15 Month Survey - Males

APPENDIX E
SUPPORTING DOCUMENTATION FOR 15-MONTH FOLLOW-UP SURVEY

QUESTION-BY-QUESTION JUSTIFICATION OF 15-MONTH SURVEY QUESTIONS
Question #

Constructs/Items

Justification

IN1-IN7,
FS1-FS4

Information for
proper identification
of sample members

This information ensures that we are speaking to the right person.

FS5-FS9

Whether and when
BSF focal child was
born

This demographic information helps to identify the BSF focal child and
clarify the age of the BSF child. In addition, the timing of the second BSF
follow-up survey will be determined by the age of the BSF child
(conducted when the child is 36 months old), making the child’s date of
birth an essential piece of information.

FS10-FS13,
FS17.1

BSF focal child’s
name

This information will be used to identify the BSF focal child for
subsequent follow-ups and to fill in the BSF child’s name in later survey
questions.

FS17

BSF focal child’s
gender

This key demographic characteristic has been linked to the satisfaction and
stability in the parents’ relationship (Lundberg and Rose 2003). It can be
used for subgroup analysis and to identify the BSF focal child for
subsequent follow-ups.

FS17.3FS17.6

BSF child’s birth
weight

This measure is a good indicator of the child’s health at birth. It will be
used for subgroup analysis. We will examine whether participants whose
children are born with low birth weight benefit more or less from BSF.

FS19

BSF partner’s name

This information will be used to fill in the BSF partner’s name in later
survey questions.

FS25-FS26

Marital and
relationship status of
BSF couple

A central goal of BSF is to encourage healthy relationships and marriage
among participants. Therefore, documenting the current status of the BSF
couple’s romantic relationship is essential to the BSF impact analysis.
These questions are adapted from the Fragile Families and Child WellBeing Study, Surveys of New Parents.

FS26.1FS26.3

When and why
romantic relationship
with BSF partner
ended

Since encouraging healthy relationships and marriage is a key goal of the
program, it is important to understand when and why the romantic
relationship between the BSF couple ended. These questions were adapted
from the Fragile Families surveys.

FS27-FS29

Likelihood of
marriage with BSF
partner

It is important to examine these measures in the BSF impact analysis
because they are key indicators of movement toward marriage. Questions
involving these measures were asked in the Fragile Families surveys.

FS33

Whether living with
BSF partner

Whether the BSF couple is sharing a residence is an important element of
the status of their romantic relationship and is therefore an important
measure for the BSF impact analysis. This question is from the Fragile
Families and Child Wellbeing study.

FS33.2

Frequency of contact
with BSF partner

Frequency of contact with the BSF partner is another important element of
the couple’s relationship status and therefore an important measure for the
BSF impact analysis. This question is from the Fragile Families and Child
Wellbeing study.

E-3

Question #

Constructs/Items

Justification

FS33.2.1FS33.2.3

Reasons for
romantic couples not
seeing each other

These questions will clarify the status of the BSF couples relationship in
two circumstances: (1) when they are married and living apart and (2)
when they are romantically involved but not seeing each other regularly.
To understand the couple’s relationship status fully, it is important to
determine whether these situations represent voluntary or involuntary
separations (due to military service or incarceration, for example). These
questions will determine whether these separations are involuntary and, if
they are, the reasons for these involuntary separations.

FS37-FS40

Whether in a new
romantic relationship
and the status of that
relationship

Understanding the nature of new romantic relationships is important to the
BSF impact analysis because of its implications for child well-being.
Research suggests that living with both biological parents is generally
advantageous for children but that living with a parent and his or her new
partner is not (McLanahan and Sandefur 1994). In fact, exposure to a
parent’s new romantic partner can put the child at risk for adverse
outcomes (Radhakrishna et al. 2001). Asking about new romantic partners
will allow us to examine whether BSF had an impact on the likelihood that
participants entered into these new relationships. Similar questions were
asked in the Fragile Families surveys and in Marriage in Oklahoma: 2001
Baseline Statewide Survey on Marriage and Divorce.

FS42FS43.1

Living arrangements
of BSF focal child

BSF aims to increase the likelihood that children will live with both
biological parents, since this family structure has been shown to have
positive effects on child well-being (McLanahan and Sandefur 1994).
Examining whether BSF has this effect on the family structure of
participants requires us to ask questions about the BSF child’s living
arrangements. Similar questions were used in the Fragile Families and
Child Wellbeing Study.

FS45FS46.1

Whether any contact
with BSF child in
past year/month

These questions are needed for skip logic for questions about recent
interactions with the BSF child (CO2-CO5).

FS50-FS52

Number of children
born or conceived
after random
assignment; whether
BSF partner is the
other parent

These questions will allow us to examine BSF potential impacts on
subsequent childbearing and multiple partner fertility. Multiple partner
fertility has been shown to have negative consequences for child wellbeing (McLanahan and Sandefur 1994; Radhakrishna et al. 2001; Carlson
and Furstenburg 2006; Harknett and Knab 2005). It is hoped that BSF will
reduce multiple partner fertility by promoting fidelity and the likelihood
that participating couples stay together. These questions were drawn from
Fragile Families follow-up surveys.

FS53FS54.2

Other members of
the household

The questions on the number of children in the household will allow us to
measure family size for determining poverty status. Poverty has been
shown to have adverse effects on a wide array of child outcomes (BrooksGunn and Duncan 1997; Mayer 1997). Poverty is therefore is an important
aspect of child well-being and an important outcome to examine in the
BSF impact analysis.
The questions on adults in the household will indicate other financial
supports available to the BSF child through co-resident relatives.
Collecting this information will allow us to examine whether BSF has had
any impact on the availability of this type of support.

E-4

Question #

Constructs/Items

Justification

FS55-FS68

Marriage start and
end dates

These questions will allow us to examine the amount of time respondents
have been married during the follow-up period. The questions will also
allow us to examine whether subsequent births occurred within a marriage
or not, making it possible to examine the program’s effects on nonmarital
childbearing. In addition, a complete marital history will allow us to
construct baseline variables for subgroup analysis based on whether and
how often respondents were married before random assignment. Questions
of this type were asked in the Current Population Survey Fertility and
Marital History Supplement.

CO1

Quality of coparenting
relationship

These questions will allow us to examine BSF’s potential impacts on the
BSF couple’s co-parenting relationship. BSF may affect co-parenting by
improving the couple’s communication skills and emphasizing that
parenting is a shared task. Questions “a” through “j” are from the
Parenting Alliance Measure, a well-validated scale developed by Richard
Abidin (Abidin and Brunner 1995). Other co-parenting measures (“m”
through “s”) were developed by Child Trends for ACF’s Healthy Marriage
Initiative studies.

CO2-CO3

Time spent with BSF
child in the past
month

An important aspect of child well-being is the quantity and quality of time
children spend with their parents. These measures have been used in
several large studies, including Fragile Families surveys, the Early Head
Start Research and Evaluation Project, and the Early Childhood
Longitudinal Study-Birth Cohort (ECLS-B).

CO4

Spanking of BSF
child

This item is a measure of harsh discipline. It is also a measure of child
maltreatment, a key aspect of child well-being. By improving conflict
management skills and overall parental well-being, BSF may reduce child
maltreatment and the use of harsh discipline. This measure was used on
Early Head Start follow-up surveys.

RR0

Attitudes toward
marriage

Attitudes toward marriage have been shown to be highly predictive of
whether low-income, unwed parents marry (Carlson et al. 2004). BSF may
encourage participants to enter into healthy marriages by changing their
attitudes toward marriage. These five items come from the Fragile
Families surveys, the 2003 Baseline Survey of Family Experiences and
Attitudes in Florida, and the Louisiana Fragile Families survey.

RR0.1

Whether friends are
ready to settle down

This item measures an aspect of social support. See discussion of social
support measures later in the table at WW56-WW62.

RR1

Happiness with
relationship with
BSF partner

Overall happiness and satisfaction is one of the most frequently used
measures of relationship quality. It is highly correlated with the likelihood
of later breakup (Karney and Bradbury 1995). In addition, interventions
with married couples similar to the BSF program have been shown to
improve relationship happiness and satisfaction in the short run (Carroll
and Doherty 2003; Markman et al. 1988; Wampler 1990).

E-5

Question #

Constructs/Items

Justification

RR2

Conflict and conflict
management

Poorly managed conflict is highly correlated with relationship dissolution
(Stanley 2003). In addition, high conflict between parents has been shown
to have adverse consequences for child well-being (Stanley 2003;
Cummings and Davies 1994; Cummings et al. 1991). For these reasons,
the BSF curriculum focuses largely on conflict management, making this a
particularly important outcome to examine in the impact analysis. The
conflict management items are drawn from three sources: (1) John
Gottman’s Sound Relationship House; (2) the Interpersonal Relationship
Scale developed by Bernard Guerney (Guerney 1977); and (3) the StanleyMarkman Relationship Dynamics Scale (Stanley et al. 2002). The StanleyMarkman scale (items RR2aa through RR2dd) was used in the Oklahoma
Marriage Survey.

RR1.1; RR4

Friendship, intimacy,
and supportiveness

Positive aspects of relationships—such as friendship, intimacy, and
supportiveness—have been shown to counteract some of the negative
effects of high conflict on romantic relationships (Huston and Chorost
1994). These positive aspects of relationships are also highly predictive of
whether couples remain together (Carlson et al. 2004). All BSF curricula
focus on building intimacy, supportiveness, and friendship in relationships,
making these aspects of relationship quality particular important to the
BSF impact analysis. Item RR1.1 measures the amount of time the BSF
couple spends together and is drawn from the National Survey of Families
and Households. Two items measure friendship: (1) RR4b drawn from
Gottman’s Sound Relationship House and (2) RR4c developed by Child
Trends for ACF’s Healthy Marriage Initiative studies. Four items measure
supportiveness and intimacy: items RR4q and RR4r, developed by Child
Trends for ACF’s Healthy Marriage Initiative studies, and items R4w and
R4x, from Fragile Families follow-up surveys. Item RR4t, on respect, is
from John Gottman’s Sound Relationship House. Item RR4y, on sexual
intimacy, was drawn from Marital Instability Over the Life Course.
Similar questions concerning sexual intimacy were also included in both
the Fragile Families surveys and the Florida marriage survey. Item RR4z
is designed to measure the supportiveness of friends and family members
for the relationship.

RR4

Commitment and
trust

Commitment and trust are important aspects of a successful relationship.
Higher levels of commitment have been shown to help couples weather
bad times and avoid breakup (Amato 2003). In addition, a lack of trust has
been shown to be a significant barrier to marriage for low-income couples
(Edin and Kefalas 2005). Our three items related to commitment (RR4o,
RR4p, and RR4v) make up the interpersonal commitment scale developed
by Scott Stanley. These items were used in the Oklahoma Marriage
Survey. Our three items on trust (RR4i, RRj, and RR4n) are drawn from
the Dyadic Trust Scale developed by Larzelere and Huston (1980), which
has been widely used in many surveys.

RR8-RR11

Infidelity and
cheating

Infidelity has been found to be a major obstacle to marriage for unwed
parents (Edin and Kefalas 2005; Smock and Manning 2003). The BSF
curriculum aims to address this issue by discussing with participating
couples the importance of fidelity and trust in building a healthy
relationship. Several large surveys— such as the Study of Marital
Instability Over the Life Course, the Louisiana Fragile Families Study, and
the Florida marriage survey—have included similar questions on infidelity.

E-6

Question #

Constructs/Items

Justification

RR14RR15.1

Domestic violence

The BSF intervention aims to improve relationship quality and increase the
likelihood that couples enter into a healthy marriage. The absence of
violence is a key characteristic of a healthy romantic relationship.
Therefore, to fully assess BSF’s success in achieving its goal of
encouraging healthy relationships and marriage, it is necessary to collect
information on intimate partner violence. These questions are drawn from
the revised Conflict Tactic Scale (CTS2), the most widely used tool for
measuring intimate partner violence in research studies (Strauss and
Douglas 2004). The CTS2 has been both well validated and shown to have
good internal consistency (Strauss et al. 1996). Versions of these CTS
questions have been used in many surveys, including the National Family
Violence Survey, the National Violence Against Women Survey, and
surveys conducted in six states as part of the ASPE-funded TANF
Caseload Project.

WB0

Parental Aggravation
Scale

High levels of parental stress and aggravation are associated with poor
cognitive and socio-emotional development in young children (McGroder
2000). BSF may reduce parental stress and aggravation by encouraging
parents to support each other in their parenting roles. These four items
represent the Aggravation in Parenting Scale used as part of the National
Survey of America’s Families (NSAF), a large survey of low-income
families. This scale has been shown to have good psychometric properties,
such as internal reliability and construct validity (Ehrle and Moore 1999).

WB1-WB3

Symptoms of
depression

Parental depression has been shown to adversely affect child outcomes
(Gelfand and Teti 1990, Downey and Coyne 1990). Given BSF’s ultimate
goal of improving child well-being, the link between parental depression
and child well-being makes this outcome particularly relevant. BSF may
reduce symptoms of depression among participants by reducing stress and
conflict in relationships. These questions represent the 20-item Centers for
Epidemiologic Studies Depression Scale (CES-D), a widely used measure
with well-established psychometric properties (Radloff 1997). The CES-D
has been used as part of many large surveys, including the survey in the
Early Head Start Evaluation, the National Longitudinal Survey of Youth,
and the Project on Devolution and Urban Change.

WB4-WB6

Alcohol and drug
use

Substance abuse and addiction can have major negative effects on the wellbeing of individuals and their families. If BSF improves relationship
quality and stability, it may also reduce substance abuse among
participants. Collecting information on alcohol and drug use will allow us
to examine whether BSF has such an effect. The question we include on
binge drinking was developed by Henry Wechsler and is recommended as
a screening tool by the National Institute on Alcohol Abuse and
Alcoholism (Wechsler et al. 1995; Wechsler 1998). It has been used in
several large national surveys, including the National Survey on Drug Use
and Health and the Youth Risk Behavior Surveillance System. The two
questions on functional impairment resulting from substance use come
from Fragile Families surveys.

E-7

Question #

Constructs/Items

Justification

SE1-SE20

Receipt of services
since random
assignment

Asking program and control group couples identical questions in the
follow-up surveys about service receipt will be an important part of the
BSF impact analysis. Their responses will allow us to understand the
“counterfactual”—what services would have been received in the absence
of BSF—and how the kind and amount of services actually received by
BSF participants differed from what they would have received in the
absence of the intervention. Understanding the additional services received
by BSF participants will help us understand and interpret BSF’s impacts on
other key outcomes. For example, if impacts are modest, we will be able
to understand whether this result is a function of the fact that services
actually do have little effect on key outcomes or that the intervention
offered few services that are distinct from those already available in the
community. Most large random assignment impact studies include
questions of this type. For example, the Early Head Start and Rural
Welfare-to-Work evaluations included similar questions on service receipt.

PA1-PA1.1

Paternity
establishment

Establishing paternity is an important step in ensuring that unwed fathers
provide financial support for their children. If BSF augments the extent to
which participating fathers are committed to their children, it may increase
the rate of paternity establishment. In addition, if BSF increases the
likelihood that paternity is established through a voluntary process, this
may improve relationships between parents and between fathers and their
children. Similar questions on paternity establishment have been asked in
several large national surveys, including the Fragile Families and Child
Well-Being Study and the National Survey of America’s Families.

PA2PA11.11

Formal child support
and other financial
support provided by
nonresidential
parents

The level of financial support provided by parents is an important element
of child well-being. Children born to unwed parents are at high risk of
receiving little financial support from their parents, particularly their
fathers (McLanahan and Sandefur 1994; Rangarajan and Gleason 1998).
BSF may increase the financial support provided to the children of
participating couples—primarily by increasing the likelihood that the
couples remain together but also by increasing the likelihood that
nonresidential parents provide financial support for their children. Similar
questions on material support provided by nonresidential parents were
asked as part of the Fragile Families Study and the Early Childhood
Longitudinal Study, Birth Cohort (ECLS-B).

PA12

Level of material
support for BSF
child provided by
father

We are including a measure of father’s material support provided to the
child that can be asked of all parents—even if the parent lives with the BSF
child. This measure has the advantage of being defined for all respondents,
so we can compare all program-group couples to all control-group couples,
preserving the advantages of random assignment. Measures of this sort
also allow us to examine whether the income of resident fathers is indeed
available to the BSF child and whether the intervention has any effect on
this availability.

E-8

Question #

Constructs/Items

Justification

WW1WW30

Family income in the
past month

Family income and poverty are important determinants of child well-being
(Brooks-Gunn and Duncan 1997; Mayer 1997). BSF aims to enhance
child well-being by improving the parental relationship and the likelihood
that the parents remain together as a couple. Since two-parent families
generally have higher incomes than single-parent families, increases in
family income may be an important avenue through which BSF improves
child outcomes. Data on family income have been collected in many
national surveys, including NSAF and Fragile Families. The particular
questions we use are drawn from the Work First New Jersey study, a large
longitudinal study of welfare recipients.

WW36WW37.2

Sharing resources
and expenses

These measures will help us understand the extent to which the income of
resident parents (particularly fathers) is available to other family members
and whether BSF has had any effect on this availability. Similar questions
concerning resource and expense sharing were asked as part of the Fragile
Families and Child Wellbeing Study.

WW32,
WW33,

Bank accounts, car,
and home ownership

These items are measures of asset accumulation. Research has shown that
couples in more committed and stable relationships are more likely to save
and plan for the future, suggesting that BSF may have an effect on these
outcomes. Similar questions have been included in many surveys of lowincome populations, including NSAF, Fragile Families, and Rural Welfareto-Work.

WW53

Material hardship

Measures of material hardship are a useful addition to income and poverty
measures because they reflect a broader concept of economic well-being
that is not captured by income or poverty status and that takes into
consideration other factors that affect economic well-being, such as wealth,
debt, and access to credit (Ouelette et al. 2004). Analyzing BSF’s impacts
on these measures will help to complete the picture of the program’s
overall effect on economic well-being. The material hardship measures we
are including in the BSF survey have been used in several other national
studies of low-income families, including Rural Welfare-to-Work and
Fragile Families.

WW54WW55.2.1

Health insurance
coverage of
respondent and BSF
child

Given BSF’s possible effects on family structure and family income, it is
possible that the program may have effects on the level and type of health
insurance coverage BSF participants and their children have. Similar
questions have been asked in ECLS-B, Fragile Families, and NSAF.

WW55.3

General health status

A large body of evidence points to a causal link between a happy marriage
and health (Wilson and Oswald 2005). Given this link and the fact that
BSF aims to improve relationship quality, it is possible that the
intervention may have positive effects on health. This general health
question has been widely used in many surveys of low-income families,
including ECLS-B, Fragile Families, and NSAF.

WW38WW40

E-9

Question #
WW56WW62,
RR0.1,
RR4z

Constructs/Items
Social support

Justification
Social support has consistently been shown to have a powerful impact on
health and well-being (House et al. 1988; Turner and Turner 1999).
Moreover, social isolation and a lack of support networks can be common
among low-income families (Edin and Kafalas 2005). Attending BSF
group sessions may increase participants’ sense of social support, if bonds
are formed with other members of the group.
There are three main types of perceived social support that may be affected
by BSF: (1) emotional support; (2) tangible support ; and (3) validation
support. Emotional support, or having close, confiding relationships with
others, is measured by items WW60 and WW62. Tangible support, or
having access to practical help like emergency child care or a small loan, is
measured by items WW56 to WW59. Validation support, or having a
social network that makes one feel accepted or normal, is measured by
items RR0.1 and RR4z. Similar questions on social support were included
in Fragile Families surveys and in surveys conducted as part of the Work
First New Jersey evaluation.

WB9-WB30

Involvement with
the criminal justice
system

Recent research suggests that a history of incarceration and involvement in
the criminal justice system may be fairly common among fathers in the
BSF target population (Western 2004). Parental incarceration has major
negative effects on child and family well-being, reducing the financial and
other support otherwise provided by parents for their children and families.
BSF may reduce criminal involvement through its potential effects on
relationship stability and quality. Information about incarceration before
random assignment will allow us to examine whether BSF has differential
impacts depending on whether the parents have a criminal history. Similar
questions have been included in other large national studies, such as
Fragile Families and the National Job Corps Study.

CC1-CC2.2

Country of origin

Several BSF sites are likely to serve a substantial number of immigrants,
who may face a set of cultural and legal barriers to marriage that is
different from what other BSF participants face. Moreover, cultural
differences may make the BSF intervention more or less effective for
certain immigrant groups. Therefore, information on country of origin will
be important both for describing the population served by BSF and in
examining whether the program’s effectiveness varies for certain groups.
Similar questions were asked in NSAF and Fragile Families.

BP1-BP3

Structure of family
of origin

Research suggests that whether individuals lived with both their biological
parents while growing up is highly predictive of their likelihood of
marriage (Carlson et al. 2004; Wood et al. 2003). Therefore, these
measures will be useful in the BSF impact analysis to create subgroups.
Similar questions were included in the Fragile Families surveys.

BP7-BP8

History of physical
or sexual abuse as a
child

A history of physical and sexual abuse during childhood has been shown to
reduce the likelihood of entering into and sustaining healthy relationships
and marriages as an adult (Cherlin et al. 2004). This research also
indicates that a history of childhood abuse is fairly common among lowincome populations. For these reasons, those with a history of childhood
abuse will be an important subgroup to examine as part of the BSF impact
analysis. These questions were deemed too sensitive to ask at BSF sample
intake, making it necessary to gather this information on follow-up
surveys. These two questions are from surveys conducted as part of
Welfare, Children, and Families: A Three-City Study.

E-10

Question #

Constructs/Items

Justification

BP9

Age of first
intercourse

The BSF curriculum aims to build commitment and trust among unmarried
couples with young children as a means of strengthening and preserving
their romantic relationships. Individuals with a large number of sexual
partners prior to entering the program may have difficulty establishing the
necessary level of commitment and trust to build a healthy and lasting
romantic relationship. Therefore, the number of sexual partners prior to
random assignment is a variable of potential interest for subgroup analysis.
Because of recall difficulties, however, asking about the number of sexual
partners prior to random assignment on the 15-month follow-up survey is
not practical. Therefore, we will ask instead about the age of first
intercourse, which has been shown to be a good proxy for the number of
sexual partners (USDHHS 1997). This question is drawn from the National
Survey of Family Growth.

BP10

Number of sexual
partners since
random assignment

Children who are exposed to the new romantic partners of their parents are
placed at increased risk of abuse and other adverse outcomes
(Radhakrishna et al. 2001). It is hoped that by increasing the likelihood
that participating couples remain together, BSF will reduce the exposure
that their children have to the new romantic partners of their parents.
Therefore, the number of sexual partners since random assignment is an
important variable to examine as part of the impact analysis. This question
is drawn from the National Survey of Family Growth.

LO1-LO18

Locating information

This information will be crucial to high response rates in the 36-month
survey.

E-11