2004). Measures about psychological abuse and physical violence were drawn from the Conflicts Tactics Scale, the SHM baseline instrument, and the Psychological Maltreatment of Women Inventory.

• Time in Shared Experiences/Interaction. This proposed construct is primarily intended to capture the “quantity” dimension of couples’ interactions. The theoretical interest in this construct is in the extent to which couples share experiences—including time spent interacting with one another—that could provide opportunities for relationships to deepen and grow (as opposed to simply spending time in the same room together, perhaps doing different things). Sub-constructs distinguish types of shared activities and with whom the couple shares activities. Fincham (2003) and Hawkins et al. (2006) argue that more research attention be given to more positive aspects of relationships, such as the amount of time the couple spends together, particularly time spent together alone, which has been shown to be related to relationship stability (Shapiro et al., 2000). Furthermore, the SHM curricula encourage couples to spend time together, so it is important to capture this dimension of couple relationships. Items have been drawn from the SHM baseline, BSF, the National Survey of Families and Households, and developed by the SHM team through cognitive testing.

• Fidelity. Being faithful to a spouse includes avoiding sexual relationships outside of the marriage and also intense romantic and emotional attachments outside of the marriage that compete with the marriage. Infidelity is the most commonly reported reason for relationship breakup (Smock & Manning, 2004). We therefore propose including two items tapping whether the respondent had been unfaithful to his or her spouse and whether the respondent believes that his or her spouse had been unfaithful in the past three months. These items are drawn from the BSF 15-month survey.

• Intimacy. Research shows that intimacy is correlated with feeling understood as a person, validated, and cared for by your partner, as well as good marital outcomes (Gable, Gonzaga & Strachman, 2006; Bradbury et al., 2000; Huston & Chorost, 1994). In addition, prior studies suggest that expressions of support and affection can increase marital satisfaction by creating a reservoir of positive feelings towards a spouse that assuages the potential ill effects of negative communication patterns (Reis-Shaver, 1998). Indeed, the SHM curricula focus on building interpersonal skills related to intimacy and social support, like supporting one's partner or dealing with a partner’s stress. Therefore, we propose including a series of measures aimed at understanding the level of intimacy in couple relationships. Researchers refer to non-contingent positive affect and physical intimacy as two distinct types of couple intimacy. Non-contingent positive affect refers to support and mutual understanding, closeness, and shared goals (Gottman et al., 1998). In addition, relationship-enhancing attributions contribute to intimacy, such that the partner’s behaviors and motives are seen as positive, dependable, and trustworthy (Kurdek, 1998). Finally, recent research emphasizes the importance of couples developing a deeper understanding of each others’ hopes, dreams and fears, and of developing a shared worldview (Kurdek, 1998; Sprecter et al., 1995). Items have been drawn from the SHM baseline, BSF, the Enrich and Nurturing Relationship Issues, Communication, and Happiness inventory, the Index of Sexual Satisfaction (Walker & Thompson, 1983), and developed by the SHM team through cognitive testing.

• Satisfaction and Dissatisfaction. Marital satisfaction is one of the most widely studied dimensions of marital relationship quality, with the literature demonstrating that relationship happiness is highly predictive of later divorce or separation among married couples (Karney & Bradbury, 1995). We propose including a series of measures of marital satisfaction that have a long-standing history in the literature and have been validated and tested in numerous studies. The first item asks respondents to assess how happy they are with their marriage (adapted from Bradbury, Fincham & Beach, 2000; Clements, Stanley & Markman, 2004). This measure of marital satisfaction is also used in the BSF 15-month survey. The second series of marital satisfaction items asks respondents to rate the extent to which they are satisfied with various dimensions of the couple relationship such as communication, conflict resolution, time spent together in shared activities, emotional supportiveness, commitment to children, and their sexual relationship, as well as other common problems such as division of household labor and finances. Items have also been drawn from the Relationship Evaluation Study, the Enrich and Nurturing Relationship Issues, Communication, and Happiness inventory, the SHM baseline, the BSF 15-month follow-up survey, and developed by the SHM team through cognitive testing.

• Commitment to the Couple. Psychologists have found that commitment to the marriage is an important mediator of both marital satisfaction and stability (Amato & Rogers, 1999; Rusbult & Buunk, 1993; Stanley et al, 1999; Stanley & Markman, 1992; Van Lange et al., 1997; Whitton, Stanley & Markman, 2002). Confidence refers to the degree to which both partners believe that they will be able to handle challenges facing their relationship. Commitment refers to the degree to which they are willing to persevere and make sacrifices for the relationship. A primary target of the SHM curricula is fostering a greater commitment and confidence in marital relationships, and strengthening levels of respect between spouses. Therefore, we propose including a series of measures aimed at tapping these constructs. Measures here have been drawn and adapted from the SHM baseline instrument, the Spouse Treatment Mediation Inventories, the Relationship Rating Form, the Dyadic Trust Scale, the Commitment Inventory, and developed by the SHM team through cognitive testing.

• Joint Commitment to Children and Extended Family. Distinct from co-parenting skills, this element of a healthy marriage encompasses a long-term and joint commitment to caring for any children that are being raised in this relationship. Measures of commitment to extended family have also been included due to their cultural relevance with the black and substantial Hispanic population served in SHM sites. Items in this section have been drawn from the BSF 15-month follow-up survey and were further developed by the SHM team through cognitive testing.

Marital Insights, Values, Expectations, and Beliefs. A key hypothesis in the SHM logic model is that marriage education programs will affect couples’ attitudes towards, and understanding of, marriage in ways that lead to improvements in marital interaction and satisfaction. Our distinction between attitudes and beliefs is similar to Baucom et al.’s (1989; 1996; 2002) distinction between “standards,” or “beliefs about what relationships… should be like,” and “assumptions,” or “beliefs about the ways relationships actually operate, as well as what men, women and one’s partner are like.”

• Attitudes and Values. We propose including items that assess the degree to which couples see marriage as a long-term proposition. Couples who see marriage in a long-term sense and who have unfavorable views of divorce fare better in the long term (Thomson & Colella, 1992; Amato & Rogers, 1997; Moors, 2000). Diverging trends in attitudes towards divorce by social class mirror diverging trends in divorce rates (Martin & Parashar, 2003). SHM programs encourage couples to see marriage as a long-term commitment and varyingly include material on the costs of divorce.

We also propose including items that tap wives’ and husbands’ support for gender equality in household roles such as work, childrearing, household chores, decision making, and communication. Research suggests that egalitarian values are associated with greater marital satisfaction and stability, especially for wives and when the work is actually shared (Fein et al., 2007; Frisco & Williams, 2003; Hochschild, 1989; Gerson, 1993; Greenstein, 1996; Perry-Jenkins & Crouter, 1990). Although SHM programs do not prescribe gender equality, they generally assume that spouses should see themselves and each other as partners who will work through differences in views of appropriate family roles and responsibilities. This may therefore result in some couples exploring and perhaps redefining their roles.

• Knowledge and Beliefs. We propose including a set of items that assess knowledge of the key precepts most marriage education programs address. These precepts include the ideas that people and relationships can change, that good relationships take work and willingness to sacrifice, that communication is important both when things are going well and not so well, and that negative reciprocity and violence are toxic. We cite some empirical research documenting the importance of these elements in the section of this document on marriage outcomes. A separate literature has stressed the importance of beliefs about relationships as influences on behavior (e.g., Baucomb et al., 1989, 1996, 2002). Because the psycho-educational approaches tested in the SHM demonstration seek to impart knowledge, as well as skills, it is important to test their understanding of these precepts. A different type of belief is general mistrust by women that men will be sexually faithful. So-called “gender mistrust” has been found to be an important influence on unmarried, largely African-American couples’ relationships (Carlson et al., 2004). SHM provides an important opportunity to learn whether this belief is prevalent also among low-income married couples, and whether efforts to build trust within participating couples leads to more trusting views generally.

The final set of questions in this module measure the degree to which SHM affects the reasons people see for marriage. Marriage sociologists believe that changing perceptions of the purposes – or meanings – of marriage are one important force underlying long-term increases in marital stability. Trends suggest that the strengthening of a consumerist culture has strengthened the value attached to individual happiness, leading to increased expectations for marital fulfillment (Fowers, 2004), at the same time other normative supports for marriage have weakened (Thornton & DeYoung, 2001; Pew, 2007). In particular, the increased acceptability of non-marital cohabitation and weakened link between childbearing/raising and marriage have eroded two important normative rationales for marriage. Low-income couples are especially likely to see financial circumstances as linked to the meaning of marriage (Tucker, 2000). On the one hand, the Fragile Families literature indicates that low-income couples see a certain level of financial success as essential for marriage to be acceptable (Edin, 2000; Edin & Kefalas, 2005). On the other hand, the economic arguments for marriage (specialization, economies of scale) are thought to be a long-standing perceived “purpose” of the institution, and one would think that such a purpose might be especially important for low-income couples near the margins of self-sufficiency.

Values such as these occupy an important place in the SHM conceptual framework, as potential influences on marital interaction and commitment that marriage education may affect. Marriage education curricula encourage couples to explore what marriage means to them and may thereby lead to reassessments of its perceived purposes. It is not possible to anticipate exactly which perceived purposes will be affected and how such changes might influence relationship quality. Because emphases vary somewhat across curricula, individual curricula address multiple purposes (e.g., encourage communication of individual needs but also the importance of making sacrifices for the sake of one’s spouse or children) and the effects of changes in values are likely to be contingent on changes in skills and behaviors. (For example, an increased pre-occupation with personal fulfillment might be beneficial for partners who also learn how to be more emotionally supportive of one another, but increase dissatisfaction for those whose interaction does not improve.) Notwithstanding this complexity and ambiguity, it is important to measure the meanings couples invest in marriage precisely so that the demonstration has the capacity to measure impacts on meanings and trace the connections to marriage outcomes.

Receipt of Marriage Education Services and Other Services. The survey will be administered to all participants in the program and control groups. As such, the survey will be a key source of information about control group members’ receipt of services in the full-scale SHM evaluation and serves as a key source of information for the implementation and impact analyses. We propose including questions in the survey aimed at assessing receipt of marriage education and counseling services, as well as participation in other types of services available in the community, such as job training, welfare, housing and child care assistance, and parenting education services. These measures are drawn from the SHM control group services survey (OMB No: 0970-0330) which is used during the pilot phase of the SHM project to assess the services differential between program and control groups, the BSF 15-month follow-up survey (OMB No: 0970-0304), and the Employment, Retention and Advancement Project (OMB No: 0970-0242). For respondents randomly assigned to the program group, information about participation in SHM services will also be gathered through the program’s MIS.

Co-Parenting. We propose a series of items aimed at assessing the co-parenting relationship between the respondent and spouses/former spouses. These measures will be administered to both partners in intact and separated couples who have some contact with each other at the 12-month follow-up. They will assess several key dimensions of co-parenting, such as the extent to which parents support and show mutual respect for each other in their childrearing, the degree to which parents are able to effectively communicate and problem solve with each other when disagreements about childrearing occur, and the extent to which each parent is involved in key aspects of raising children and running a household (e.g., cooking, household chores, running errands, managing household finances, and caring for children). These measures are drawn from the BSF 15-month follow-up survey, the Fragile Families and Child Well-Being Surveys, and adapted from the ECLS-B questionnaires, which included low-income and racially and ethnically diverse populations.

A growing body of research has found links between marital quality and aspects of co-parenting (Katz & Low, 2004), suggesting that the development and sustenance of a healthy couple relationship is likely to promote improved co-parenting. In addition, evidence indicates that interventions aimed at improving marital relationship quality have the potential to spill over into both parenting and co-parenting domains, since adults’ relationships with partners consistently influence relationships with their children (Cowan & Cowan, 1987; Erel & Burman, 1995; Florsheim, Moore, Zollinger, MacDonald, & Sumida, 1999; Lindahl & Malik, 1999). By educating couples about positive modes of communication and problem solving, marriage education programs have the opportunity to improve co-parenting, and ultimately, outcomes for children.

Parenting and Family Functioning Outcomes. We propose including several measures of self-reported parenting behaviors and key aspects of family functioning, such as family emotional climate, the use of warm parenting behaviors and harsh discipline techniques, parental involvement and engagement, regularity of family routines, and parental stress and aggravation. Items are asked with regard to the couple’s Focal Child, who will be selected at random from the children between 0 and under 15 years old who were present in the household at baseline. Items in the survey have been selected for their age-appropriateness and will vary depending upon the age of the Focal Child at the point of follow-up. The proposed items have been drawn from prior studies, which included low-income and racially and ethnically diverse populations, including the ECLS-B, the Fragile Families and Child Well-Being Study, the PSID-CDS-II, the Minnesota Family Investment Project, and have been adapted from the Moos Family Environment Scale (Moos & Moos, 1994) and Family Expressiveness Scale (Halberstadt, 1983). Both husbands and wives in intact and separated couples at the 12-month follow-up will be asked the parenting and family functioning measures, based upon literature which suggests that the maintenance of a healthy marriage increases the probability that fathers will have a direct and positive relationship with their children (McBride & Rane, 2001). Furthermore, by asking both parents such questions, the current study has the potential to make an important contribution to the literature about the extent to which fathers’ parenting practices influence child well-being, because this question has been relatively understudied, particularly among low-income fathers, to date (Cabrera et al., 2004).

According to the “spillover hypothesis,” programs that attempt to reduce the level of conflict in and increase the quality of the couple relationship may positively affect the family climate and parent-child relationship (Erel & Burman, 1995; Haberstadt, 1983). For example, prior research has documented that higher levels of marital conflict are linked with more frequent use of harsh discipline, less parental warmth, and less parental involvement and engagement with children (Buehler & Gerard, 2002; White 1999; Carlson et al., 2006). Marital distress can also lead parents to have high levels of overall stress and parenting stress in particular, both of which have been found to be related to internalizing (withdrawn, depressed) and externalizing (aggressive, impulsive) behavior problems among children living in low-income families (Conger et al., 2002; McLoyd, 1990). SHM programs could reduce parental stress through improvements in the quality of couple relationships and the provisions of family support services and referrals to other services for stressors that families might face. Lastly, marital distress has also been shown to lead parents to interact less consistently with their children (e.g., due to depression) or to be less effective in coordinating their time spent with their children (Hops, 1992; McHale & Cowan, 1996; Lindahl et al., 1997). Thus, SHM programs could have indirect effects on the regularity of family routines, which has been linked with child well-being (Edin & Lein, 1997).

Non-Residential Parental Involvement. In addition to the parenting and co-parenting measures that will be asked of all parents, residential and non-residential, we propose to include several measures of non-residential parental involvement (namely, the frequency with which non-resident parents interact and see their children and the extent to which non-resident parents provide child support). Items will be asked regarding the Focal Child. The proposed items have been drawn from a variety of sources including the 2002 National Survey of Family Growth, the Fragile Families and Child Well-Being Study, and the ECLS-B.

Even if an SHM couple separates or divorces, a program that helps parents improve their relationship or maintain civil relationships with their partners after divorce or separation could enable them to remain involved in the lives of their children (Carlson & McLanahan, 2006). Thus, it could be that marriage education programs help improve the quality of the co-parental relationship, such that non-residential parents are able to maintain higher levels of involvement and child support payment compliance even if the marriage ends (Tishler et al., 2003; Emery et al., 2001; Bronte-Tinkew, Carrano, & Guzman, 2006; Minton & Pasley, 1996).

Child Outcomes. The ultimate goals of SHM are the support of healthy marriages and the improvement of child well-being. In addition to effects on parenting, healthy marriages might directly benefit children by exposing them to good models for healthy relationships, increasing their likelihood of living with both parents, increasing family income, and reducing exposure to parental conflict – one of the clearest risk factors for less favorable child outcomes (Cummings & Davies, 1994; Emery, 1982; Morrison & Coiro, 1999; Hetherington et al., 1992). The proposed survey instrument includes a short battery of parent reports on how all children in the family are doing. These questions are primarily focused on social, emotional and behavioral outcomes, as prior research suggests that the short-term associations of divorce and high parental conflict with child well-being have been most consistently observed in these domains (Gottman & Katz, 1989). These parent-reported measures of child well-being will be administered to all participants in the SHM evaluation. This parent-reported information on child well-being will be supplemented with direct child assessments and an adolescent survey that will be collected about the Focal Child at the 36-month follow-up. Together, these data will allow the research team to understand the program’s impacts on children’s emotional and behavioral adjustment, as well as cognitive development and academic skills, as prior research suggests that these domains of children’s functioning are related (Alexander, Entwisle & Dauber, 1993).

Individual Adult Well-Being. We propose including a series of measures on the 12-month survey aimed at assessing respondents’ perceived stress, mental health, physical health, and alcohol and substance abuse. These items will be administered to all participants in the SHM evaluation.

• Mental health and stress. The literature suggests that there are a number of reasons why we should include measures of mental health and stress in the 12-month survey. Poor mental health can substantially impair relationship quality and stability (Fein et al., 2007; Karney & Bradbury, 2005), and undermine the development of children (Hair, McPhee, Moore & Vandivere, 2005); Moore et al., 2006). At the same time, SHM may affect mental health in several ways: SHM could improve participants’ understanding of, and skills for providing emotional and practical support in response to the difficulties that their partners face. Family support coordinators will also provide couples with referrals to address some of the stressful situations and living conditions (e.g., financial hardships, crowded and noisy living environments) that they might face. Items tapping individual perceived stress are drawn from the Perceived Stress Scale, which is a self-reported global measure of perceived stress (Cohen & Williamson, 1983). Items tapping individual mental health are drawn from the Kessler 6-item Psychological Distress Scale (K-6), a measure of generalized distress, that is used in the U.S. National Health Interview Survey (NHIS) and the National Household Survey on Drug Use (Kessler et al., 2003).

• Physical health. Effects of the SHM program on physical health outcomes are likely to be indirect and would most likely be a consequence of program-induced improvements in the quality of couple relationships and marital outcomes (Waite & Gallagher, 2000; Ribar, 2004). At the same time, it is possible that family support coordinators could help some families obtain health insurance, which could also affect health outcomes for program group members. Therefore, the survey instrument includes a single item tapping respondents’ self-reported physical health recommended by the National Health Interview Survey, which was also included in the SHM baseline instruments. It also includes a single item about whether or not the respondent and anyone in his or her family is covered by health insurance.

• Alcohol and substance use. A number of studies have also documented an association between marriage and substance abuse. Substance-abusing couples are more likely to experience divorce (Yamaguchi & Kandel, 1985). In addition to a substance user’s behavior having an impact on his or her spouse, the responses of the non-abusing spouse can also influence the substance user, e.g., “enabling” or “codependent” behaviors (Rotunda & Doman, 2001; Yoshioka, Thomas, & Ager, 1992). When substance abuse is present in a marriage, there are often various adverse effects on relationship, family, and parental functioning (Rotunda & Doman, 2001; Roosa & Tein, 1993). Therefore, the survey includes a short set of measures intended to tap self-reported alcohol and substance use. We also propose including a single measure about a respondent’s perceptions of his or her (former) spouse’s alcohol or drug use. These items are based on recommendations of the National Institute on Alcohol Abuse and Alcoholism (Taj et al., 1998) and the National Institute on Drug Abuse, and were also used in the SHM baseline instruments.

Employment, Income, Material Hardship, and Perceived Financial Strain. A number of studies suggest that couples’ financial circumstances and financial strain may be related to the quality of couple relationship, marital satisfaction and stability. For example, studies show that married adults have better financial outcomes than unmarried adults (Waite & Gallagher, 2000), that couples from disadvantaged backgrounds are more likely to separate and divorce after marrying (Ellwood & Jencks, 2001; Bramlett & Mosher, 2002), and that marital break-up is associated with substantial increases in poverty and economic stress (Spain & Bianchi, 1996). Thus, the SHM program could have indirect effects on families’ economic circumstances through impacts on marital stability, and could directly help to improve couples’ financial circumstances by connecting them with employment, education, and training opportunities and services, as well as forms of governmental assistance (e.g., Food Stamps and TANF) and income supports. Therefore, the survey instruments include several measures aimed at capturing families’ total income, husbands’ and wives’ employment behaviors, and material hardship (e.g., measures of the extent to which the families’ basic needs, like food, shelter and health care, are unmet). Perceived financial strain (or individuals’ assessments of their financial situations and the degree to which they are concerned and worried about these circumstances) has been linked with marital instability through individual emotional distress, the occurrence of couple disagreements, and quality time couples spend together (Gudmunson, Beutler, Israelen, McCoy, & Hill, 2007), and has also been linked with less favorable child outcomes (Conger et al., 1992; Conger, Conger & Elder, 1997). Furthermore, prior research suggests that individual employment experiences such as work hours and schedules can be linked with marital instability and quality of marital relationship (Presser, 2000). These items were drawn from a variety of sources including the BSF 15-month follow-up survey, the New Hope Project, and the Minnesota Family Investment Program.

Social Support Outcomes. A number of studies indicate that social support provided to couples can ease strain or buffer conflict and therefore, can influence how the couple responds to marriage education (House, Umberson, & Landis, 1988; Umberson, Chen, House, Hopkins & Slaten, 1996; Wheaton, 1985; Amato, 2004). Studies also show that couples can internalize positive or negative attitudes about marriage and partners depending on their social network (Oliker, 1989, as cited in Bryant & Conger, 1999). For instance, individuals who perceive support for marriage or disapproval of divorce from their friends and families may adopt similar positions, or their behavior may be influenced. Elsewhere, studies of dating couples further suggest that approval or disapproval from social networks can shape the outcome of the relationship (Bryant & Conger, 1999). Levels of perceived support for a couple’s marital relationship have been associated with better relationship quality and with greater marital success two years later, independent of earlier levels of marital success (Sprecher & Felmlee, 1992; Bryant & Conger, 1999) as well.

The SHM program might have direct effects on the availability of social support and the size of participating couples’ social networks by bringing couples together in the marriage education workshops. It could be that couples become friends and sources of social support for each other in group classes. Furthermore, because program participants will be exposed to more positive communication techniques, such couples may be better able to get along with others and maintain friendships outside of the SHM program activities. The curricula also directly encourage couples to think about and improve their social networks.

For these reasons, we propose including a series of measures aimed at assessing whether respondents have sources of instrumental and emotional support available in their social networks. These items are drawn from the SHM baseline instruments, the BSF 15-month follow-up survey, and the Chapin Hall Community Partnerships for Protecting Children. We also propose including several measures about social network size and level of community involvement, as prior research suggests that individuals who are more connected to others and more involved in the community may be better able to cope with stressors in their lives (Bryant & Conger, 1999). These items tapping social networks and community involvement are drawn from the Social Capital Community Benchmark Survey and the Pennsylvania State Marital Instability Study. Finally, we propose including several questions aimed at understanding the extent to which families and extended family members interfere with the couple’s relationship. These items were adapted from the SHM baseline instruments.

Demographic information. The 12-month follow-up survey contains several items assessing basic demographic information that were not assessed at baseline about each member of the couple.  We propose including two questions assessing the country of origin of immigrants to the U.S. In addition, we propose including two items aimed at assessing the respondents’ cultural values and acculturation with American culture.  Identification with particular cultural norms, attitudes and expectations could influence the effectiveness, as well as take up rates of a marriage education program. For example, cultural values are likely highly correlated with one’s expectations and attitudes about marriage, as well as views on what constitutes acceptable modes of couple and family communication patterns; such values could either contradict or support the style, format or delivery of the marriage education services being tested in SHM.

Contact Information Sheet. At the end of the survey, respondents will be asked to complete a contact information sheet to help ensure that they can be tracked during the follow-up period for the 36-month follow-up data collection effort. Contact information will be sought for three close relatives and/or friends, including their names, addresses, phone numbers and their relationships to the respondent. In addition, approximately 6 months after random assignment, the survey firm will send study participants a postcard asking to update their contact information to help ensure that they can be tracked at the 12-month follow-up point. Having access to high quality contact information is critical for achieving high response rates during each of the follow-up surveys.

12-MONTH OBSERVATIONAL STUDY

The SHM team proposes to conduct videotaped observations of couple, parent-child, and/or co-parenting interactions with a random subsample of participants in both the program and control groups at the 12-month follow-up. These observations will provide critical information about the quality and nature of couple interactions, as well as parenting behaviors. The protocol for conducting these videotaped observations can be found in Appendix C.

Having both self-reported measures and independent observations of couple, co-parenting, and parent-child interactions will enhance our ability to appropriately measure these constructs of interest by understanding how the SHM program affects different aspects of couple and parent-child relationship quality. First, recent findings from several experimental evaluations of marriage education group sessions have been found to have positive impacts on couple interactions and parent-child relationships that were assessed using observational methodologies. These impacts, however, were not always apparent in self-reported data of couple and parent-child interactions, suggesting that it is important to have both sources of measurement when evaluating these kinds of programs (Cowan et al., in press).

Second, independent observations can provide more objective measures of the nature of these types of interactions because self-reports may be biased depending upon the individual’s own characteristics, such as stress and financial well-being. Furthermore, some theoretical constructs in couple and parent-child interactions that involve contingent connections between one person’s actions with another person’s actions can only be fruitfully assessed with observations (e.g., Patterson’s concept of coercive cycles; Dishion, Patterson & Kavanagh, 1992).

While there is a strong argument for capturing videotaped observations of couple and parent-child interactions, it is also critical to capture both observational data and self-reported data. Self-reports may result in better or more complete information about how a couple and family members interact with each other than a videotaped observation, since the set up of these interactions can be somewhat unnatural. Moreover, self-reported data can capture how the couple and family members interact with each other across different contexts and time, which is not as easily captured by a brief snapshot of couple and family functioning taken at a specific point in time, and self-report data illuminate subjective perceptions, which cannot be observed.

Sample selection and sample size for the 12-month observational study. The research team would like to collect videotaped observational data in all eight SHM sites with a subset of couples with children aged 0 to under 15 at random assignment in the program and control groups at the 12-month follow-up as described earlier. We propose targeting 306⁴ couples with children in this age range in each site to participate in the observational study at the 12-month follow-up, resulting in a respondent sample of approximately 220 couples in each site, which is approximately one-third of the survey respondent sample. We propose conducting videotaped observations of couple, parent-child and co-parenting interactions in the homes of SHM participants. One of the couple’s children (the Focal Child) will also be asked to participate in the co-parenting and parent-child interactions. This child will be randomly selected from all of the eligible children in the family (for whom the parents have given initial permission at baseline). In the cases of couples that are intact at the 12-month follow-up, the interviewer will ask the primary parent (self-identified at the time the observations are conducted) to interact with the Focal Child, and ask the couple to engage in a series of couple interactions⁵ with and without the Focal Child. In the case of couples who are separated at the 12-month follow-up, the research team will be asking only one of the spouses to participate in the observational study because of the impracticalities of bringing separated spouses together. Here, the primary custodial parent of the Focal Child will be asked to interact with this child alone.

Description of semi-structured interactions. Below, we provide a brief overview of these planned interactions:

Couple interactions. For the observational study, couples who are still together will be asked to engage in three semi-structured 7-minute discussions while being videotaped, so that the research team can learn about how couples interact and communicate with one another. These discussions are designed to capture both positive and negative aspects of couple communication and interaction patterns and have been used extensively in prior research. Prior studies also suggest that a high proportion of couples participating in these types of videotaped observations report having positive experiences and commonly report having a greater awareness and appreciation for their relationships due to the exercise (Bradbury, 1994). These semi-structured interactions have been used with low-income and racially and ethnically diverse populations in prior research (Conger et al., 1990; Conger & Conger, 1992; Lindahl & Malik, 1999; Johns, Newcomb & Bradbury, 2007; Cutrona et al., 2003; Sullivan, et al., 1998;). The protocols for the collection of the videotaped interactions have been drawn from a variety of sources including the NICHD Study of Early Child Care and Youth Development, the Family Transitions Project, and the National Evaluation of Early Head Start.

The discussion topics include the following:

• Problem solving interactions. The couple will be asked to identify a topic of disagreement and then will be asked to discuss this topic for 7 minutes.

• Social support interactions. Husbands and wives will be asked to separately identify something about themselves that they would like to change (e.g., work habits, career, something about his/her personality or appearance, some problem with friendships or relationships within their family – that is, anything that is outside of the marital relationship). The couple will then be asked to discuss one topic identified by the husband for 7 minutes, followed by a discussion of the topic identified by the wife for 7 minutes.

There are a number of reasons why we propose structuring the couple interaction tasks in this way. First, studies have shown that mismanaged conflict in a relationship is likely to take a toll on peoples’ marriages and their children, and that it has predictive power in relationship outcomes (such as divorce) (Stanley, 2003). It is for this reason that SHM curricula focus heavily on building effective communication skills, particularly in conflict situations.

At the same time, while marital conflict is a risk factor for a variety of negative marital outcomes (low marital satisfaction, divorce, etc.), problem-solving is not the only skill that matters in predicting relationship outcomes. Studies have found that conflict can be especially detrimental to marriages when levels of supportive behaviors are also low (Pasch & Bradbury, 1998). Indeed, the SHM curricula also focus on building interpersonal skills related to intimacy and social support, like supporting one's partner or dealing with a partner’s stress. While these behaviors can be coded during a problem solving interaction, the meaning and significance of supportive and warm behaviors for other marital outcomes can be quite different depending upon the scenario used to elicit such behaviors (Melby et al., 2004). Thus, to reflect the current literature on marital quality, it is important to capture not only conflict resolution communication patterns, but to also assess levels of positive and supportive behaviors among couples in the SHM program and control groups.

Co-parenting and parenting interactions. Other elements of the observational study will assess co-parenting (the extent to which the couple shows supportive or negative communication patterns with each other when interacting with a child) and parenting behaviors (e.g., warmth, coercive, and harsh parenting techniques), as prior research suggests that co-parenting and parent-child interactions may be key mediators of the effects of marriage education programs aimed at improving couple interactions on child well-being (Cowan, Cowan & Kerig, 1993).

Semi-structured tasks lasting a total of 20 minutes will be used to assess parenting and co-parenting behaviors among couples who are together at the time of the 12-month follow-up. For these couples, we will assess co-parenting behaviors including both parents and the Focal Child for 10 minutes and then ask the primary caregiver (self-identified at the time the observations are conducted) to engage in parent-child interactions for another 10 minutes with the Focal Child. For couples who are separated at the time of the 12-month follow-up, the research team will only ask the primary custodial parent to engage in parent-child interactions with the Focal Child for about 10 minutes.

The tasks that will be used to assess parenting and co-parenting behaviors will vary with the Focal Child’s age. For children younger than 2 years old at the follow-up point, one or both parents will be asked to play with the child with a bag of novel toys and teach the child a new skill (e.g. how to transfer an object from one hand to the other, how to bang or throw an object, etc.) while the child is placed in a high chair, car seat, or reclining chair. For children 2 to 6 years old at the follow-up point, one or both parents will be asked to play with the child with a bag of novel toys, complete a maze using an etch-a-sketch, and/or solve a puzzle. For children 7 to 8 years old at the follow-up point, one or both parents will be asked to complete a maze using an etch-a-sketch, solve a puzzle, and/or engage in a short discussion with their child about a series of topics, such as whether or not children should be asked to complete household chores, whether children should be allowed to watch television, etc. For children 9 to 15 years old at the follow-up point, one or both parents will be asked to engage in a short discussion with their child about a topic of disagreement (common topics used in prior research include the child’s appearance, engagement in after-school activities, engagement in school-related activities, and misbehaviors). These paradigms used to assess co-parenting and parent-child relationships have been well-established in the developmental literature and have been tested and validated with low-income and racially and ethnically diverse populations (see NICHD study of early childcare; Early Head Start evaluation; Conger, et al., 1992; Lindahl & Malik, 1999; Fivaz-Depeursinge et al., 1994; Hedenbro et al., 2006; Pinderhughes, et al., 2001).

Coding scheme for videotaped interactions. The videotaped couple, co-parenting and parent-child interactions will be reviewed by the research team and coded to assess the quality and nature of couple and family interactions using the Iowa Family Interaction Rating Scales (IOWA) macro-analytic coding system (Melby & Conger, 2001). The IOWA was designed to assess 1) dyadic interaction patterns and behaviors, such as escalating and reciprocated hostility, dominance, warmth and support, listener responsiveness, and prosocial and avoidant behaviors; 2) individual characteristics, such as humor, sadness, anxiety, positive mood and externalized negative behaviors; 3) parenting behaviors, such as neglecting, permissive, consistent harsh discipline, positive reinforcement, and intrusive behaviors; and, 4) quality of individual and group problem solving behaviors, such as effective and disruptive problem solving processes, negotiation and compromise, agreement, and solution quality. The IOWA coding scheme has been tested and validated with low-income families and has been fruitfully applied in prior research of marital conflict, family functioning, and child outcomes, all of which are key constructs of interest for the SHM evaluation (see Melby, Hoyt & Bryant, 2003; Melby et al., 1995; Conger et al., 1992;).

QUALITATIVE DATA COLLECTION EFFORTS FOR IMPLEMENTATION RESEARCH

The research team currently proposes to interview a wide variety of individuals associated with the SHM program including but not limited to the following: (1) supervisory staff such as program directors and clinical supervisors; (2) program staff such as recruitment and intake workers, family support workers, marriage education facilitators, and MIS employees; (3) clerical staff such as data entry workers and receptionists; and, (4) program participants. We have attached an example of the implementation study protocol that will be used by the study team (Appendix D). This protocol provides an example of the likely research activities and questions that will be collected as part of the qualitative data collection efforts for the implementation research. The finalized protocol will be informed by the results of the pilot phase of the SHM demonstration, which is currently on-going.

Because the purpose of the implementation research is interpretive, identification of interviewees will occur prior to the site visits and will be based on purposeful, convenience sampling of people that are identified as being best able to shed light on program operations. For program participants an attempt will be made to include a diverse range of couples who can provide possible insights into issues that are particularly relevant to SHM target populations. This includes people who represent different ages, genders, length of time married, and those who use English as a secondary language and/or non English speakers. Other variables yet to be determined might also be used to select participants for the focus groups. This sampling strategy will be used to ensure that the voices of a representative cross section of SHM participants are included in the findings.

Data gathered from program staff will focus on: (1) program implementation including the design and implementation of the program, barriers to program implementation and perceptions of what works and doesn’t work; (2) program enrollment including descriptions of target population, participant recruitment and enrollment efforts, the referral process, and participant retention; (3) staff and management issues such as the roles of SHM staff, changes in staffing composition, communication mechanisms among staff, and supervision; and (4) program monitoring and evaluation such as management of couple and program data, program changes as a result of monitoring feedback, and perceptions of evaluation and collaboration in cases where there are multiple sites operating under one umbrella agency.

Data from program participants will be gathered via observations and focus groups. The focus groups will provide opportunities for participants to exchange ideas and to comment on, support, and/or refute each others’ points of view. This type of methodology is useful because it complements findings from the survey by allowing the researchers to understand how many people have a particular belief or opinion, and it provides a mechanism to better understand how and why their opinions are constructed. Information gleaned from the focus groups will be used to further develop relevant lines of questioning for the 36-month survey instrument and will also be informative for the design of the program and for better understanding whether or not the program is meeting the needs of the couples.

A3. Use of Information Technology for Data Collection to Reduce Respondent Burden

The use of information technology has been incorporated into the data collection design wherever possible to reduce respondent burden.

The CATI/CAPI technology that will be used to administer the survey can reduce respondent burden. Computer programs enable respondents to avoid inappropriate and non-applicable questions. This technology allows for “individualized” question phrasing, and thus, for more streamlined administration. For example, separated respondents can automatically be routed past questions only relevant to respondents who are still living with their spouses. Respondents who are not currently working will not be prompted to answer questions concerning their employment. Also, depending upon the gender of the respondent to the survey, the appropriate “he” or “she” pronoun and the name of the respondent’s spouse/former spouse will automatically be inserted into the stems of questions.

A4. Efforts to Identify Duplication

The survey and the videotaped observations of couple, parent-child and co-parenting interactions will focus on information that cannot be found in administrative records or other existing sources. For example, the survey and videotaped observations will facilitate the collection of data on respondents’ marital relationship functioning, expectations and ideals about marriages, experiences in accessing program services, their physical and emotional well-being, their children’s health and behavior problems, parenting outcomes, family functioning, and nuanced characteristics of employment such as work hours and schedules. These types of information are not available in program or administrative records.

The implementation research efforts will facilitate understanding and documenting program operations and work practices of SHM staff, as well as provide insights into SHM couples’ program participation experiences. These data are not available from other administrative and existing data sources, and as such no duplication between the impact and implementation research efforts, or other data sources, is expected.

A5. Burden on Small Business

Does not apply. All respondents are individuals.

A6. Consequences to Federal Program or Policy Activities if the Collection of Information is not Conducted or is Conducted Less Frequently

If the 12-month follow-up data collection is not completed, we will not be able to adequately evaluate the impact of the SHM program for adults, children, or families. The analysis of the short-term impacts would be limited because changes in many important outcomes, such as marital relationship outcomes, marital and residential stability, marital relationship outputs, experiences with participation in services, co-parenting, parenting, father involvement, child well-being, and economic outcomes cannot be captured in administrative records data. In addition, impacts for population sub-groups could not be assessed, and analyses of the mediators of impacts could not be examined.

If the data are not collected, program operators and policy makers will receive little information about whether these particular enhancements to existing child-focused or two-generational services lead to impacts on parents and children in low-income families. The implementation study also depends on the collection of survey data at the 12-month follow-up, to obtain information on the services that are received by members of the program and control groups. The survey is the only way of obtaining these data in a comparable way for members of both the program and control groups, and this information is critical to fully understanding the service receipt differential between members of the program and control groups, as both groups receive the same survey instrument.

The qualitative data collected as part of the implementation research for this study will also inform the refinement of questions on survey instruments used at subsequent waves of follow-up, assist in the development of alternative hypotheses about program impact data, pursue emerging avenues of inquiry related to couples’ participation in SHM, and better understand program operations. The qualitative data are extremely important because they document program operations and activities to help policy makers understand whether these kinds of interventions can be replicated and successfully implemented on a large-scale.

A7. Special Data Collection Circumstances

No such circumstances.

A8. Form 5 CFR 1320.8(d) and Consultations Prior to OMB Submission

The 60-day Federal Register notice soliciting comments for the SHM 12-month data collection instruments was posted in the Federal Register on October 5, 2007. The notice yielded a request for a copy of the draft instrument and a general, unspecific comment that did not warrant a change in the instrument. A copy of the 60-day Federal Register notice and a draft of the 30-day notice are located in Appendices H and I respectively.

Although the self-reported survey and videotaped observations of couple, parent-child and co-parenting interactions at the 12-month follow-up represent an effort to break new ground in assessing marriage education programs, these instruments do build upon previous research. We have consequently developed instruments that incorporate items, scales and measures from other major experimental evaluations and large-scale studies, as well as smaller-scale experimental and non-experimental research conducted across various disciplines as much as possible, in addition to the baseline instruments and control group services survey used in the SHM evaluation. To the extent possible, measures included in the survey instrument and the protocol for conducting the videotaped observations of couple, parent-child and co-parenting interactions were drawn directly from prior research. Many measures on the 12-month survey instrument and protocol for the videotaped observations will be phrased exactly as they appeared in prior research, while others were modified to reflect the goals of the SHM initiative and the current evaluation as fully as possible, and to reflect the low literacy and comprehension skills of the current evaluation’s study population.

The survey questions for the 12-month follow-up draw extensively upon previous work conducted by Child Trends over the last two years in the measurement and conceptualization of key components of marital quality and healthy marriages. This work was funded by the Administration for Children and Families (ACF) through the National Institute for Child Health and Human Development as part of the Family and Child Well-Being Research Network (Grant# HD-30930). Many of the items were drawn from a compendium compiled by Child Trends that includes an extensive inventory of measures of healthy marriage that have been used in large- and small-scale studies and evaluations, many of which have been shown to have strong psychometric properties (Carrano, Cleveland, Bronte-Tinkew, and Moore, 2003). Items were also identified through memoranda commissioned by Child Trends from experts in the field (Jekielek, Moore, Carrano, and Matthews, 2003) and a review of the literature (Bronte-Tinkew, et al., 2003). Recommended items were then tested with a racially and economically diverse sample of engaged and married individuals through iterative rounds of cognitive interviews (see Guzman et al., 2005) and revised based on that experience. In addition, comments from consultants to the project led to further revisions and refinements.

Instruments that were used in the development of survey questions and selection of protocols for the videotaped observations for the 12-month follow-up are as follows:

• The SHM baseline instruments (OMB No. 0970-0299) and control group services survey (OMB Control No: 0970-0330);

• The BSF 15-month follow-up survey (OMB No. 0970-0304) developed by Mathematica Policy Research, Inc.;

  • MDRC surveys, including those used in the following projects: Canada’s Self-Sufficiency (SSP) project; the Employment Retention and Advancement (ERA) project (OMB No. 0970-0242 and OMB No. 0970-0265); the Enhanced Services for the Hard-to-Employ (HtE) project (OMB No. 233-01-0012); the Minnesota Family Investment Program (MFIP); the National Evaluation of Welfare-to-Work Strategies (NEWWS);

• Surveys done in connection with studies, including: the Fragile Families and Child Well-Being Study, the National Survey of Families and Households (NSFH); the Early Childhood Longitudinal Study – Birth and Kindergarten Cohorts (ECLS-B/K); The Early Head Start Research and Evaluation Project; and, other national surveys, such as the Current Population Survey (CPS), National Health Interview Survey (NHIS), the National Longitudinal Survey of Youth (NLSY), the Intergenerational Panel Study of Parents and Children, and the National Co-Morbidity Study;

• Marital quality protocols developed by expert researchers, including: Dr. John Gottman of the University of Washington, Drs. Philip and Carolyn Cowan of the University of California at Berkeley, Dr. Scott Stanley at University of Colorado at Denver, and Dr. Paul Amato, Penn State University.

Instruments that were used in the development of protocol for the videotaped observations for the 12-month follow-up are as follows:

• NICHD Study of Early Child Care; the National Evaluation of Early Head Start; and, protocols and instruments developed and used by the following researchers:  Thomas Bradbury and colleagues (Pasch & Bradbury, 1998; Sullivan, Pasch, Eldridge, & Bradbury, 1998); Rand Conger and colleagues (Conger et al, 1992); Philip and Carolyn Cowan and colleagues (Cowan, Cowan & Kerig, 1993; Kerig, Cowan, & Cowan, 1993); Carolyn Cutrona and colleagues (Cutrona, 1996); Elizabeth Fivaz-Depeursinge and colleagues (Fivaz-Depeursinge & Fivaz, 2006); John Gottman and Robert Levenson (Gottman & Levenson, 1992); Kristin Lindahl and colleagues (Lindahl, Clements & Markman, 1997; Lindahl; & Malik, 1999a, 1999b); James McHale and colleagues (McHale, Kursten-Hogan, Lauretti, Rasmussen, 2000; Schoppe-Sullivan, Mangelsdorf, Frosch, & McHale, 2004); and, Ellen Pinderhughes and colleagues (Pinderhughes, Nix, Foster, Jones & The Conduct Problems Prevention Research Group, 2001).

To select the measures for various components of the survey and develop the design of the observational study protocol, we consulted Paul Amato at  Penn State University; Thomas Bradbury at the University of California at Los Angeles; Chalandra Bryant at Penn State University; Philip and Carolyn Cowan at the University of California at Berkeley; Rolando Diaz-Loving at la Universidad Nacional Autónoma de México (UNAM); Frank Furstenburg at the University of Pennsylvania; Jan Melby at Iowa State University; Charles Negy at the University of Central Florida; the Administration for Children and Families; and, lead staff of the Building Strong Families Project at Mathematica Policy Research, Inc.

A9. Justification for Respondent Payments

The 12-month follow-up instruments that will be used to collect data from SHM sample members have some unique qualities that make administration difficult and threaten response rates. We are therefore requesting clearance to offer a monetary incentive to those who complete the 12-month follow-up data collection activities. Aspects of the survey effort that may make it more difficult to obtain high completion rates are:

• The surveys include questions that could be perceived as intrusive and therefore could make respondents reluctant to participate (i.e., questions about their mental health, drug and alcohol use, relationship satisfaction, and experience of domestic violence).

• Many participants may have negative feelings about the subject matter covered by the 12-month follow-up questions, such as relationship distress or substance use.

• Other difficulties in administering the 12-month follow-up stem from the nature of the population. Educationally and economically disadvantaged groups, such as those in the SHM sample, have been found to be more difficult than the general population to contact and to convince to participate in surveys.

These difficulties interact to make this survey of SHM sample members more difficult to conduct than surveys of the general population.

Thus, we are requesting clearance to use respondent payments for those who complete the 12- month follow-up surveys to obtain completion rates that will yield credible results, to avoid the bias that could result from selective non-response, and to reduce item non-response. In addition, providing an incentive at follow-up will also increase the likelihood that these sample members will respond to later follow-up surveys because sample members who receive monetary incentives for completing a past survey are more likely to respond to subsequent surveys (Singer, et al., 1998). We believe that the studies summarized here, and MDRC’s previous experiences with surveys of welfare recipients and other disadvantaged populations, make a strong case for the use of gifts and respondent payments for completing the SHM 12-month follow-up data collection activities.

To be effective, the amount of the incentives must fit the burden of the survey. We have based the amount to be paid to SHM respondents on prior research, and MDRC’s and Abt Associates’ experience collecting data from similar populations. We propose a $30 incentive for each member of the couple who completes the SHM 12-month follow-up survey. This amount reflects current practice in surveys using similar instruments and is also in line with the size of the incentive found to be effective for the ERA 42-month survey effort (OMB control number 0970-0285). (For the ERA study, a $25 incentive was given to respondents who completed the 30-minute interview in 2007). We believe that the increase in the amount of the financial incentive is warranted due to inflation and the increase of burden associated with the longer survey instrument. Based on prior fielding experiences of direct observational assessments with low-income populations, the research team also plans to offer incentives of $55 to each spouse, for couples who participate in both the observational study and the survey at the 12-month follow-up.

Approximately 6 months after random assignment, study participants will also be contacted by the survey firm through the mail and will be asked to update their contact information and will be given a small gift or giftcard valued at up to $2. Having access to high quality contact information is critical for achieving high response rates.

The purpose of the payment to parents for the survey and videotaped observations is to improve response rates by decreasing the number of refusals, enhancing respondent retention, speeding the data collection process, and providing a gesture of goodwill to acknowledge respondent burdens. The payments are being proposed in addition to many of the techniques suggested by OMB to improve response rates that have been incorporated into our data collection effort and are described in Section B.3, because our experience has shown that small monetary incentives are useful when fielding data collection instruments with low-income populations as part of a complex study design. The best statement of current thought on incentives is the Symposium on Providing Incentives to Survey Respondents convened in October 1992 by the Council of Professional Associations on Federal Statistics (COPAFS) for OMB. COPAFS asked Richard Kulka of NORC to write a review of the literature in light of what was learned at the symposium. Kulka concluded, “the greatest potential effectiveness of monetary incentives appears to be in surveys that place unusual demands upon the respondent, require continued cooperation over an extended period of time, or when the positive forces on respondents to cooperate are fairly low.” Kulka also wrote, “there is evidence that increasing the size of a monetary incentive will result in increases in survey response and/or response quality, although there is also consistent evidence that this benefit may rather quickly reach 'diminishing returns', whereby large incentives no longer result in appreciable increases in survey response” (Kulka, 1992).

In addition, more than two decades of survey research support the benefits of offering incentives. Hazard, citing evidence from a 1974 study by Ferber and Sudman found that the effects of incentives are contingent upon respondent burden (i.e., the effort needed to cooperate), the amount of the incentive, and the economic level of the respondent (Hazard, 2002). A study by Berlin and colleagues found that incentives increased the response rates of respondents with low levels of literacy, as well as lowering interviewer costs (Berlin et al., 1992). James also found that an incentive was effective in lowering non-response rates and that any incentive lowered the number of interviewer visits per case (James, 1997). The Mack et al. study of responders to the Survey of Income and Program Participation (SIPP) found that incentives reduced non-response rates in initial and subsequent interviews, and were particularly effective in reducing non-response rates in poor and African-American households (Mack, Huggins, Keathley & Sudukchi, 1998). Moreover, the use of incentives has been found to be efficacious for increasing the response rates of in-home and sensitive subject matter surveys (Mosher et al., 1994). Finally, our prior experience fielding data collection instruments with economically disadvantaged and TANF-receiving populations also supports the evidence that incentives increase response rates. For example, in a follow-up interview with Jobs Corps applicants, experimental evidence showed that incentives increased response rates and greatly increased search efficacy. Experience in these and similar studies of disadvantaged populations suggest that incentives can help convince reluctant respondents to participate (Moffitt, 2004). We believe that the studies summarized here, and MDRC’s previous experiences with fielding surveys and other kinds of assessments with low-income populations, make a strong case for the use of respondent payments for completing the survey and direct child assessments. We have based the amount of the incentive to be paid for these data collection elements on prior research conducted in this area, and the prior experience of MDRC and Abt Associates Inc. in interviewing similar populations.

For the qualitative data collection activities that are part of the implementation research for this study, we will not be paying financial incentives to program staff and study participants who participate in the naturalistic observations of program activities by research staff and open-ended interviews and small group discussions. However, study participants will be reimbursed child care and transportation expenses to facilitate their involvement in this part of the qualitative data collection efforts.

A10.    Confidentiality

Each potential participant in the SHM 12-month survey and observational study, and implementation research, will have been read the agreement to take part in the SHM study when they enrolled in SHM (see Appendix G for the entire informed consent form – OMB Control Number: 0970-0299).  This statement will explain the study and will assure them of their privacy and rights as respondents.  Specifically, the reference to confidentiality reads:

 

If you agree to be in the study, researchers will collect information about you and your children.

 

The information you share with the study team is important. It could help make these services available to other couples like you.  At the start of the study, you and your spouse will be asked to answer some questions in private.  These questions will ask you how well you get along with your spouse, how happy or sad you are, and what makes you upset.

 

If you agree to be in the study, you and your spouse will be interviewed one or more times over the next seven years by a survey company called Abt Associates.  Abt Associates is part of the research team for this study.  You will be asked about your marriage, how well you are getting along with your spouse, your experiences with [Local program], and your children.  You might also be asked to let us do some activities with your children.  You do not have to answer any question that you don’t want to answer. You will get [gift amount] for each interview.

 

If you agree to be in the study, [Local Program] program will share information with the research team about the services you get over the next five years as well.  We might also collect data from [State] about things like your wages and benefits. We might also collect data from [State] about services your children get, and your children’s school test scores.

 

Taking part in the study is your choice.  You may stop being in the study at any time. If you stop being in the study, we will use any information that we have collected before then.

 

Your Answers Will Be Kept Private

 

Only the study staff will be able to see information you give them. Your name will never appear in any public document.  All the study staff is trained to protect privacy.  Information gathered from [State] about you or your children will be marked with a code number, not names.  We also have a Confidentiality Certificate (CC) from the US government that adds special protection for the research information about you.   It says we do not have to identify you, even under a court order or subpoena.   Still, if keeping your answers private would put you, someone else or your child in serious danger, then we will have to tell government agencies to protect you or the other person.  And, the government may see your information if it audits us. 

 

At the outset of the 12-month follow-up effort, respondents will be reminded that they can refuse to answer the questions, their answers will be kept private, and that their agreement or refusal to participate will not affect their participation in the study or the ability to get services now or in the future.

 

The SHM Confidentiality Certificate from the National Institute on Alcohol Abuse and Alcoholism authorizes anyone connected with any information collections that are part of the SHM project to withhold the identity of subjects of the research.  The Confidentiality Certificate protects the privacy of all research data gathered by researchers from MDRC, its subcontractors and cooperating agencies, and anyone else who may come into contact with research information about SHM study participants.

 

A10.1  Confidentiality and Follow-Up Data Collection Activities

 

Abt Associates will be responsible for administering the survey and observational study effort.  Interviewers have access to the sample member’s name, address and telephone number which have been stored in their laptop.  All laptops used by Abt Associates’ interviewers and other members of the SHM project research team meet OMB’s and ACF’s data security and encryption standards. Abt’s proprietary Bellview CATI software allows interviewers to read and add information to files.  They cannot print or change information.  Specifically:

 

       Interviewers are not given the SSN’s of sample members. They are given Abt-generated ID numbers.

       Cases are delivered electronically through a sample control subsystem that is part of the Bellview CATI system.

       Interviewers can obtain sample member’s name, address, and telephone information from the Bellview CATI system.  This system only allows the interviewer to read information, not print or extract it by other means.

 

Handling case material.  Interviewers are sent cases via the CATI Data Collection System, which is integrated into Abt’s proprietary Field Management System (FMS).  The FMS is a major application composed of a set of interrelated applications that control all aspects of sampling, data collection, data cleaning and delivery of survey data.  Interviewers are instructed to keep the Abt ID number, respondent name, contact address, telephone information and answers private.  The interviewers are also instructed not to disclose any information to anyone not associated with the project.  Interviewers are allowed to discuss the interviews during interviewer meetings and during one-to-one supervisory sessions, but the interviews must be discussed in general terms, not identifying the individual.

 

In general, the interviewers do not have hard-copy files and all survey activities are completed electronically.  If they have any handwritten notes used during the tracking and locating process, these notes must be shredded at the end of each interview.

 

Training procedures for interviewers.  Abt Associates has a zero tolerance policy with regard to falsification or violation of respondent confidentiality/privacy.  Confidentiality requirements are reviewed with all project employees and in addition, in project specific trainings which include modules on confidentiality and the protection of privacy covered.  Abt Associate employees must also sign confidentiality pledges as a condition of their employment.

A11. Questions of a Sensitive Nature

This section contains additional justification for questions of a sensitive nature included in the 12-month follow-up. Several types of questions that will potentially be “sensitive” for respondents will be asked during the interview, including some regarding employment and income, mental health, substance abuse, family composition, participants’ own experiences with abuse or neglect, and attitudes and experiences regarding marriage and relationships.

To improve understanding of how the SHM program affected low-income married couples and their families, it will be necessary to ask these types of sensitive questions. For example, as discussed in Section A2, subgroups such as adults with mental health problems, high levels of relationship conflict or existing substance abuse may see different impacts due to the interdependence between these factors and marital satisfaction. Therefore, it is important to acquire a measure of these characteristics 12 months after the intervention. Questions will thus touch on the role of such factors as the quality of couple relationships, religion, sexual fidelity and similar matters that are commonly considered private. In cognitive testing, we found that couples were very comfortable responding to the draft instrument with this question, despite the sensitive nature of some questions. However, couples will not be pressed to divulge details of how these factors play out in their own lives and can choose not to answer any question. Furthermore, we are exploring whether sensitive questions, such as fidelity and the experience of emotional and physical abuse items at the 12-month follow-up can be asked in a self-administered questionnaire format using CAPI/CATI technology, where respondents complete the questions in private in order to maximize the privacy of respondents.

Due to particular concern on the SHM team and at the urging of MDRC’s institutional review board, we have also undertaken several measures to ensure that questions of a sensitive nature dealing with relationship conflict and violence will not exacerbate any dangerous situations for the respondents.

• The study participation warns participants that they will be asked sensitive questions and that they do not need to answer anything that makes them uncomfortable.

• As discussed in Section A10, we have provided with a Confidentiality Certificate that will allow us to protect sensitive research information from disclosure, even under a subpoena or court order. Participants are advised in the consent form, however, that we may still release information to prevent harm to a participant or another person.

• Lastly, we consulted with Dr. Richard Heyman of the State University of New York at Stony Brook, who recently completed a study of the effects of partner aggression research on couples, including those in unhappy relationships and experiencing physical violence (Heyman et al., 2006). In Dr. Heyman’s study, couples were asked to complete questionnaires about intimate partner violence, engage in conflict conversations, and be interviewed individually about anger escalation and de-escalation in these conversations. His research found that, in almost all cases, participants rated these activities as helpful to them personally and to their relationships. Those participants who did not see positive effects reported no impact, and there was no evidence whatsoever of dangerous behaviors resulting from these questions.

A12. Estimates of the Hour Burden of Data Collection to Respondents

Participation in all the 12-month follow-up data collection activities is completely voluntary. No sanction or penalty will be applied to those participants receiving state or federal assistance that choose not to provide information.

The respondents for the SHM 12-month survey will include both partners in married couples randomly assigned to the program and control groups in all eight sites of the SHM evaluation. This will be a fielded sample of 800 couples (1,600 individuals) per site. A subset of these couples (306 couples per site) who have at least one child who is between the ages of 0 and up to 15 at baseline will also be selected to participate in the observational study at the 12-month follow-up. One Focal Child will be selected for each couple (as described earlier) to participate in the observational study, and the ages of these children will range from newborns to adolescents up to 15 years old at baseline. For the process and implementation research, qualitative data collection efforts will target SHM program staff and study participants in the eight SHM sites.

A summary of the estimated response burden by 12-month follow-up component is illustrated below. Total burden hours are calculated as the number of respondents multiplied by the projected response rate, multiplied by the length of each follow-up component. MDRC projects an 80% response rate for the 50-minute survey instrument, which will be administered to all SHM participants in both the program and control groups. For the subset of couples randomly selected to participate in the observational study, MDRC anticipates a 72% response rate. In addition, by using national statistics on separation and divorce, burden for the observational study was estimated based on the projection that 90% of couples would be married at follow-up, while 10% would be separated or divorced. While both parents of the separated couple will only participate in the 12-month follow-up survey, the custodial parent in separated couples will be asked to engage in the parent-child interaction, making their burden significantly less than couples who are still intact at follow-up and engage in a couple, co-parenting, and parent-child interactions. This projection was taken into account when calculating burden. For the observational study estimate, the total burden is 5120 respondents with an average burden hour per response of approximately 32 minutes for a total annual burden estimate of 2758 hours. MDRC took the burden to SHM couples’ focal children into account distinguishing in the chart below between intact and separated couples and between children with intact and separated parents. The burden estimates for the implementation research are based on the time and number of program staff and study participants who we expect will be involved in the one-on-one interviews and small group discussions. The estimated burden hours are as follows:

ANNUAL BURDEN ESTIMATES

Instrument	Annual Number of Respondents	Number of Responses per Respondent	Average Burden Hours per Response	Estimated Annual Burden Hours
12-month survey	10,240	1	0.83 hrs	8,499.2
12-month observational study	5120	1	0.53875 hrs	2, 758.4
The process and implementation field research guide	504	1	1 hr	504

Estimated Total Burden Hours: 11761.6 hrs

A13. Estimates of Capital, Operating, and Start-Up Costs to Respondents and Record Keepers

Not applicable. The 12-month data collection instruments will be administered by Abt Associates and MDRC. The qualitative interviews and small group discussions as part of the implementation research will be conducted by MDRC and its research partners.

A14. Estimates of Costs to Federal Government

ACF is funding these activities. The estimated cost for designing the 12-month instruments, preparing submissions for OMB and for MDRC’s institutional review board, data entry and processing, and monitoring 12-month data collection is $4,709,865. This estimate also includes an incentive to maximize response rates of $30 per person to compensate respondents for their time for completing the survey and $55 per adult to compensate respondents for participating in the observational study and survey, as well as the tracking incentive of up to $2 sent to study participants approximately 6 months after random assignment. We expect these costs to spread out over three years as we administer the 12-month follow-up instruments and process the resulting data in each site. These costs also include any reimbursement to study participants for childcare and transportation costs.

A15. Changes in Burden

The efforts are all new data collection efforts and do not involve a change in burden.

A16. Tabulation, Analysis, and Publication Plans and Schedule

Tabulation of data from survey and observational study at 12-month follow-up. None of the tables will present individual-level data, all of the results and sample characteristics will be presented in aggregate form.

A16.1. Analysis Plans for 12-Month Data Collection Efforts

Assessing and monitoring the quality of the data from the 12-month survey. The follow-up survey will go through a rigorous series of tests for completeness and quality. Staff at the survey firm will review the initial cases completed by each interviewer, as well as perform occasional spot checks after that. Editing/coding staff will review questionnaires for quality and consistency after this initial period. Interviewers will be apprised of any problems found and retrained as needed. During the coding of data, coder reliability checks will be undertaken repeatedly to verify that coding procedures are being followed correctly. Data entered into computer files will be assessed for missing information, outliers, and other data problems according to standard procedures. If necessary, questionnaires will be recoded. The survey firm will deliver to MDRC data sets of completed cases at agreed-upon intervals, along with marginal frequencies. The data and frequencies will be reviewed for outliers, unusual distributions and inconsistencies between data items. Lastly, because some of the measures of relationship quality, expectations, values and ideals have been developed specifically for the SHM evaluation, and have not been tested and validated with low-income couples, one of the roles of the survey and observational data will be to help us understand how different items relate to each other, as well as to validate survey items with data from the observational study.

Assessing, monitoring the quality of, and coding the data from the videotaped observational study. The videotaped observations will be conducted by the same field interviewers who administer the 12-month survey. They will receive extensive training in how to administer these assessments. This training will consist of pre-training exercises and mock set ups of the videotaping equipment and protocol for the activity and discussion-based interactions. Finally, each interviewer will undergo a certification process prior to fielding to ensure that the interviewer is qualified to set up the videotaped observations. To ensure that the interviewers are qualified to conduct the observations, they will undergo a certification process, in which they must meet select administration criteria on two separate practice administrations (Goyette et al., 2006). The SHM research team will also work with Abt Associates to monitor early interviews and conduct periodic reviews of the observations over the course of fielding the data collection instruments to ensure that interviewers are following procedures and protocols with a high degree of fidelity. In addition, 20 percent of all videotaped observations will be reviewed by MDRC in order to monitor reliability and to ensure that a high quality of data is being collected once fielding of the assessments has begun. Upon reviewing the videotapes, MDRC will be able to provide interviewers with feedback and guidance in order to trouble-shoot administration problems and deviations from the prescribed procedures and protocols for administering the direct child assessments. This will allow MDRC to make any mid-course corrections within a relatively short timeframe.

The videotaped interactions from the observational study will be sent to a coding lab, where trained coders will observe and code the individual’s and couple’s behaviors during the taped interactions. The videotaped couple, co-parenting and parent-child interactions will be reviewed by the research team and coded to assess the quality and nature of couple and family interactions using the Iowa Family Interaction Rating Scales (IOWA) macro-analytic coding system (Melby & Conger, 2001). To establish interrater reliability of the coding, two coders will independently code a randomly-selected portion of the tapes (20% of the tapes). The double-coding helps to catch idiosyncrasies associated with any coder’s ratings or those associated with any of the couple’s style of communication. Intra-Class Correlations will be used to assess inter-rater reliability on continuous measures and Cohen’s Kappa statistics will be used to assess inter-rater reliability on all categorical measures.

Data file construction for survey and observational data collected at 12-month follow-up. Data from the 12-month survey and videotaped observations will then be merged with data from other sources. That is, data from the 12-month follow-up will be combined with the other data collected for this project, including the baseline instruments, the MIS data on program participation, administrative records information relating to welfare receipt, earnings, child welfare, and other program tracking (if available) and data collected from subsequent waves of follow-up with study participants and their children.

As previously indicated, the SHM evaluation utilizes a random assignment analytic design. We offer a brief outline of how we will address the project’s long-term analytical goals, with a focus on how the follow-up survey and videotaped observational data will be useful in that process.

Estimating overall impacts. Although the use of a randomized design will ensure that simple comparisons of experimental and control group means will yield unbiased estimates of program effects, the precision of the estimates will be enhanced by estimating multivariate regression models that control for factors at baseline that also affect the outcome measures. Such impacts are often referred to as “regression-adjusted” impacts. Examples of factors that may affect outcomes are the sample members’ age, number of children, prior employment, and baseline levels of marital distress.

As suggested by the range of questions on the proposed survey, a range of outcomes will be examined in the impact analysis, including marital stability, relationship quality, expectations, ideals and attitudes, economic and employment outcomes, receipt of services, child care outcomes, aspects of family functioning and routines, parental psychological well-being and health, parenting, parent-child relationships, father involvement, mental and physical health outcomes – stress, distress, substance and alcohol use, and social support. We are still developing an exact analysis plan to guide how we will construct various outcomes.

In drawing inferences about estimated impacts, standard statistical tests such as the two-group t-test (for continuous variables such as an index of marital quality) or chi-square tests (for categorical measures, such as marital status) will be used to determine whether estimated effects are statistically significant (Green, 1999). Since we will analyze multiple outcomes, we will explore the possibility of adjusting estimates to account for this fact, for example, by using a Bonferroni correction (Darlington, 1990) or other omnibus test (such as those discussed in Cooper & Hedges, 1994).

Subgroup analyses. In some cases, impacts can vary for certain subgroups based on their demographic characteristics or circumstances at baseline. For this reason, it is essential to go beyond the examination of overall impacts of the program to examine impacts among subgroups defined by level of disadvantage and other characteristics. Impacts might differ for participants and their children according to a variety of baseline characteristics. Program impacts, particularly on family functioning, parenting practices, and child outcomes, might also vary by children’s characteristics, such as gender and age at study entry.

An analysis of subgroup impacts involves estimating the program’s effects for each subgroup separately, using the regression-adjusted model mentioned earlier, and then comparing the two impacts. The standard errors of each of the impacts are used to assess whether the two impacts are statistically significantly different from each other. Subgroup impacts estimated in this way are referred to as unconditional subgroup impacts, because they show the gross effect of a particular characteristic, such as length of marriage, on a program’s impacts. As an example, impacts on marital stability may be smaller for participants who have experienced childhood abuse or neglect, as compared with their counterparts who did not share such experiences. However, this difference may arise not because of education per se, but because those who have experienced childhood abuse or neglect are also less likely to take up SHM services, which also affects how they benefit from the program. In this case, it would be of interest to estimate conditional subgroup impacts, or impacts by education level that also control for prior work experience. These impacts would be obtained by pooling the sample and estimating one impact model, in which particular background characteristics are interacted with all of the other variables in the model and with the program group dummy variable. For example, if the coefficient on the interaction of program status and whether the participant experienced childhood abuse is reduced in size once the interaction of program status and prior relationship quality is included, we can conclude that some part of the effect of education on the program’s impacts is due to its correlation with these prior experiences.

A16.2 Storing, analyzing and presenting qualitative data for implementation research.

Qualitative data will be collected by members of the implementation research team, who will make field notes with observations of SHM program activities and interviews. Field notes will include illustrative descriptions of the setting, program staff and SHM participants including but not limited to: descriptive data about the individuals involved in the observation; their actions and activities; the amount of time spent in these activities; the goals that they were trying to accomplish; whether or not the goals were met; and any unanswered questions the researcher might have about the observations. After the observation is complete, the field notes will be transformed into diagrams with boxes that contain key concepts and arrows indicating the relationships between the concepts and/or an organized framework/matrix will be used that is reflective of the goals of the research and that can be ordered for analysis. Open-ended semi-structured interviews and small group discussions with program staff and study participants may be conducted in English or Spanish and may also be audiotaped and transcribed by the implementation research team. NVIVO software will be used to assist with storing and coding data of focus group data. Coding of the focus group data will be entered into the NVIVO database and frequencies of coded responses will be computed. Matrices will be developed to create classification systems, examine salient patterns, and allow the researchers to map relationships among the codes and themes or the purpose of producing graphics. Lastly, the implementation research team will review various documents produced by the SHM sites, such as program literature and brochures, attendance records, emergency assistance and incentive records, and program policy and procedure manuals, to examine similarities and differences between various SHM sites. 

For all of the data collection efforts the team will hold regularly scheduled meetings to develop and discuss strategies for confirming the research findings. This will include cross checks (triangulation) with all of the data collection efforts to look for evidence or counter evidence about the findings. This process helps the researchers to gain insight into whether a particular point or explanation holds true across the variety of data sources. If, for example, the finding is supported by observations, field notes, interviews, and/or site documents – then the level of confidence in the findings is higher, and the researchers can surmise that what they have identified is integral to understanding a particular program site or sites, or a group of program participants.

SCHEDULE

ACF expects that MDRC will begin administering the 12-month follow-up instruments in April 2008. The exact timing of the 12-month data collection however will depend on receipt of OMB clearance and on progress in site development and program pilots. Summaries of the 12-month data will be prepared within a few months after survey completion, approximately three years after the beginning of random assignment in each site.   

As noted earlier, the information obtained through the 12-month data collection will be critical to the overall SHM project. An interim impact report scheduled for 2011 will include analyses from the 12-month survey and observational data to assess early impacts on key outcomes (marital satisfaction and stability, couple relationship quality, child well-being, and so on), as well as an update on program implementation in each site (particularly attendance, completion, and quality of services). The project also plans additional publications, including a cross-site implementation report in 2010 that will draw on the qualitative implementation research proposed in this submission, and a final impact analysis report (including updated impacts based on a later follow-up survey) scheduled for 2013, as well as research briefs and special topic reports as requested.

A17. Reasons for Not Displaying the OMB Approval Expiration Date

Not applicable. We intend to display the OMB approval number and expiration data on all 12-month follow-up materials and protocols for the qualitative data collection on efforts as part of the implementation research.

A18. Exceptions to Certification Statement

Not applicable. We have no exceptions to the Certification Statement.

REFERENCES

Alexander, K. L., Entwisle, D. R., and Dauber, S. L. 1993. “First-grade classroom behavior: Its short-and long-term consequences for school performance.” Child Development, Vol. 64, No. 3: 801-14.

Amato, P.R. 2004. “Tension between institutional and individual views of marriage.” Journal of Marriage and the Family, 66: 959-965.

Amato, P. R. 2000. “Consequences of divorce for adults and children.” Journal of Marriage and the Family 62: 1269-1287.

Amato, P. R. and S.J. Rogers. 1999. “Do attitudes toward divorce affect marital quality?” Journal of Family Issues 20, 1: 69-86.

Amato, P.R. and S.J. Rogers. 1997. “A longitudinal study of marital problems and subsequent divorce.” Journal of Marriage and the Family 59, 3: 612-624.

Baucom, D.H., Epstein, M., Sayers, S., and Sher, T.G. 1989. The role of cognition in marital relationships: Definitional, methodological, and conceptual issues. Journal of Consulting and Clinical Psychology, Vol. 57: 31-38.

Baucom, D.H., Epstein, N., Rankin, L.A., & Burnett, C.K. 1996. Assessing relationship standards: The inventory of specific relationship standards. Journal of Family Psychology, 10: 72-88.

Belcher, H.M., Butz, AM, Pulsifer M., Marano, N. 2001. Effectiveness of a home intervention for perceived child behavioral problems and parenting stress in children with in utero drug exposure. Archives of Pediatric and Adolescent Medicine. Vol.155:1029-1037

Berlin, M., Mohadjer, L., Waksberg, J., Kolstad, A., Kirsch, D., Rock, D., and Yamamoto, K. 1992. “An experiment in monetary incentives.” Proceedings of Survey Research Methods Section of American Statistical Association, 393-8.

Bradbury, T.N. & Fincham, F.D. 1992. Attributions and behavior in marital interaction. Journal of Personality and Social Psychology, Vol. 63, 613-628.

Bradbury, T. N., F.D. Fincham, and S.R.H. Beach. 2000. “Research on the nature and determinants of marital satisfaction: A decade in review.” Journal of Marriage and the Family 62: 964-980.

Bramlett, M. D. and W.D. Mosher. 2002. “Cohabitation, marriage, divorce, and remarriage in the United States.” In Vital and Health Statistics 23, 22. Hyattsville, MD: National Center for Health Statistics.

Bringing Baby Home. 2006. Research Evaluation of the Bringing Baby Home Program. Unpublished findings presented on www.bbhonline.org/research.shtml

Bronte-Tinkew, J., et al. 2003. “Conceptualizing and measuring healthy marriages: A review of the literature and annotated bibliography.” Washington, DC: Child Trends.

Bronte-Tinkew, J., Carrano, and Guzman, L. 2006. Resident fathers’ perception of their roles and links to involvement with infants. Fathering: A Journal of Theory, Research, and Practice about Men as Fathers, Vol. 4, No. 3, 254-285.

Bryant, C. M. and R.D. Conger. 1999. “Marital success and domains of social support in long-term relationships: Does the influence of network members ever end?” Journal of Marriage and the Family 61: 437-450.

Buehler, C. and Gerard, J.M. 2002. Marital conflict, ineffective parenting, and children’s and adolescents’ maladjustment. Journal of Marriage and Family, Vol. 64, No. 1: 78-92.

Cabrera, N.J. et al. 2004. Low-income fathers’ involvement in their toddlers’ lives: Biological fathers from the Early Head Start Research and Evaluation Study. Fathering: A Journal of Theory, Research, and Practice about Men, Vol. 2, No. 1: 5-30.

Carlson, M., McLanahan, S., and England, P. 2003. “Union formation and dissolution in Fragile Families.”

Carlson, M.J., and McLanahan, S. 2006. Strengthening unmarried families: Could enhancing couple relationships also improve parenting? Social Service Review, Vol. 80, No. 2: 297-321.

Carrano, J., K. Cleveland, J. Bronte-Tinkew, and K. Moore. 2003. “Conceptualizing and measuring healthy marriages.” CD-ROM. Washington, DC: Child Trends.

Carrere, S., Buehlman, K.T., Gottman, J.M., Coan, J.A., and Ruckstuhl, L. 2000. Predicting marital stability and divorce in newlywed couples. Journal of Family Psychology, Vol. 14, No. 1: 42-58.

Clements, M.L., S.M. Stanley, and H.J. Markman. 2004. “Before they said "I do": Discriminating among marital outcomes over 13 years based on premarital data.” Journal of Marriage and Family 66: 613-626.

Cohen, S. & Williamson, G. 1988. Perceived stress in a probability sample of the United States. In S.Spacapan & S. Oskamp (Eds.) The social psychology of health: Claremont Symposium on applied social psychology. Newbury Park, CA: Sage.

Coleman, M., L. Ganong, and M. Fine. 2000. “Reinvestigating remarriage: Another decade of progress.” Journal of Marriage and the Family, 62: 1288-1307.

Conger, R., Conger, K, and Elder, G. 1997. "Family Economic Hardship and Adolescent Adjustment: Mediating and Moderating Processes." Consequences of Growing Up Poor, edited by G. Duncan and J. Brooks-Gunn. New York: Russell Sage, 288-310.

Conger, R.D., G.H. Elder, F.O. Lorenz, K.J. Conger, R.L. Simons, L.B. Whitbeck, S. Huck, and J.N. Melby. 1990. “Linking economic hardship to marital quality and instability.” Journal of Marriage and the Family 52: 643-656.

Cooper, H. & Hedges, L.V. (Eds.). 1994. Handbook of Research Synthesis. New York:Russell Sage Foundation.

Cowan, C.P., & Cowan, P.A. 1988. Who does what when partners become parents: Implications for men, women, and marriage. Marriage and Family Review, 13: 1-12.

Cowan, C. P. and Cowan, P.A. 1992. When partners become parents: The big life change for couples. New York: Basic Books.

Cowan, P.A., Cowan, C.P, and Kerig, P.K. 1993. Marital quality and gender differences in parent-child interaction. Developmental Psychology, Vol 29, No. 6: 931-939.

Cowan, C.P. and Cowan, P.A. 2006. The Case for Preventive Intervention to Strengthen Couple Relationships: Good for Couples, Good for Children. Presentation at the Evolving Families Conference entitled Marriage and Family: Complexities and Perspectives. Cornell University, April 7-8, 2006.

Cowan, C.P., Cowan P.A., Pruett, M.K., and Pruett, K. 2006. The Supporting Father Involvement Study: The first public report. Department of Social Services, Sacramento, CA, March 27, 2006.

Cummings, E.M. & Davis, P. 1994. Children and marital conflict. New York: Guilford.

Cummings, E. M., Ballard, M., El-Sheikh, M., and Lake, M. 1991. Resolution and children's responses to interadult anger. Developmental Psychology, 27: 462-470.

Cutrona, C. E. 1996. Social support in couples: Marriage as a resource in times of stress. Thousand Oaks, CA: Sage Publications

Cutrona, C. E., Russell, D.W., Abraham, W.T., Gardner, K.A., Melby, J.N., Bryant, C., and Conger, R.D. 2003.  Neighborhood context and financial strain as predictors of marital interaction and marital quality in African American couples.  Personal Relationships, 10: 389-409.

Darlington, R. B. 1990. Regression and Linear Models, New York: McGraw-Hill.

Dishion, T. J., Patterson, G. R., and Kavanagh, K. A. 1992. An experimental test of the coercion model. In J. McCord & R. Trembley (Eds.), The interaction of theory and practice: Experimental studies of interventions (pp. 253-282). New York: Guilford Press.

Edin, K. 2000. “What do low-income single mothers say about marriage?” Social Problems 47, 1: 112-133.

Edin, K. and Kefalas, M. 2005. Promises I Can Keep: Why Poor Women Put Motherhood Before Marriage. University of California Press, Berkeley.

Edin, K. and Lein, L. 1997. Making Ends Meet: How Single Mothers Survive Welfare and Low-Wage Work. New York: Russell Sage Foundation.

Ellwood, D.T. and C. Jencks. 2004. The spread of single-parent families in the United States since 1960. KSG Faculty Research Working Paper Series RWP04-008. Cambridge, MA: Harvard University, Kennedy School of Government.

Emery, R.E. 1982. Inter-parental conflict and the children of discord and divorce. Psychological Bulletin, 92: 310-330.

Erel, O. and Burman, B. 1995. Interrelatedness of marital relations and parent-child relations: A meta-analytic review. Psychological Bulletin, 110: 108-132.

Fein, D. et al. 2003. The determinants of marriage and cohabitation among disadvantaged Americans: Research findings and needs. Report prepared for the Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research, and Evaluation.

Fincham, F.D. 2003. Marital conflict: Correlates, structure and context. Current Directions in Psychological Science, 12: 23-27.

Fincham, F.D., Bradbury, T.N., and Grych, J.H. 1990. Conflict in close relationships: The role of intrapersonal phenomena. In S. Graham and V.S. Folkes (Eds.), Attribution theory: Applications to achievement, mental health, and interpersonal conflict. Hillsdale, NJ: Erlbaum.

Fivaz-Depeursinge, E. and Favez, N. 2006. Exploring triangulation in infancy: Two contrasted cases. Family Process, Vol. 45, No. 1: 3-18.

Fivaz-Depeursinge, E. and Bungin, D., Corboz-Warnery, A., Lebovici, S., Stern, D., Byrng-Hall, J., and Lamour, M. 1994. The dynamics of interfaces: Seven authors in search of encounters across levels of description of an event involving a mother, father, and baby. Infant Mental Health Journal, 15: 69-89.

Florsheim, P., Moore, D., Zollinger, L., MacDonald, J., Sumida, E. 1999. The transition to parenthood among adolescent fathers and their partners: Does antisocial behavior predict problems in parenting? Applied Developmental Science, Vol. 3, No. 3: 178-191.

Fowers, B. J. 2004. “Conceptualizing and measuring healthy marriages and positive relationships.” Conceptualizing and measuring “healthy marriages” for empirical studies: Recommendation memos from experts in the field. Washington, DC: Child Trends.

Frisco, M.L. and Williams, K. 2003. “The Division of Housework, Marital Happiness and Marital Status among Dual-Earner Couples.” Journal of Family Issues, 22(1): 51-73.

Gable, S. L., Gonzaga, G., & Strachman, A. 2006. Will you be there for me when things go right? Social Support for Positive Events. Journal of Personality and Social Psychology, 91: 904-917.

Gerson, K. 1993. No man's land. New York: Basic Books.

Gottman, J.M., Coan, J., Carrer, S. & Swanson, C. 1998. “Predicting martial happiness and stability from newlywed interactions.” Journal of Marriage and the Family, 60, 1: 5-22.

Gottman, J. 1994. Why marriages succeed or fail. New York: Simon & Schuster.

Gottman, J.M., and Levenson, R.W. 1992. Marital processes predictive of later dissolution: Behavior, physiology, and health. Journal of Personality and Social Psychology, Vol. 63, No. 2: 221-233.

Gottmann, J.M. and Katz, L. 1989. Effects of marital discord on young children’s peer interaction and health. Developmental Psychology, 25: 373-381.

Greenstein, T.N. 1996. Gender ideology and perceptions of the fairness of the division of household labor: Effects on marital quality. Social Forces, Vol. 74, No. 3: 1029-1042.

Grych, J.H. and Fincham, F.D. 2001. Interparental conflict and child development: Theory, research, and applications. New York: Cambridge University Press.

Gudmunson, C.G., Beutler, I.V., Israelsen, C.L., McCoy, J.K, and Hill, E.J. 2007. Linking financial strain to marital instability: Examining the roles of emotional distress and marital interaction. Journal of Family and Economic Issues, Vol. 28, No. 3: 1573-3475.

Guzman, L., G. Matthews, J. Hamilton, and K. Moore. 2005. “Summary Report on Cognitive Interviews for Supporting Healthy Marriage.” Washington, DC: Child Trends.

Hair, E., McPhee, B., Moore, K.A., Vandivere, S. 2005. How do parenting behaviors mediate the link between maternal depressive symptoms and child well-being? APPAM Annual Meeting, Washington, DC.

Halberstadt, A.G. 1983. Family expressiveness styles and nonverbal communication skills. Journal of Nonverbal Behavior, 8: 15-26.

Halford, W.K., Markman, H.J., Kline, G.H., and Stanley, S.M. 2003. Best practice in couple relationship education. Journal of Marital and Family Therapy, Vol. 29, No. 3, 385-406.

Hawkins, A.J., Fawcett, E.B., Carroll, J.S., Gilliland, T.T. 2006. The marriage moments program for couples transitioning to parenthood: Divergent conclusions from formative and outcome evaluation data. Journal of Family Psychology, Vol. 20, No. 4: 561-570.

Hedenbro, M., Shapiro, A., and Guttmann, J. 2006. Play with me at my speed: Describing differences in tempo of parent-infant interaction in the Lausanne Triadic Play Paradigm in two cultures. Family Process, 24: 145-167.

Hetherington, E.M. and Clingempeel, W.G. 1992. Coping with marital transitions: A family systems perspective. Chicago: University of Chicago Press for the Society for Research in Child Development.

Heyman, R.E., D.J. Owen, and A.M. Smith Slep. In press. “The Risk of Partner Aggression Research: Impact of Laboratory Couples Conflict Protocols on Participants.”

Heyman, R. E., Owen, D.J. & Slep, A. M. 2006. “The risk of partner aggression research: Impact of laboratory couples conflict.” Violence and Victims, 21: 483-497.

Hochschild, A. 1989. The second shift. New York: Avon.

Horwitz, A.V. and H. Raskin. 1996. “Becoming married and mental health: A longitudinal study of a cohort of young adults.” Journal of Marriage and the Family, 58, 4.

Hops, H. 1992. Parental depression and child behavior problems: Implications for behavioral family intervention. Behavior Change, Vol. 9: 126-138.

House, J.S., D. Umberson, and K. Landis. 1988. “Structures and processes of social support.” Annual Review of Sociology 14: 293-318.

Hughes, D., E. Galinski, and A. Morris. 1992. “The effects of job characteristics on marital quality: Specifying linking mechanisms.” Journal of Marriage and the Family 54: 31-42.

Huston, T. L. & Chorost, A. F.  1994.  Behavioral buffers on the effect of negativity on marital satisfaction: A longitudinal study,  Personal Relationships, 1, (3): 223-239.

Interagency Forum on Child and Family Statistics. 1999. America’s Children: Key National Indicators of Well-Being. Washington, DC: US Government Printing Press.

James, T. 1997. Results of the Wave 1 incentive experiment in the 1996 survey of income and program participation. Proceedings of the Survey Research Section of the American Statistical Association.

Jekielek, S.M., K. Moore, J. Carrano, and G. Matthews (eds). 2003. Conceptualizing and measuring ‘healthy marriages’ for empirical research and evaluation studies: Recommendation memos from experts. Washington, DC: Child Trends.

Johns, A.L., Newcomb, M.D., Johnson, M.D., & Bradbury, T.N. 2007. Alcohol-related problems, anger, and marital satisfaction in monoethnic Latino, biethnic Latino, and European American newlywed couples. Journal of Social and Personal Relationships, 24: 255-275.

Jordan, P.L., Silliman, B., Stanley, S.M., Coffin, W., and Markman, H.J. 2001. Preventive interventions for couples. In H. Liddle, D. Santisteban, R. Levant, and J. Bray (Eds.), Family Psychology: Science-Based Interventions (pp. 123-146). Washington, D.C.: APA Publications.

Karney, B.R. and T.N. Bradbury. 1995. “The longitudinal course of marital quality and stability: A review of theory, method and research.” Psychological Bulletin 118: 1.

Kessler, R.C., et al. 2003. Screening for serious mental illness in the general population. Arch Gen Psychiatry, 60: 184-189.

Kurdek, L.A. 1993. “Predicting marital dissolution: A 5-year prospective longitudinal study of newlywed couples.” Journal of Personality and Social Psychology, 64: 221-242.

Mack, S., Huggins, V., Keathley, D., and Sundukchi, M. 1998. Do monetary incentives improve response rates in the survey of income and program participation? U.S. Bureau of the Cnesus, Demographic Statistical Methods Division, Washington D.C.

Martin, S. and Parashar, S. 2003. Divergent trends in educational attainment and attitudes towards marital dissolution from 1974-2000: An emergent source of societal inequality? Paper presented at the annual meeting of the American Sociological Association, Atlanta Hilton Hotel, Atlanta, GA. <http://www.allacademic.com/meta/p106925_index.html.>

McBride, B.A., Rane, T.R., and Bae, J. 2001. Father/male involvement in prekindergarten at-risk programs: An exploratory study. Early Childhood Research Quarterly, Vol. 16, 77-93.

McHale, J.P. and Cowan, P.A. 1996. Understanding how family-level dynamics affect children’s development: Studies of two-parent families. Jossey-Bass Publishing, 112.

McHale, J.P., Kuersten-Hogan, R., Lauretti, A., Rasmussen, J.L. 2000. Parental reports of coparenting and observed coparenting behavior during the toddler period. Journal of Family Psychology, Vol. 14, No. 2: 220-236.

McLanahan, S. and G. Sandefur. 1994. Growing up with a single parent: What hurts, what helps. Cambridge, MA: Harvard University Press.

McLoyd, V. 1990. “The impact of economic hardship on African-American families and children: Psychological distress, parenting, and socio-emotional development.” Child Development 61, 2: 311-46.

Melby, J.N., Hoyt, W.T., and Bryant, C.M. 2003. A generalizability approach to assessing the effects of ethnicity and training on observer ratings of family interactions. Journal of Social and Personal Relationships, 20: 171-191.

Melby, J.N., and Conger, R.D. 2001. The Iowa Family Interaction Rating Scales: Instrument summary. In Kerig and Lindahl (Eds.). Family observational coding systems: Resources for systematic research (pp. 33-58). Mahway, NJ: Erlbaum.

Melby, J.N., Ge, X., Conger, R.D., and Warner, T.D. 1995. The importance of task in evaluating positive marital interactions. Journal of Marriage and the Family, 57: 981-994.

Minton, C. and Pasley, K. 1996. Fathers’ parenting role and father involvement: A comparison of nondivorced and divorced, nonresident fathers. Journal of Family Issues, Vol. 17, No. 1: 26-45.

Moffitt, Robert A. 2001. Policy interventions, low-level equilibria, and social interactions. In Social Dynamics, edited by S.N. Durlauf and H.P. Young. Cambridge: MIT Press, 25-82.

Moore, Kristin A., J. Bronte-Tinkew, S. Jekielek, L. Guzman, S. Ryan, Z. Redd, J. Carrano, G. Matthews.  2007. Developing Measures of Healthy Marriages and Relationships.  Handbook of Measurement Issues in Family Research, S. Hofferth and L. Casper, Eds. Mahwah, NJ:  Lawrence Erlbaum Associates.

Moore, K. A., D. Hogan, B. Brown, L. Lippman, G. Sandefur, F. Goldscheider, and M. Msall. 2005. “Indicators of child well-being, positive outcomes measures and indicators of child disability: Summary of accomplishments (1993-2005).” Report prepared for the NICHD Family and Child Well-Being Research Network.

Moore, K. A., Hair, E. C., Vandivere, S., McPhee, C. B., McNamara, M., & Ling, T. 2006. Depression among moms: prevalence, predictors, and acting out among third grade children. research brief. Washington, D.C.: Child Trends.

Moore, K.A. & Vandivere, S. 2005. Turbulence: The Effects of Change. In Cosby, A, R. Greenberg, L. Southward and M. Weitzman (Eds.), About Children. Mississippi State University: The Social Science Research Center.

Moos, R.H. and Moos, B.S. 1994. Family environment scale manual: Development, applications, research (3^rd ed.). Palo Alto, CA: Consulting Psychologists Press.

Morrison, D.R., and Corio M. 1999. Parental conflict and marital disruption: Do children benefit when high-conflict marriages are dissolved? Journal of Marriage and the Family, Vol. 61, No. 3: 626-637.

Mosher, W.D., Pratt, W.F., and Duffer, A.P. 1994. CAPI, event histories, and incentives in the NSFG cycle 5 pretest. Proceedings of the Survey Research Methods Section of the American Statistical Association, 1: 59-63.

Oliker, S.J. 1989. Best friends and marriage. Berkeley: University of California Press

Pasch, LA., and Bradbury, T.N. 1998. Social support, conflict, and the development of marital dysfunction. Journal of Consulting and Clinical Psychology, Vol. 66, No. 2: 219-230.

Perry-Jenkins, M., and Crouter, A.C. 1990. Men’s provider role attitudes: Implications for household work and marital satisfaction. Journal of Family Issues, Vol. 11: 126-156.

Pinderhughes, E.E., Nix, R.E., Foster, M., and Jones, D. Parenting in context: Impact of neighborhood poverty, residential stability, public services, social networks, and danger of parental behaviors. Journal of Marriage and the Family, Vol. 63, No. 4: 941-953.

Presser, H.B. 2000. Nonstandard work schedules and marital instability. Journal of Marriage and Family 62 (1): 93–110.

Reardon-Anderson, J., Stagner, M., Macomber, J.E., and Murray, J. 2005. Systematic review of the impact of marriage and relationship programs. Washington D.C.: Administration for Children and Families, U.S. Department of Health and Human Services.

Reichman, N. E., H. Corman, and K. Noonan. 2003. “Effects of child health on parents’ relationship status.” Bendheim-Thoman Center for Research on Child Well-being Discussion Paper #03-21-FF. Princeton, NJ: Princeton University.

Reis, H.T., and Shaver, P. 1988. Intimacy as an interpersonal process. In S. Duck (Ed.), Handbook of personal relationships (pp. 367-389). Chichester, England: Wiley.

Ribar, D. C. 2004. What Do Social Scientists Know About the Benefits of Marriage? A Review of Quantitative Methodologies. Discussion Paper No. 998. Bonn, Germany: Institute for the Study of Labor. Available from www.iza.org

Rusbult, C.E. and Buunk, B.P. 1993. Commitment processes in close relationships: An interdependence analysis. Journal of Social and Personal Relationships, Vol. 10, No. 2: 175-204.

Schoppe-Sullivan, S.J., Mangelsdorf, S.C., Frosch, C.A., and McHale, J.L. 2004. Associations between coparenting and martial behavior from infancy to the preschool years. Journal of Family Psychology, Vol. 18, No. 1: 194-207.

Shapiro, A.F. & Gottman, J.M. 2005. Effects on marriage of a psycho-communicative-educational intervention with couples undergoing the transition to parenthood: Evaluation at 1-year post intervention. Journal of Family Communication, 5: 1-24.

Shapiro, A.F., Guttmann, J.M, and Carrere, S. 2000. The baby and the marriage: Identifying factors that buffer against decline in marital satisfaction after the first baby arrives. Journal of Family Psychology, Vol. 14, No. 1: 59-70.

Silliman, B., Stanley, S.M., Coffin, W., Markman, H.J., & Jordan, P.L. 2002. Preventative interventions for couples. In H.A. Liddle, D.A. Santisteban, R.F. Levant, & J.H. Bray, Family Psychology Science-Based interventions, pp. 123-145, Washington, DC: American Psychological Association.

Smock, Pamela J. and Wendy Manning. 2004. Living Together Unmarried in the United States: Demographic Perspectives and Implications for Family Policy. Law and Policy, 26: 87-117.

Spain, D. and Bianchi, S.M. 1996. Balancing act: Motherhood, marriage and employment among American women. Russell Sage Foundation: New York.

Sprecher, S., Metts, Burleson, B., Hatfield, E., & Thompson, A.  1995.  Domains of  expressive interaction in intimate relationships:  Associations with satisfaction and  commitment.  Family Relations, 44, 203-210.

Sprecher, S., & Felmlee, D.  1992.  The influence of parents and friends on the quality and stability of romantic relationships:  A three-wave longitudinal investigation. Journal of Marriage and the Family, 54: 888-900.

Stanley, S.M. 2003. Assessing couple and marital relationships: Beyond form and toward a deeper knowledge of function. Paper presented at Healthy Marriage Interventions and Evaluation Symposium of the Measurement Issues in Family Demography Conference, Washington D.C.

Stanley, S.M., S.L. Blumberg, and H.J. Markman. 1999. “Helping couples fight for their marriages: The PREP approach.” Pages 279-303 in R. Berger and M.R. Hannah (eds.), Preventive approaches in couples therapy. New York: Brunner/Mazel.

Stanley, S.M. and Markman, H.J. 1992. Assessing commitment in personal relationships. Journal of Marriage and the Family, 54: 595-608.

Stith, S.M, Smith, D.B, Penn, C. Ward, D. and Tritt, D. 2004. Risk factor analysis for spouse physical maltreatment: A meta-analytic review. Journal of Aggression and Violent Behavior, Vol. 10, 65-98.

Stoneman, Z., Brody, G.H, Churchill, S.L, and Winn, L.L. 1999. Effects of residential instability on Head Start children and their relationships with older siblings: Influences of child emotionality and conflict between family caregivers. Child Development, Vol. 70, No. 5, 1246-1262.

Sullivan, K.T., Pasch, L.A., Eldridge, K.A., & Bradbury, T.N. 1998. Social support in marriage: Translating research into practical applications for clinicians. The Family Journal: Counseling and Therapy for Couples and Families, 6, 263-271.

Taj, N., Devera-Sales, A., and Vinson, D.C. 1998. Screening for problem drinking: Does a single question work? Journal of Family Practice, Vol. 46, 328-335.Thomson, E. and Colella, U. 1992. Cohabitation and Marital Stability: Quality or Commitment? Journal of Marriage and the Family, Vol. 54, No. 2: 259-267

Thornton, A. and L. Young-DeMarco. 2001. “Four decades of trends in attitudes toward family issues in the United States: The 1960s through the 1990s.” Journal of Marriage and Family 63, 4: 1009-1037.

Tucker, M.B. 2000. Marital values and expectation in context: Results from a 21-city survey. In L. Waite, C. Bacharach, M. Hindin, E. Thomson, and A. Thornton (Eds.). The Ties that Bind: Perspective on Marriage and Cohabitation. Aldine de Gruyter, 166-187.

Turner, C.F., Ku, L, Rogers, S.M., Lindberg, L.D., Pleck, J.H., Sonenstein, F. L. 1998. Adolescent sexual behavior, drug use, and violence: Increased reporting with computer survey technology. Science, Vol. 8: 867-873.

Umberson, D., M.C. Chen, J.S. House, K. Hopkins, and E. Slaten. 1996. The effect of social relationship on psychological well-being: Are men and women really so different? American Sociological Review, 61: 837-857.

U.S. Department of Health and Human Services, National Institute on Alcohol Abuse and Alcoholism. 2000. “Highlights from the Tenth Special Report to Congress: Health risks and benefits of alcohol consumption.”  Alcohol Research and Health 24, 1. Washington, DC: U.S. Government Printing Office. Available on the World Wide Web at www.niaaa.nih.gov/publications/arh24-1/toc24-1.htm.

U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration (SAMHSA), Office of Applied Studies. 2003. 2003 National Survey on Drug Use & Health:  Results. Washington, DC: U.S. Government Printing Office. Available on the World Wide Web at www.drugabusestatistics.samhsa.gov/NHSDA/2k3NSDUH/2k3 results.htm#ch2.

Van Lange, P. A. M., Rusbult, C. E., Drigotas, S. M., Arriaga, X. B., Witcher, B. S. & Cox, C. L. 1997. Willingness to sacrifice in close relationships. Journal of Personality and Social Psychology, 72: 1373-1395.

Waite, L. J., & Gallagher, M. 2000. The case for marriage: Why married people are happier, healthier, and better off financially. New York: Doubleday.

Walker, A.J. and Thompson, L. 1983. Intimacy and intergenerational aid and contact among mothers and daughters Journal of Marriage and the Family, Vol. 45, No. 4: 841-849.

Webster-Stratton, C. and Taylor, T. 2001. Nipping early risk factors in the bud: Preventing substance abuse, delinquency, and violence in adolescence through intervention targeted at young children (0-8). Prevention Science, 2: 165-191.

Wheaton, B. 1985. “Models for the stress-buffering functions of coping resources.” Journal of Health and Social Behavior 26: 352-364.

Whitton, S. Stanley, S.M., Markman, H.J. 2002. Communication, conflict, and commitment: Insights on the foundations of relationship success from a national survey. Family Process, 41(4), 659-675.

Yamaguchi, K. and D.B. Kandel. 1985. “On the resolution of role incompatibility: A life event history analysis of family roles and marijuana use.” American Journal of Sociology 90: 1284-1325.

Yoshioka, M., E. Thomas and R. Ager. 1992. “Nagging and other drinking control efforts of spouses of uncooperative alcohol abusers: Assessment and modification.” Journal of Substance Abuse 4: 309-318.

1 The Building Strong Families (BSF) project is another multi-site random assignment evaluation funded by the Department of Healthy and Human Services in 2002 as part of the Healthy Marriage Initiative. BSF is an initiative to develop and evaluate programs designed to help interested unwed parents achieve their aspiration for healthy marriage and a stable family life.

2 Throughout this document, the age range 0 to younger than 15 years old includes the children of women who are pregnant at random assignment.

3 The cognitive testing in Oklahoma, Texas, and Washington, DC was conducted in three iterative rounds with samples of nine or fewer participants. Different measures were used in each round of the cognitive testing and none of the respondents to the cognitive testing will be part of the actual study sample. Because the samples for the cognitive testing included less than ten respondents, it is our understanding that these efforts do not require a separate OMB review and approval process.

4 A targeted sample size of 306 was selected for the observational study, so that the resulting respondent sample would be approximately one-third of the survey respondent sample. This is based on the assumption that 90 percent of respondents who completed the survey will agree to participate in the observational study, bringing the respondent rate for the observational study to approximately 72 percent.

5 We are currently exploring whether sample sizes will allow us to collect father-child interactions for portions of the observational study sample. However, doing so will not increase the burden to participants as the total number of respondents will be the same.

58