Citation
Adult children accounts of parental infidelity and divorce

Material Information

Title:
Adult children accounts of parental infidelity and divorce associations with own infidelity, risky behaviors, and attachment
Creator:
Spence, Aaron Michael
Publication Date:
Language:
English
Physical Description:
1 electronic file : ;

Subjects

Subjects / Keywords:
Adultery ( lcsh )
Risk-taking (Psychology) ( lcsh )
Adult children ( lcsh )
Adult children ( fast )
Adultery ( fast )
Risk-taking (Psychology) ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Although there is a great deal of research that examines the effect that divorce has on children, there are very few studies that explore the impact that parental infidelity may have on this population. In this study, we examined undergraduates who reported information about parental infidelity and divorce and measured their own infidelity behaviors, attachment, and risk-taking. We hypothesized that participants with parents who were known to have committed infidelity and were divorced would have the highest rates of their own infidelity, more risk-taking behavior, and more insecure attachment styles. Data was analyzed using ANOVAs and regression. Results partially supported the initial hypotheses. Specifically, participants whose parents engaged in infidelity and remained married had the highest rates of their own infidelity. Additionally, participants with divorced parents had the highest rates of risk-taking. Parental infidelity may serve as a model promoting children to also engage in this behavior in their own relationships. However, more work needs to be conducted in this area in order to establish causal connections.
Bibliography:
Includes bibliographical references.
Statement of Responsibility:
by Aaron Michael Spence.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
859204629 ( OCLC )
1249894190 ( ISBN )
ocn859204629
Classification:
HQ519 .S64 2012 ( lcc )

Auraria Membership

Aggregations:
Auraria Library
University of Colorado Denver

Downloads

This item has the following downloads:


Full Text
ADULT CHILDRENS ACCOUNTS OF PARENTAL INFIDELITY AND DIVORCE:
ASSOCIATIONS WITH OWN INFIDELITY, RISKY BEHAVIORS, AND
ATTACHMENT
by
Aaron Michael Spence
B.A., Bowling Green State University, 2009
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Masters of Arts
Clinical Psychology Program
2012


This thesis for the Master of Arts degree by
Aaron Michael Spence
has been approved for the
Clinical Psychology Program
by
Peter Kaplan, PhD, Chair
Elizabeth Allen, PhD, Advisor
Michael Zinser, PhD
Joan Bihun, PhD
Date: 11/05/2011


Spence, Aaron Michael (M.A., Clinical Psychology Program)
Adult Childrens Accounts of Parental Infidelity and Divorce: Associations with Own
Infidelity, Risky Behaviors, and Attachment
Thesis directed by Assistant Professor Elizabeth Allen, Ph.D.
ABSTRACT
Although there is a great deal of research that examines the effect that divorce has
on children, there are very few studies that explore the impact that parental infidelity may
have on this population. In this study, we examined undergraduates who reported
information about parental infidelity and divorce and measured their own infidelity
behaviors, attachment, and risk-taking. We hypothesized that participants with parents
who were known to have committed infidelity and were divorced would have the highest
rates of their own infidelity, more risk-taking behavior, and more insecure attachment
styles. Data was analyzed using ANOVAs and regression. Results partially supported
the initial hypotheses. Specifically, participants whose parents engaged in infidelity and
remained married had the highest rates of their own infidelity. Additionally, participants
with divorced parents had the highest rates of risk-taking. Parental infidelity may serve
as a model promoting children to also engage in this behavior in their own relationships.
However, more work needs to be conducted in this area in order to establish causal
connections.
The form and content of this abstract are approved. I recommend its publication.
Approved: Elizabeth Allen, Ph.D.
m


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION........................................................... 1
Impacts of Divorce................................................... 1
Health Compromising and Risky Behaviors and Divorce.............4
Attachment Styles and Divorce...................................7
Dissolution of Marriage among Adult Children of Divorce....... 10
Impact of Infidelity................................................ 11
Risky Behaviors and Infidelity................................ 12
Attachment and Infidelity..................................... 14
Parental Infidelity and Adult Childrens Own Infidelity....... 16
II. METHODS............................................................... 13
Participants.........................................................19
Procedures...........................................................20
Measures.............................................................21
Parental Infidelity and Divorce................................21
Attachment Style...............................................21
Risky Behaviors................................................22
Own Infidelity.................................................23
Hypotheses...........................................................23
Overview of Data Analytic Approach...................................24
III. RESULTS................................................................26
Parental Status and Own Infidelity...................................26
IV


Parental Status and Risky Behaviors............................28
Parental Status and Attachment.................................30
Parental Infidelity and Gender Interactions....................32
IV. DISCUSSION........................................................34
REFERENCES............................................................39
APPENDIXES............................................................43
A. Announcement for Study Participation........................43
v


LIST OF TABLES
Table
II. 1 Recoded values of open-ended item responses into 0-5 scale score.............22
III. 1 Cell sizes for parent and participant genders regarding infidelity behavior.33
vi


LIST OF FIGURES
Figure
III. 1 Relationship between Gender of Parent and Child with Infidelity


CHAPTER I
INTRODUCTION
Divorce may have significant impacts on the adults involved in the divorce, as
well as important intergenerational effects on the parents offspring. Related to this,
parental infidelity, which may or may not end in divorce, may have important additional
impacts on children. There are major gaps in the literature on this topic, however, but
some noteworthy results have still been found. This paper will seek to discuss and
integrate research findings regarding important intergenerational effects of divorce on
children, factors that may influence these effects, and intergenerational impacts of
infidelity.
Impacts of Divorce
Divorce is extremely prevalent in the United States today, and it is estimated that
50% of children in America will experience parental divorce (Lansford, 2009). This
overwhelming statistic may have far-reaching implications, as marital dissolution may
impact children in a variety of ways. Some researchers have hypothesized that divorce
may have important psychosocial effects on children, resulting in insecure attachment
styles, behavioral problems, health-compromising behaviors, cognitive and social deficits,
psychological distress, poor academic achievement, and poor self-concepts (Crowell,
Treboux, & Brockmeyer, 2009; Huure, Junkkari, & Aro, 2006; Lansford, 2009; Zill,
Morrison, & Coiro, 1993). Divorce may also impact children throughout adulthood, and
may lead to higher divorce rates among adult children whose parents have divorced.
1


A variety of studies throughout the years have provided empirical support for the
idea that divorce has a variety of impacts on children (Duncan & Hoffman, 1985;
Guidubaldi, Cleminshaw, Perry, & Mcloughlin, 1983; Lansford, 2009; Zill, Morrison, &
Coiro, 1993). These studies have also found that the impacts that children face are not
always readily apparent. Children may appear quite resilient to the impacts of divorce,
but may be suffering major emotional problems beneath the surface (Laumann-Billings &
Emery, 2000). Laumann-Billings and Emery (2000) also provided evidence that the
distress felt by children from divorced families may be considerable decades later; some
children even speculated that they might be a different person today if their parents had
not divorced. Some caution needs to be used in interpreting these findings, however, as
Clarke-Stewart, Vandell, McCartney, Owen, and Booth (2000) hypothesized that parental
variables (i.e. income, depression, education, ethnicity, and childrearing beliefs) may
mediate the effects between divorce and the consequences felt by children.
Other studies that have examined the impacts of divorce on children failed to find
any significant results (Baydar, 1988; Mechanic & Hansell, 1989). These findings raise
noteworthy concerns regarding the divorce literature; some of the studies were
overpowered with very large sample sizes, and, yet, few significant differences were
found (Baydar, 1988). Further, the significant differences that were found between
children from divorced and intact families had negligible effect sizes. Results from these
studies also showed that neither recent nor early divorce had an impact on health
outcomes (Mechanic & Hansell, 1989).
To examine these contradicting findings, Amato and Keith (1991a) performed a
meta-analysis on over 13,000 children from preschool to college age. This technique
2


allowed Amato and Keith to look at effect sizes and account for differing design features
of 92 studies. The meta-analysis revealed that children from divorced families generally
experience lower academic achievement, poorer psychological adjustment, more
behavioral problems, more negative self-concepts, increased social difficulties, and more
relationship problems with their mothers and fathers (Amato & Keith, 1991a). These
results were echoed in multiple reports by the National Center for Health Statistics (2002;
2008), that stated that children from single parent households do worse in terms of
academic achievement, depression, and behavioral problems than children in two parent
households.
Amato and Keith (1991b) carried out an additional second meta-analysis that
analyzed over 80,000 adults to see if the impacts of divorce carry into adulthood. This
analysis revealed that adult children who experienced parental divorce had more
behavioral problems, less education, lower job status, impaired psychological well-being,
a lower standard of living, lower marital satisfaction, a heightened risk of divorce, and
worse physical health (Amato & Keith, 1991b). These findings illustrate major
intergenerational implications, and it is interesting to note how children from divorced
families are more likely to follow in their parents footsteps and have relationships that
end in divorce as well. It is also worth noting that Amato (2001) followed up these meta-
analyses a decade later and incorporated 67 studies conducted in the 1990s and found
similar results. This study, as well as others (e.g. Lansfords review (2009) that
examined the effects of divorce in more recent years), helped demonstrate that impacts of
divorce seem to be consistent and stable throughout time.
3


Amato (1994) urges not to be overly concerned with these findings, however, as
the decline in functioning is small, and there is a great deal of overlap in well-being
between children of divorced and intact families. Also, psychosocial impacts on children
as a result of divorce seem to be the exception rather than the rule, as most children do
not experience significant negative outcomes. Following this, another, more recent study,
which followed individuals over 15 years after their parents divorced also found that
divorce carried effects into adulthood (Angame-Lindberg & Wadsby, 2009). This study
demonstrated that adult children of divorce experienced more negative life events, which
resulted in difficult adjustment. However, these findings and effect sizes were small,
with minimal between group differences. The small differences found in these studies
are still important to note, nonetheless, so that clinicians know where to focus and can
work to try to make differences between children from divorced and intact families as
diminutive as possible.
Although Amatos (1994) review gives hope to children of divorce (the overall
difference in well-being between children whose parents have divorced and children from
intact families is small), there may be other variables that show a stronger impact on
children of divorce. Three particular variables of interest that have received attention in
the divorce literature are health compromising and risky behaviors, attachment styles, and
subsequent divorce of adult children, all of which seem to be impacted by parental
divorce. Each variable will be reviewed below.
Health-Compromising and Risky Behaviors and Divorce
Health compromising and risky behaviors encompass a wide range of behaviors
that may lead to many poor health outcomes and early morbidity and mortality.
4


Examples of these behaviors include smoking; heavy drinking; using illicit drugs; and
engaging in risky sexual behaviors such as having unprotected sex, having sex under the
influence of alcohol or drugs, or engaging in sexual activities with multiple partners; the
psychosocial implications of all of these behaviors have been documented thoroughly in
the literature. Currently, there have been a handful of studies conducted that have
examined the link between parental divorce and childrens involvement in risky health
behaviors.
Huure, Junkkari, and Aro (2006) conducted a 16-year longitudinal study that
examined the impacts of divorce on a variety of outcomes, including health-
compromising behaviors. The study examined 1,471 participants between the ages of 16
to 32 whose parents either divorced or did not. Results showed that daily smoking and
hazardous alcohol consumption were more prevalent in the children from divorced
families than those from intact families for both males and females. The findings of this
study also help provide evidence for the long-standing impacts that divorce can have well
into adulthood.
Another study conducted by Schwartz, Friedman, Tucker, and Tomlinson-Keasey
(1995) looked at archival data in order to examine mortality differences between children
from divorced and intact families. It was found that children from divorced families had
decreased longevity, and, on average, died four years earlier than those from intact
families; a finding that may be a result of increased engagement in health-compromising
behaviors. Martin, Friedman, Clark, and Tucker (2005) followed up this study to assess
potential mediators and moderators of the decreased longevity among children of divorce
finding, and found that smoking was the strongest mediator of this relationship; a finding
5


that provides further evidence that children of divorce are at an increased likelihood of
engaging in risky health behaviors.
Further evidence that children from divorced families engage in more risky health
behaviors than children from intact families comes from the work of Spruijt and
Duindam (2005). The authors used data gathered from the longitudinal study Utrecht
Study of Adolescent Development, which is a representative random sample of
adolescents and young adults (age range from 12-30) to examine differences between
males and females from divorced and intact families. The results showed that males from
divorced families engaged in more risky behaviors (defined as smoking, drinking, and
using soft drugs), and had a higher number of sexual partners.
There are a variety of theories as to why children of divorce may engage in
health-compromising and risky behaviors more so than children from intact families.
One theory that attempts to explain the link between divorce and health-compromising
behaviors focuses on the disparities between children from divorced and intact families as
a result of parental divorce. It has been found that children from divorced families have
lower education, higher unemployment, and higher prevalence of depression and minor
psychiatric disturbance (Huure et al., 2006), all of which are predictive of health-
compromising behaviors (Bachman, et al., 2008; Downing-Matibag & Geisinger, 2009;
Schwinn, Schinke, & Trent, 2010). These findings suggest that divorce may increase an
adult childs propensity to engage in such behaviors through the mediating factors of
lower education, unemployment, and impaired mental health status.
Another popular theory as to why children of divorce may engage in more risky
health behaviors than their counterparts from intact families stems from a lack of parental
6


supervision and involvement. Children from divorce are often under the primary custody
of one parent and do not receive as much parental regulation as children who have two
parents to share the burden. Resources may also be strained after divorce, and adequate
child-care services are not available. This lack of sufficient supervision may give
children the opportunity to engage in risky behaviors and form bonds with peers who
engage in these behaviors as well.
An additional theory states that children may engage in risky behaviors in order to
cope with parental separation. Parental divorce creates a major shift in family dynamics,
and children may feel quite distressed from the change. Children may engage in poor
coping techniques, such as using drugs or engaging in risky sex, to deal with the loss and
displace any depression they may feel. It has also been widely documented that
adolescents may also engage in risky behaviors through acting out (Allison, &
Furstenberg, 1989; Lansford, 2009; Zill et al., 1993); a behavior that may serve as
another means to cope with parental separation.
Attachment Styles and Divorce
Attachment, broadly defined, is the way in which individuals cognitively
conceptualize their internal working models of self and others in interpersonal
relationships (Carranza, Kilmann, & Vendemia, 2009). Attachment theory was first
conceptualized by Bowlby (1969) in the context of infant/caretaker relationships, and was
based on the manner in which the infant interacted with and without his or her primary
caretaker. Attachment theory has also been extended to adult intimate relationships
(Hazan & Shaver, 1987) in order to assess how individuals function in and conceptualize
their romantic relationships.
7


The concept of attachment flows from two continuum of anxiety and avoidance
regarding interpersonal relationships. Anxiety is indicative of a desire for extreme
closeness in relationships, as well as a reliance of self-worth that is completely
maintained by others. At the same time, anxious attachment is colored with worries
about abandonment (Allen & Baucom, 2004). Conversely, individuals with avoidant
attachment are overly concerned with their autonomy and are uncomfortable with being
dependent or too close to others (Allen & Baucom, 2004). Put another way, individuals
with high anxiety are said to have a negative view of self, while individuals with high
avoidance are said to have a negative view of others (Brennan, Clark, & Shaver, 1998).
Stemming from the two dimensions of anxiety and avoidance are four separate
attachment categories that an individual can fall into. An individual that is high on the
anxious continuum and low on the avoidant dimension is said to have a preoccupied
attachment style, while an individual who is high on the avoidant continuum and low on
the anxious continuum is said to have a dismissive attachment style (Brennan et al., 1998).
Individuals who rank highly on both scales have a fearful attachment style; a style
theoretically based on a negative view of both others and self, so that extreme closeness
is desired but feared and avoided due to concerns regarding rejection (Fraley, Waller, &
Brennan, 2000). Finally, an individual who rates low on each scale is said to have a
secure attachment style with a positive view of self and others.
The measurement of attachment has changed and improved throughout the years.
Attachment has been measured in a variety of ways, including the use of observational
and interview techniques. However, it has become a quite popular and accepted method
to measure attachment from self-report scales (Brennan et al., 1998). Two of the
8


commonly used self-report measures in studies that assess attachment are the
Relationship Styles Questionnaire (RSQ) and the Experiences in Close Relationships
Inventory (ECRI; Fraley et al., 2000).
A myriad of studies have been conducted throughout the years that have assessed
how attachment may impact various outcomes, ranging from high-risk drinking (Molnar,
Sadava, DeCourville, & Perrier, 2010) to developing posttraumatic symptomatology after
a traumatic event (Sandberg, Suess, & Heaton, 2010) to infidelity (Allen & Baucom,
2004). These studies have primarily found that insecure attachment styles (preoccupied,
dismissive, and fearful) lead to poorer outcomes. Attachment styles that are developed in
childhood are also thought to influence adult attachment, and are believed to be relatively
stable throughout life. However, significant disruptions, changes, or events in an
individuals life may impact his or her attachment style and cause it to change (Hamilton,
2000). For this reason, many researchers have been interested in how divorce impacts
childrens attachment styles, especially because it has been tied to many adverse
outcomes. Further, it is easy to theorize how divorce may impact attachment styles by
disrupting parent/child relationships, and by potentially causing children to view
relationships with feelings of anxiety and avoidance.
Many studies have found that divorce may play an important role in affecting
individuals attachment styles, with divorce leading to more insecure attachment styles
among children (Beckwith, Cohen, & Hamilton, 1999; Crowell, Treboux, & Brockmeyer,
2009; Lewis, Feiring, & Rosenthal, 2000). In a study of 202 couples consisting of 404
individuals between the ages of 19 and 35, Jacquet and Surra (2001) also found that
women from divorced families had more distrust, dissatisfaction, and ambivalence in
9


their own relationships; all of which are characteristics of insecure attachment styles.
Men did not seem to be as impacted, however, and only seemed to demonstrate more
distrust and ambivalence if their partners parents were divorced; gender disparities of
this study and other findings will be discussed later. It seems clear in the literature that
the major shift in parental caretaking and family dynamics that characterize divorce is a
phenomenon that readily impacts childrens attachment styles.
Dissolution of Marriage among Adult Children of Divorce
There are multiple theories that speculate why adult children from divorced
families may be more likely to divorce themselves. One theory posits that modeling may
offer the potential explanation (Crowell et al., 2009). Parents who divorce may illustrate
maladaptive behaviors to maintain a committed marriage such as poor communication,
negative emotion, and withdrawal from conflict resolution (Crowell et al., 2009).
Subsequently, their children may model these behaviors in their own marriages,
characteristics that are likely to lead to their own divorce.
Another hypothesis states that parents may convey, intentionally or
unintentionally, positive beliefs about divorce and a shallow commitment to marriage
(Crowell et al., 2009). Children may foster these beliefs, and see divorce as an
acceptable alternative in their own lives. This hypothesis was supported in a large study
of 2,033 married individuals by Amato and DeBoer (2001); they found that it was
divorce itself that predicted the intergenerational risk of divorce, apart from any conflict
in the parents relationship. The authors concluded that individuals from divorced
families have a weaker commitment to the norm of lifetime marriage than children from
intact families (Amato & DeBoer, 2001). Children of divorce have also been shown to
10


understand that life goes on after divorce (Greenberg & Nay, 1982), are more accepting
of alternatives to marriage (Amato, 1988), and they also hold less optimistic views of
marriage (Franklin, Janoff-Bulman, & Roberts, 1990).
There is a plethora of research to support these hypotheses, and the
intergenerational transmission of divorce has been widely documented (Amato & DeBoer,
2001; Amato & Keith, 1991b; Teachman, 2002; Wolfinger, 2000). The consequence of
parental divorce also has a potent effect on the intergenerational transmission of divorce;
Amato and DeBoer (2001) note that parental divorce doubles the chances that their adult
children will also divorce. Multiple family disruptions may further exacerbate the
problems, as children who undergo multiple parental divorces and transitions may be at
an increased risk for divorce, behavioral problems, poor educational outcomes, and
unstable relationships (Cavanagh, Crissey, & Raley, 2008; Fomby & Cherlin, 2007; Ryan,
Franzetta, Schelar, & Manlove, 2009; Wolfinger, 2000). Overall, it seems safe to
conclude that parental divorce is a strong predictor of adult childrens own divorce.
Impacts of Infidelity
Infidelity, also known as extradyadic involvement (EDI), can be operationalized
as having sexual encounters with an individual outside of ones primary relationship in
which monogamy is expected. EDI is a prevalent phenomenon in romantic relationships,
and when discovered, can elicit a great deal of distress and conflict in both partners
(Allen, Atkins, Baucom, Snyder, Gordon, & Glass, 2005). Infidelity is also a common
precipitant of divorce (Amato & Rogers, 1997), and it is important to hypothesize how
EDI may have significant intergenerational impacts on children. However, there have
been very few empirical studies that have examined the intergenerational impacts of
11


infidelity, and, therefore, it is important to utilize past research from a related topic.
Drawing from the divorce literature, then, it would seem important to examine the
impacts that infidelity has on a variety of outcomes, namely childrens risky behaviors,
attachment styles, and own infidelity. Evidence for infidelitys impact on each outcome,
and theories of why these variables may be affected, will be considered in turn.
Risky Behaviors and Infidelity
There is currently no empirical evidence to support the link between parental
infidelity and childrens risky behaviors. However, a variety of sources have speculated
about infidelitys impact in this area. Wallerstein and Kelly (1996) theorized that
adolescents who find out about parental infidelity may regress and be extremely anxious
in the wake of their sexual and aggressive impulses; effects that could potentially lead to
acting out and serious behavior changes. Wallerstein and Kelly provided a case example
of a girl named Jean. Jean became sexually active at the age of 14, which coincided with
her discovery of her fathers affair. With her mother, Jean formed strategies for
punishing her father and fantasized about his sexual involvements. After her parents
divorce, she continually flirted with her older male teachers and soon began abusing
alcohol and drugs.
Sori (2007) also commented on the idea that infidelity may impact childrens
risky behaviors. She stated that older children may react to infidelity through
externalizing behaviors, and, for adolescents or young adults, this may mean using
alcohol, drugs, or sex to act out. Using sex as an outlet may especially be pronounced in
older children as they have a heightened sense of their own sexuality; increased acting
out, linked with insufficient parent/child communication about safe-sex practices as a
12


result of family disruption, may further lead to increased sexual risk behaviors such as
not using condoms (Hadley et al., 2009). Further, a clinical case study illustrated by
Lusterman (2005) illustrated the impact that infidelity had on a teenage girl. In the case,
the girl found out about her fathers infidelity, but kept it a secret at his request. However,
she began acting out and dressing provocatively; signs that she may be engaging in risky
behavior. Caution must be given in interpreting this and other case examples, as
generalizability may be limitied; however, these case studies and speculations provide
important direction and insight on how infidelity may impact childrens risky behaviors.
Empirical studies are needed to see if these results generalize and are a common finding
among children who experience parental infidelity.
There have been multiple theories developed to hypothesize why children may
engage in risky behaviors following parental infidelity. One hypothesis states that
infidelity causes a shattered superego in the child. The case study of Priscilla
(Wallerstein & Kelly, 1996) illuminated this idea vividly with her statement, The beliefs
which gave me the ability to effectively deal with life were blown apart for me by my
dads [infidelity]. Parental infidelity may undermine many of the ideals that the child
was taught by the parent, with the result that the child does not know what is right or
wrong anymore. This may lead the child to engage in behaviors that were previously
considered bad, such as drinking, smoking, doing drugs, or having sex.
Another hypothesis that links parental infidelity to risky behaviors in children
states that an affair may destroy a childs sense of security and trust, and they may no
longer see the world as safe or predictable (Sori, 2007). This shattered trust may lead a
child to seek security elsewhere in order to cope and feel safe. However, this search may
13


lead to risky behaviors such as drug use or sex. Additionally, adolescents may return this
shattered sense of trust with anger and hostility towards the parent(s) (Sori, 2007;
Wallerstein & Kelly, 1997). This may lead adolescents to engage in risky behaviors in
order to cope or to lash out at parents.
One additional theory states that children may engage in risky behaviors, namely
risky sexual behaviors, due to modeling. Children may believe that their parent(s) are
engaging in unsafe sexual practices, and model this behavior in their own relationships.
They may believe that this behavior is acceptable if their parent(s) do it and follows in
their footsteps.
Attachment and Infidelity
Theories regarding the impact of infidelity on attachment are relatively
straightforward. Generally, they state that infidelity may be a major disruption or shift in
an individuals life and can precipitate a change in the childs attachment style; usually
with the end result being an insecure attachment style. This may occur because infidelity
may diminish an individuals self-esteem and shatter trust and conceptions of intimacy
(Sori, 2007). Following this, Lusterman (2005) points out that the daughters trust in
boys was severely impacted by the discovery of her fathers infidelity, and all of her
subsequent relationships were affected. Further, children may no longer hold the same
beliefs about the importance of intimate relationships or the stability of partnerships (Sori,
2007). Therefore, seeing a parents infidelity may lead the child to develop insecure
attachment for a couple of reasons. Anxious attachment may occur due to a desperate
need for closeness, while maintaining an acute fear of abandonment. Conversely,
parental infidelity may lead to avoidant attachment, as an individual may take on a
14


negative view of others and avoid closeness. Although the literature relating to the
intergenerational impacts of infidelity is sparse, there have been two empirical studies
conducted that have assessed parental infidelitys impact on adult childrens attachment.
Walker and Ehrenberg (1998) conducted an advent study that examined the
intergenerational impacts of infidelity. This study sought to examine if a childs
perceptions of parental divorce had an impact on his or her attachment style. Walker and
Ehrenberg (1998) found that children who attributed their parents divorce to infidelity
were significantly more likely to have an insecure attachment style, which may have
important implications for their own intimate relationships later in life.
Another important study conducted by Platt, Nalbone, Casanova, and Wetchler
(2008) examined if parental conflict and infidelity could serve as predictors for adult
childrens attachment style and own infidelity. Platt et al. (2008) only found partial
evidence that parental conflict predicted attachment style, and they found no evidence
that parental infidelity predicted attachment. This seems to run contrary to Walker and
Ehrenbergs (1998) findings, but parental conflict may have been a confounding variable
in Walker and Ehrenbergs study. The two studies also used different measures to
determine attachment style, which may explain the difference.
Both of the studies conducted by Walker and Ehrenberg (1998) and Platt et al.
(2008) have important limitations that need to be noted. Both studies had relatively small
sample sizes and the participants consisted only of undergraduate students. This means
that smaller effects may not be able to be detected and that the samples may have been
unrepresentative. Another major limitation of these studies is the failure to include
participants whose parent(s) have committed infidelity, but remain married.
15


Parental Infidelity and Adult Childrens Own Infidelity
The intergenerational impact of parental EDI on adult childrens subsequent
infidelity is theoretically tied to the intergenerational effects of divorce. Parental
infidelity may lead children to develop positive views of infidelity and a shallow
commitment to being faithful to their partner. Subsequently, children may believe that
infidelity is not a big deal, and is a common and acceptable practice.
Other theories regarding the intergenerational transmission of infidelity are
similar to the impacts of parental EDI on risky behaviors. This theoretical link may be
tied to the idea that infidelity is a form of risky behavior. Thus, children may act out and
engage in risky sexual practices, including having multiple partners or committing
infidelity. The intergenerational transmission of infidelity may also be tied to theories of
risky behavior through modeling. Children may observe parental infidelity and take on
the belief that if my parents do it, it must be ok. Thus, children will see their parents
engaging in EDI and follow in their parents footsteps and actions.
Although few empirical studies have been conducted, there has been some
evidence concerning the intergenerational transmission of infidelity. Platt et al. (2008)
found that children that know their father has had EDI are more likely to engage in
infidelity themselves. When further examined, it was found that this relationship only
remained significant for boys. This may illustrate a modeling effect where boys know of
their fathers EDI, and subsequently find the behavior to be more acceptable. However,
this link needs to be studied much further in depth before conclusions can be made.
The impacts of parental divorce and conflict on children have been well
documented in a plethora of studies throughout the years. Research that focuses on the
16


impacts that infidelity has on children is in its infancy, however, and there is a paltry
amount of literature on this issue. The studies conducted by Walker and Ehrenberg
(1998) and Platt et al. (2008), as well as the case studies by Lusterman (2005) and
Wallerstein and Kelly (1997) provide helpful directions to future researchers interested in
exploring this new research topic. The findings regarding the intergenerational effects of
divorce should also be utilized, and the childs gender is important to include as a
potential moderator on the impacts of infidelity. Studies of divorce have also pinpointed
that children are generally affected in terms of risky behaviors, attachment, and
subsequent divorce, and infidelity research should focus on these outcomes as well.
Further, the divorce literature seems to agree that the impacts of divorce carry well into
adulthood, a finding that should be explored in the infidelity literature as well. A great
deal of research needs to be conducted to fully explain the intergenerational impacts of
infidelity, but with time and effort it can eventually be well understood.
A study that I propose to conduct would build off of the research already conducted
regarding the intergenerational impacts of infidelity. Currently, there is a discrepancy in
the research regarding the impacts of infidelity on attachment style, little investigation
regarding an adult childs subsequent infidelity with knowledge of parental infidelity, a
gap in the literature regarding the intergenerational impacts of infidelity on children
engaging in risky behaviors, and a lack of studies that use samples that include children
whose parents committed infidelity but remain together. Therefore, I propose to conduct
a study with four main purposes: (1) given the conflicting findings in the literature, to
further investigate the impact of parental infidelity on an adult childs attachment style,
(2) to explore if parental infidelity results in higher rates of subsequent infidelity
17


behaviors from adult children, (3) to examine whether or not parental infidelity leads to
children engaging in more risky behaviors, and (4) to assess the potential moderating
impact that the childs gender has on these outcomes. For all these research questions, I
will use the adult childrens self-report, thus, these links between parental infidelity and
current functioning will only relate to parental infidelity which is known by the
respondent.
18


CHAPTER II
METHODS
In order to conduct this study, I conducted secondary analyses using a data set
collected by Dr. Elizabeth Allen. An undergraduate sample (described below) was used
to test the hypotheses.
Participants
Participants in Dr. Allens study consisted of undergraduate students from a local
university campus (N= 484). The undergraduate students were recruited from
psychology courses, and were given the option to participate in the study for credit. The
age of the students ranged between 18 to 67 years (M= 21.53). Males constituted 29% of
the sample, while females composed the other 71%. The sample contained African
Americans (5.9%, N= 26), American Indian or Alaskan Natives (.5%, N= 2), Pacific
Islanders (.9%, N= 4), whites (65.8%, N= 291), Asians (15.9%, N= 70), and individuals
who identify with another racial group (5.7%, N= 25) or more than one racial group
(4.5%, N= 22). Further, 12.2% (N= 55) of the undergraduate sample identified as
having a Hispanic ethnicity, seven of which identified as white Hispanic. There were 44
participants who did not select any race category.
In order to participate in the larger undergraduate study, students had to be over
18 years of age and a student in the given course. For the current study, participants data
were not included in the study if their parents were never married, there was missing data
regarding parental divorce or infidelity, or if it was determined that their responses to the
survey were invalid. These inclusion and exclusion criteria left a total of 435 participants
with data that could be analyzed.
19


Procedure
Participants were recruited from psychology courses from the Auraria campus.
The participants instructor or the investigator announced the research opportunity to the
class. Instructors either offered the opportunity to volunteers without any course
incentives or in exchange for class credit; the instructor had the option of choosing which
method to use for recruitment. Students were told that the survey was about experiences
with infidelity or monogamy and contained personal and sensitive questions regarding
sexual practices and drug use (see appendix A for full announcement for study
participation). This was done to help minimize potential psychological risks for the
students that were thinking about participating.
Individuals who decided to participate in the study were asked to review a consent
form, complete a questionnaire, and then given a debriefing pamphlet. The informed
consent form again clarified the nature of the questions in the survey, and that the
participants could skip any questions they found too personal. Participants completed the
questionnaires in a proctored on-campus location, which was usually a classroom.
Several times were scheduled when participants were able to take the survey. Separate
seating was enforced in order to protect participants answers from being read or seen by
others. The debriefing forms addressed issues that may have been raised by the
questionnaire, and listed personal and relationship counseling resources for those who
needed it.
All questionnaires were kept confidential, and names were not attached to any
survey. The questionnaires did contain demographic variables, but no identifying codes
were assigned to the surveys. After completing the study, participants placed their
20


questionnaires in an envelope, sealed it, and returned it to the study coordinator. The first
page that contained the demographic data was then removed and kept separate from the
survey; this was done so that participants would not be able to be identified during data
entry due to any unusual demographic combinations.
Measures
Parental Infidelity and Divorce
Parental divorce was measured by the question If your parents were married, did
they divorce (yes/no). Parental infidelity was assessed by asking the participants to
check boxes indicating who they have known that have engaged in infidelity. The check
boxes included My father (biological, adoptive, step), At least one of my mothers
boyfriends, My mother (biological, adoptive, step), At least one of my fathers
girlfriends, Other relatives (uncles, aunts, grandfather, grandmother, etc.), and close
friends. The gender of the parent who committed infidelity was determined by the
participants selection of My father, My mother, or both on the checklist.
Attachment style
The Experiences in Close Relationships Inventory (ECRI; Hazan & Shaver, 1988)
was used to measure the attachment of participants. The scale consists of two subscales -
Avoidance and Anxiety with each subscale containing 18 items. Individuals who score
highly on the avoidant subscale are overly concerned with their autonomy, and are
uncomfortable being dependent or too close to others. Those who score highly on the
anxiety subscale are said to have a desire for extreme closeness in relationships, as well
as a reliance of self-worth that is based on the approval of others. For the current sample,
scores on the Avoidance Subscale averaged = 2.85 (SD = .97) out of a possible range of 1
21


to 7. The internal consistency was excellent (a = .92). The current sample scored an
average of 3.59 on the Anxiety Subscale (SD = 1.11) out of a possible range of 1 to 7.
This scale also had excellent internal consistency (a = .92).
Risky Behaviors. A novel scale that contained items about various risky sexual
behaviors, drinking, and drug use was used to assess participants risky behaviors. The
scale consisted of five questions on a 0-5 Likert scale. This items included In the past
year, how often have you had sexual intercourse without using a condom (scale 0-5), In
the past year, how often were you under the influence of alcohol when having sex (scale
0-5), Please estimate the number of sexual partners you have had in the past year
(open-ended), Over the last year, about how many alcoholic beverages do you drink a
week (open-ended), and Please estimate the number of times over the last year you
have used recreational drugs (open-ended). Items on the 0-5 scale were presented as 1 =
never, 2 = almost never, 3 = sometimes, 4 = almost always, 5 = always, and 0 = 1 have
not had sexual intercourse in the past year. Open-ended questions were restricted to a 0-5
response in order to form the scale. The risky behavior scale had an average score of
1.77 (SD = 1.12), and demonstrated adequate internal consistency (a = .75). The scales
alpha only marginally increased to .76 if the question In the past year, how often have
you had sexual intercourse without using a condom was removed, so all items were
included.
Table II. 1 Recoded values of open-ended item responses into 0-5 scale score
Scale
Score
0
Please estimate the Over the last year, Please estimate the number
number of sexual about how many of times over the last year
partners you have had alcoholic beverages do you have used recreational
in the past year_________you drink a week__________________drugs____________
0 0 0
22


Table II.l (cont.)
1 1 <1 1
2 2 1 1.99 2-4
3 3 2-3.99 5-10
4 4-5 4-9.99 11-45
5 6+ 10+ 46+
Own Infidelity. Ones own infidelity behavior was measured by the participants
response to Total number of other people that you have had sexual contact with while in
a serious and steady relationship. The average number of encounters was .88 (SD =
1.40). Due to the skewed distribution of this variable (data range from 0 to 40
encounters), the responses were restricted to a 6-point scale. The restricted infidelity
scale was recoded as follows: 0 on the scale meant that the individual had no infidelity
encounters, 1 on the scale meant that the individual had one previous infidelity encounter,
2 on the scale meant that the individual had two previous infidelity encounters, 3 on the
scale meant that the individual had three previous infidelity encounters, 4 on the scale
meant that the individual had four or five previous infidelity encounters, 5 on the scale
meant that the individual had six or more previous infidelity encounters.
Hypotheses
I hypothesized that individuals whose parents have committed infidelity would
have more insecure attachment styles, engage in more risky behaviors, and would be
more likely to have infidelity encounters themselves than those whose parents have not
engaged in infidelity. Further, I hypothesized that divorce would compound these
23


psychosocial effects; therefore, individuals whose parents are both divorced and have
committed infidelity would have the highest levels of insecure attachment, risky behavior,
and own infidelity.
I also hypothesized that there would be an interaction between the gender of the
parent who committed infidelity and the gender of the participant. Specifically, I
predicted that males whose fathers have engaged in infidelity would have more
extradyadic involvements than those whose mother has committed infidelity. Conversely,
females whose mothers have engaged in infidelity would have more extradyadic
involvements than females whose father has engaged in infidelity. Males and females
who have parents that have both engaged in infidelity would have the highest rates of
extradyadic involvements.
Overview of Data Analytic Approach
Using the questions described above, participants whose parents have not
committed infidelity nor divorced were in the no infidelity/no divorce group (NI/ND, N =
267). Checking father or mother on the infidelity question without the parents being
divorced placed the participant in the infidelity/no divorce group (I/ND, N = 35),
Participants who answered yes to the divorce question, but whose parent(s) have not
engaged in infidelity were placed in the divorce/no infidelity group (D/NI, N= 74).
Finally, participants whose parent(s) have committed infidelity and divorced were in the
infidelity/divorce group (I/D, N= 59). These were the four groups that served as the
independent variable in a series of ANOVAs.
The dependent variable for the first ANOVA was Avoidance as measured by the
ECRI. Anxiety was the dependent variable for the second ANOVA. The third ANOVA
24


had the risky behavior scale as the dependent variable. Finally, ones own infidelity was
the dependent variable for the fourth ANOVA. Post hoc analyses allowed for direct
comparisons of each group or main effect regarding relative levels of insecure attachment
styles, risky behaviors, and infidelity encounters.
In order to test the hypotheses about the interaction between the gender of the
parent who committed infidelity and the gender of the participant, a 2 x 3 ANOVA was
used with participant gender and gender of the parent who committed infidelity as the
independent variables. Gender of the participant was male or female, and gender of the
parent who committed infidelity was father only, mother only, or both parents. I
predicted that there would be an interaction between gender of the participant and gender
of the parent who committed infidelity such that participants whose same sex parent
committed infidelity would have the highest levels of own infidelity, while opposite sex
parental infidelity would result in lower levels of own infidelity.
During the proposal defense, committee members were interested in alternate
ways to statistically test the findings in order to demonstrate competence in this area and
to compare and contrast findings using different statistical procedures. Thus, as an
additional major goal of this project, multiple analyses were run for each research
question. Factorial ANOVAs and regressions were conducted to help illustrate any
significant findings of the study. Any discrepancies between these analyses and the
proposed analyses were evaluated and discussed.
25


CHAPTER III
RESULTS
Parental Status and Own Infidelity
Initially, a One-Way ANOVA was ran with parental status as the independent
variable (NI/ND, I/ND, NI/D, and I/D) and the participants own infidelity encounters as
the dependent variable. Results demonstrated that parental status was associated with
ones own infidelity, F(3, 417) = 3.41,/) < .05. Post hoc analysis indicated that
individuals whose parents remained married but had engaged in infidelity (I/ND) had the
highest rates of own infidelity (M= 1.47, SD = 1.83). This group was significantly
different from individuals whose parents remained married without infidelity (NI/ND; M
= .76, SD = 1.26). However, there were no significant differences between the I/ND
group and the NI/D group (M= .90, SD = 1.39) and the I/D group (M= 1.12, SD = 1.45).
Following up the initial ANOVA, a 2 x 2 ANOVA was run with parental divorce
status and parental infidelity status as the independent variables. A significant main
effect of the model was found for infidelity, F( 1, 416) = 7.22,/) < .05, partial r\2 = .01.
Thus, the results demonstrated that individuals whose parent(s) had engaged in infidelity
had significantly more infidelity encounters themselves. However, there were no other
significant main effects or interactions in the model; neither parental divorce status nor
the interaction between parental divorce and parental infidelity were significant.
Next, a regression was run with dummy coded parental status variables (NI/D,
I/ND, and I/D) as the independent variables. The NI/ND group was used as the reference
group for this analysis. Results demonstrated that the overall model predicted a
26


significant, yet small amount, of an individuals own infidelity encounters, F(3, 416) =
3.41 ,p < .05, R2 = .02. Consistent with the one way ANOVA results, the only significant
predictor in the model was the I/ND group, Std. fi = .14, p < ,05, This demonstrated that
the only group with significantly more infidelity encounters than participants in the
NI/ND group was participants whose parent(s) had engaged in infidelity, but did not
divorce.
Another regression was conducted with parental infidelity status (0 = no infidelity,
1 = infidelity) and parental divorce status (0 = no divorce, 1 = divorced) as predictors.
The overall regression model was also significant, but had a small effect size, F(2, 417) =
4.08,p < .05, R2 = .02. Consistent with the 2 X 2 ANOVA, parental infidelity status was
the only significant predictor in the model, (fi = .14, p < .05). This indicated that
individuals whose parent(s) engaged in infidelity were more likely to have more infidelity
encounters themselves.
Finally, due to the fact that the distribution of own infidelity encounters was
significantly skewed, nonparametric tests were run on both the scaled infidelity
encounters (responses confined to 0-5) and the nonscaled infidelity encounters (responses
were used as given, 0 to 40 range). A Kruskal-Wallis test was conducted with parental
status group (NI/ND, I/ND, NI/D, and I/D) as the independent variable. The test
demonstrated that for both the scaled and nonscaled infidelity encounters, parental status
significantly impacted participants own infidelity (scaled H(3) = 10.07, p <.05;
nonscaled H(3) = 9.68, p < .05). This provided further evidence that parental divorce and
infidelity status affected participants infidelity behavior.
27


Mann-Whitney tests were used to further explore this finding on the scaled
infidelity encounters to determine where the differences between groups occurred.
Consistent with the one way ANOVA, it was found that children in the NI/ND group and
children in the I/ND group significantly differed, with the I/ND group having more
infidelity encounters, U= 3340.5, z = -2.55,p< .05, r = -.15. Additionally, children in
the I/D group had significantly higher infidelity encounters than children in the NI/ND
group, U= 6285.5, z = -231,p < .05, r = -.13.
In summary, the I/ND group had the highest rates of own infidelity. Following
this was the I/D group, which had the second highest rates of infidelity. Next, the NI/D
group had the third highest rates. Finally, the NI/ND group had the lowest rates. Thus,
when comparing groups, it is found that the I/ND group is significantly higher than the
NI/ND group. Additionally, interaction effects are not found. Rather, there is a main
effect of infidelity when collapsing the groups across the levels of parental status
(infidelity and divorce).
Parental Status and Risky Behaviors
Initially, a One-Way ANOVA was run with parental status as the independent
variable (NI/ND, I/ND, NI/D, and I/D) and the participants risky behaviors as the
dependent variable. Results demonstrated that parental status significantly affected risk-
taking behavior, F(3, 429) = 3.34,p < .05. Specifically, individuals whose parents
divorced and had not engaged in infidelity (NI/D) had the highest rates of risky behaviors
(M= 2.08, SD= 1.18). This group was significantly different from individuals whose
parents remained married without infidelity (NI/ND; M= 1.68, SD = 1.15). However,
28


there were no significant differences between the NI/D group and the I/ND group (M=
1.79, SD = 1.00) and the I/D group (M= 2.04, SD = .99).
Exactly as had been done with the infidelity outcome variable, the same analyses
were run with risky behaviors as an outcome variable. Following the first ANOVA, a 2 x
2 ANOVA was run with parental divorce status and parental infidelity status as the
independent variables. A significant main effect of the model was found for divorce, F( 1,
429) = 5.08,p < .05 partial r|2 = .01. Results suggested that individuals whose parents
had divorced had significantly higher levels of risk-taking behavior than individuals
whose parents did not divorce. There were no other significant main effects or
interactions in the model; neither parental infidelity status nor the interaction between
parental divorce and parental infidelity significantly impacted risky behaviors.
Next, a regression was run with dummy coded parental status variables (NI/D,
I/ND, and I/D) as the independent variables. The NI/ND group was used as the reference
group for this analysis. Results demonstrated that the overall model predicted a
significant, yet small amount, of an individuals risk-taking behaviors, F(3, 429) = 3.34,p
< .05, R2 = .02. There were two significant predictors in the model, the NI/D group, Std.
= .13, p < .05 and the I/D group, Std. /? = .11 ,p< .05. This showed that individuals
whose parents had divorced, with or without infidelity, had higher levels of risk-taking
compared to individuals whose parents had neither divorced nor engaged in infidelity.
Another regression was conducted with parental infidelity status (0 = no infidelity,
1 = infidelity) and parental divorce status (0 = no divorce, 1 = divorced) as predictors.
This overall model was also significant, but had a small effect size, F(2, 430) = 4.88, p
< .05, R2 = .02. Parental divorce status was the only significant predictor in the model, (fi
29


= -14, p < .05). This result demonstrated that parental divorce status significantly
predicted ones risk-taking behavior, as parental divorce was associated with an increase
in risky behaviors.
Parental Status and Attachment
Anxious attachment. Initially, a One-Way ANOVA was run with parental status
as the independent variable (NI/ND, I/ND, NI/D, and I/D) and the participants anxious
attachment score as the dependent variable. Results demonstrated that parental status did
not significantly affect anxious attachment, F( 3, 410) = 1.74 ,p> .05. There were no
differences between individuals in the NI/ND group (M= 3.51, SD = 1.07), individuals in
the NI/D group (M= 3.58, SD = 1.12), individuals in the I/ND group (M= 3.84, SD =
1.10), and individuals in the I/D group (M= 3.82, SD = 1.27).
Exactly as had been done with the previous outcome variables, the same analyses
were run with anxious attachment as an outcome variable. Following up the first
ANOVA, a 2 x 2 ANOVA was run with parental divorce status and parental infidelity
status as the independent variables. Neither of the groups significantly impacted anxious
attachment. However, there was a trend for a main effect of infidelity, F{ 1, 410) = 3.83,
p = .05. Individuals whose parent(s) engaged in infidelity tended to have more anxious
attachment.
Next, a regression was run with dummy coded parental status variables (NI/D,
I/ND, and I/D) as the independent variables. The NI/ND group was used as the reference
group for this analysis. The overall model was not significant, F{3, 410) = 1.74,/) > .05,
R2 = .01. However, the I/D dummy coded group showed a trend, ft = 10, p = .06.
30


Another regression was conducted with parental infidelity status (0 = no infidelity,
1 = infidelity) and parental divorce status (0 = no divorce, 1 = divorced) as predictors.
This model also did not achieve significance, F(2, 411) = 2.57, p > .05, R2 = .01.
However, parental infidelity status demonstrated a trend towards significance, /? = .10,p
= .05. This result demonstrated a tendency for parental infidelity to be associated with
higher levels of anxious attachment.
Thus, while there was a trend shown for a main effect of infidelity in both the
ANOVA and the regression, when comparing the four groups there was not one single
group that emerged as highest in anxious attachment.
Avoidant attachment. Initially, a One-Way ANOVA was run with parental
status as the independent variable (NI/ND, I/ND, NI/D, and I/D) and the participants
avoidant attachment score as the dependent variable. Results demonstrated that parental
status did not significantly affect avoidant attachment, F(3, 411) = 1.90, p > .05. There
were no differences between individuals in the NI/ND group (M= 2.84, SD = 1.04),
individuals in the NI/D group (M= 2.68, SD = .83), individuals in the I/ND group (M =
3.05, SD = .98), and individuals in the I/D group (M= 2.62, SD = .81).
Exactly as had been done with the previous outcome variables, the same analyses
were run with avoidant attachment as an outcome variable. Following up the first
ANOVA, a 2 x 2 ANOVA was run with parental divorce status and parental infidelity
status as the independent variables. A significant main effect of the model was found for
divorce, F(l,411) = 5.31,/><.05 partial r\2 = .01. Contrary to expectations, it was found
that individuals whose parents had not divorced had significantly higher ratings of
avoidant attachment. There were no other significant main effects or interactions in the
31


model; neither parental infidelity status nor the interaction between parental divorce and
parental infidelity significantly impacted risky behaviors.
Next, a regression with dummy coded parental status variables (NI/D, I/ND, and
I/D) as the independent variables was run. The NI/ND group was used as the reference
group for this analysis. The overall model was not significant, F(3, 411) = 1.90, p > .05,
R2 = .01.
Another regression was conducted with parental infidelity status (0 = no infidelity,
1 = infidelity) and parental divorce status (0 = no divorce, 1 = divorced) as predictors.
This model overall also did not achieve significance, F(2, 412) = 2.22, p > .05, R2 = .01.
However, parental divorce status was a significant predictor, /? = -.1 \,p < .05. This result
indicated that individuals whose parents never divorced had higher levels of avoidant
attachment.
Parental Infidelity and Gender Interactions
In order to test the relationship between the gender of the parent who engaged in
infidelity and the gender of the participant, a 2 x 3 ANOVA would be run with the gender
of the participant (male or female) and gender of the parent who engaged in infidelity
(father, mother, or both) as independent variables, and the participants own infidelity as
the dependent variable. However, cell sizes for this analysis were quite small which
would severely limit statistical conclusions to be made. Thus, the ANOVA was not run
as the power of detecting an interaction for the model was only .24. However, see Figure
1 for an illustration of the group differences. The figure illustrates exactly what would
have been expected to happen, as males with father infidelity had the highest rates of own
32


infidelity, and it appears that there is an interaction between the gender of the parent who
engaged in infidelity and the gender of the participant.
Table III. 1 Cell sizes for parent and participant genders regarding infidelity behavior
Participant Gender
Parent who Committed Male Female
Infidelity
Father Only 14 33
Mother Only 4 16
Both Parents 11 22
Figure III. 1 Relationship between Gender of Parent and Child with Infidelity
33


CHAPTER IV
DISCUSSION
Using a large undergraduate sample, the effects of parental infidelity and divorce
were examined on adult childrens attachment styles, risky behaviors, and own infidelity.
Results demonstrated that both divorce and parental infidelity may impact children in a
variety of ways, even into adulthood. Namely, children whose parents were divorced had
elevated levels of risk-taking behaviors relative to those whose parents remained together.
Parental infidelity did not seem to contribute to higher risk-taking. Divorce may lead to
higher rates of risky behaviors, as parental resources are generally lessened in single-
parent homes; single parents cannot dedicate as many resources to supervising their
children and checking in on what behaviors they are engaging in. Furthermore, single-
parent families are often more strained in terms of financial resources compared to two-
parent homes. This may lead to a lack of extracurricular opportunities for children that
may be incompatible with, and, thus, prevent the engagement in, risky behaviors.
Another explanation as to why divorce may lead to risk-taking may be that the
children of divorce are acting out their frustrations from the shift in family dynamics.
Children of divorce may begin engaging in maladaptive coping techniques in order to
deal with the loss of a parent, and these coping strategies may become habitual patterns
of behavior throughout the individuals development. Conversely, infidelity did not seem
to impact the degree to which individuals engaged in risky behaviors. This study was the
first quantitative attempt to examine the relationship between parental infidelity and risk-
taking, and it does not appear to support the link speculated about in some case studies.
The absence of this relationship may be due to the fact that children can see the damaging
34


effects that infidelity (a risky behavior in itself) can have, and choose not to engage in
behaviors that may have similar consequences.
Another major finding of the current study was that individuals whose parents
committed infidelity had higher rates of infidelity than individuals whose parents did not
engage in infidelity. Furthermore, adult children whose parents remained married after
infidelity had the highest rates of own infidelity. This may be a result of a modeling
process in which children see their parents engage in infidelity, but remain married, and
thus there may be greater perceptions that infidelity is an acceptable behavior for them to
do as well, or they may simply have more prolonged exposure to the unfaithful parental
model.
Hypotheses regarding the impact that parental divorce and infidelity would have
on adult childrens attachment styles were generally not supported. Neither divorce nor
infidelity significantly affected participants anxious attachment styles; however,
individuals whose parents committed infidelity showed a trend towards having more
anxious attachments. Even though this effect was quite small, individuals who see their
parents engage in infidelity may develop a more anxious attachment towards their own
significant others. They may desire to be extremely close to their romantic partner, but
fear rejection (possibly through an infidelity encounter). It would be interesting to see if
individuals whose parents committed infidelity were more afraid of their own partners
engaging in infidelity than controls; however, more research needs to be conducted on
this topic.
Few effects were found regarding parental divorce and infidelity status on
avoidant attachment as well. One finding that did arise was that individuals whose
35


parents did not divorce had higher levels of avoidance. This finding was quite
unexpected, and may suggest that individuals who have parents that remain married have
an increased fear of getting close to others. One possible interpretation of this finding is
that individuals at a young age may avoid getting too close to others if the relationship
does not seem as strong as their parents. These individuals may avoid getting too close
in relationships until they are sure that their partner is the one. However, this finding
has not been established in previous literature, was a small effect, and in fact may be a
spurious, Type I, effect.
The overall lack of findings regarding parental effects on attachment was
surprising. It seems that children may not fear that their own relationships are bound to
end up like their parents, and so they do not obtain insecure attachment styles due to
parental behaviors. However, the time at which parents get divorced or engage in
infidelity may have differential effects on their children. If children are younger during
these events, they may experience stronger impacts in terms of attachment styles.
Unfortunately, this study was not able to control for the age at which the infidelity or
divorce occurred. Future research should focus on any age influences that may play a
role.
The hypothesis regarding the interaction of the parent(s) gender who committed
infidelity and the gender of the participant was partially supported. That is, it appeared
that male participants whose father engaged in infidelity had higher rates of infidelity
than if only the participants mother committed infidelity. Additionally, it appeared that
females whose mother committed infidelity had higher rates of own infidelity than if only
their fathers committed infidelity. These patterns suggest a modeling process in which
36


participants look to their same gendered parent for acceptable behaviors to engage in.
Males who see their father engage in infidelity, or females who see their mother engage
in infidelity, may model the same behavior later in life. It would be interesting to see if
this link is upheld by the gender commonality between child and parent, or if it is more
closely related to which parent the child feels closest to (which presumably occurs more
often with same gender dyads). More research with larger sample sizes should be
conducted in order to statistically examine this relationship.
It is also important to note that the effect sizes of the significant findings were
very small, and accounted for a minimal amount of variance. Thus, although it seems
that parental infidelity and divorce may play a role in shaping their adult childrens
behaviors, there seem to be other factors that carry more weight. One variable that has
consistently been found to carry stronger effects is relationship conflict between the
parents. Parents who have a high conflict relationship and get divorced or engage in
infidelity may have a stronger impact on their children. Thus, it would be important to
look at the divorce or infidelity situations more closely, including factors such as parental
conflict.
There were several limitations to the current study. First, the sample was
comprised of only undergraduates from a local university. It is unclear how these results
would generalize to the overall young adult population and to older community
populations. Second, only known parental infidelity was assessed. Participants had to be
aware of their parent(s) infidelity, and families in which the children know of an
infidelity may have different characteristics than families with unknown parental
37


infidelity. It would interesting to see whether these results would remain consistent for
families in which there is parental infidelity, but the children do not know about it.
Finally, it is unclear if it was the participants primary caretaker who engaged in
infidelity based on how the question was asked. If participants endorsed mother or father
infidelity, it could have been their biological, adoptive, or stepparent. Thus, a participant
whose parents divorced when he or she was very young may have had a biological parent
engage in infidelity, but most of his or her experiences were with the stepparent. This
may impact some of the modeling conclusions that were developed; however, the amount
of participants that this discrepancy would be applicable for would presumably be low.
More research needs to be conducted in this area to strengthen the findings of the current
study and address the limitations noted here.
38


REFERENCES
Allen, E. S., Atkins, D. C., Baucom, D. EL, Snyder, D. K., Gordon, K. C., & Glass, S. P.
(2005). Intrapersonal, interpersonal, and contextual factors in engaging in and
responding to extramarital involvement. Clinical Psychology: Science and
Practice, 72(2), 101-130.
Allen, E. S., & Baucom, D. H. (2004). Adult attachment and patterns of extradyadic
involvement. Family Process Special Issue: Meaningful Voices, Old and New,
43(4), 467-488.
Allison, P. D., & Furstenberg, F. F. (1989). How marital dissolution affects children:
Variations by age and sex. Developmental Psychology, 25(4), 540-549.
Amato, P. R. (1988). Parental divorce and attitudes toward marriage and family life.
Journal of Marriage & the Family, 50(2), 453-461.
Amato, P. R. (1994). Life-span adjustment of children to their parents' divorce. The
Future of Children, 4(1), 143-164.
Amato, P. R. (2001). Children of divorce in the 1990s: An update of the Amato and Keith
(1991) meta-analysis. Journal of Family Psychology, 15(3), 355-370.
Amato, P. R., & DeBoer, D. D. (2001). The transmission of marital instability across
generations: Relationship skills or commitment to marriage? Journal of Marriage
& the Family, 63(4), 1038-1051.
Amato, P. R., & Keith, B. (1991a). Parental divorce and the well-being of children: A
meta-analysis. Psychological Bulletin, 110(1), 26-46.
Amato, P. R., & Keith, B. (1991b). Parental divorce and adult well-being: A meta-
analysis. Journal of Marriage & the Family, 53(1), 43-58.
Amato, P. R., & Rogers, S. J. (1997). A longitudinal study of marital problems and
subsequent divorce. Journal of Marriage & the Family, 59(3), 612-624.
Angame-Lindberg, T., & Wadsby, M. (2009). Fifteen years after parental divorce:
Mental health and experienced life-events. Nordic Journal of Psychiatry, 63(1),
32-43.
Bachman, J. G., O'Malley, P. M., Schulenberg, J. E., Johnston, L. D., Freedman-Doan, P.,
& Messersmith, E. E. (2008). The education-drug use connection: How successes
andfailures in school relate to adolescent smoking, drinking, drug use, and
delinquency. New York, NY: Taylor & Francis Group/Lawrence Erlbaum
Associates.
Baydar, N. (1988). Effects of parental separation and reentry into union on the emotional
well-being of children. Journal of Marriage & the Family, 50(4), 967-981.
39


Beckwith, L., Cohen, S. E., & Hamilton, C. E. (1999). Maternal sensitivity during
infancy and subsequent life events relate to attachment representation at early
adulthood. Developmental Psychology, 35(3), 693-700.
Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York: Basic Books.
Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult
attachment: An integrative overview. In J. A. Simpson, & W. S. Rholes (Eds.),
Attachment theory and close relationships, (pp. 46-76). New York, NY, US:
Guilford Press.
Carranza, L. V., Kilmann, P. R., & Vendemia, J. M. C. (2009). Links between parent
characteristics and attachment variables for college students of parental divorce.
Adolescence, 44(114), 253-271.
Cavanagh, S. E., Crissey, S. R., & Raley, R. K. (2008). Family structure history and
adolescent romance. Journal of Marriage and Family, 70(3), 698-714.
Clarke-Stewart, K. A., Vandell, D. L., McCartney, K., Owen, M. T., & Booth, C. (2000).
Effects of parental separation and divorce on very young children. Journal of
Family Psychology, 14(2), 304-326.
Crowell, J. A., Treboux, D., & Brockmeyer, S. (2009). Parental divorce and adult
children's attachment representations and marital status. Attachment & Human
Development, 77(1), 87-101.
Downing-Matibag, T. M., & Geisinger, B. (2009). Hooking up and sexual risk taking
among college students: A health belief model perspective. Qualitative Health
Research, 19(9), 1196-1209.
Duncan, G. J., & Hoffman, S. D. (1985). Economic consequences of marital instability.
In Horizontal equity, uncertainty, and economic well-being. M. David and T.
Smeeding, eds. Chicago: University of Chicago Press.
Emery, R. E. (1982). Interparental conflict and the children of discord and divorce.
Psychological Bulletin, 92(2), 310-330.
Fomby, P., & Cherlin, A. J. (2007). Family instability and child well-being. American
Sociological Review, 72(2), 181-204.
Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item response theory analysis
of self-report measures of adult attachment. Journal of Personality and Social
Psychology, 78(2), 350-365.
Franklin, K. M., Janoff-Bulman, R., & Roberts, J. E. (1990). Long-term impact of
parental divorce on optimism and trust: Changes in general assumptions or
narrow beliefs? Journal of Personality and Social Psychology, 39(4), 743-755.
Greenberg, E. F., & Nay, W. R. (1982). The intergenerational transmission of marital
instability reconsidered. Journal of Marriage & the Family, 44(2), 335-347.
40


Guidubaldi, J., Cleminshaw, H. K., Perry, J. D., & Mcloughlin, C. S. (1983). The impact
of parental divorce on children: Report of the nationwide NASP study. School
Psychology Review, 12(3), 300-323.
Hadley, W., Brown, L. K., Lescano, C. M., Kell, H., Spalding, K., DiClemente, R.,
Donenberg, G., (2009). Parent-adolescent sexual communication: Associations of
condom use with condom discussions. AIDS and Behavior, 13(5), 997-1004.
Hamilton, C. E. (2000). Continuity and discontinuity of attachment from infancy through
adolescence. Child Development, 71(3), 690-694.
Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process.
Journal of Personality and Social Psychology, 52, 511-524.
Huurre, T., Junkkari, H., & Aro, H. (2006). Long-term psychosocial effects of parental
divorce: A follow-up study from adolescence to adulthood. European Archives of
Psychiatry and Clinical Neuroscience, 256(4), 256-263.
Jacquet, S. E., & Surra, C. A. (2001). Parental divorce and premarital couples:
Commitment and other relationship characteristics. Journal of Marriage & the
Family, 63(3), 627-638.
Lansford, J. E. (2009). Parental divorce and children's adjustment. Perspectives on
Psychological Science, 4(2), 140-152.
Laumann-Billings, L., & Emery, R. E. (2000). Distress among young adults from
divorced families. Journal of Family Psychology, 14(4), 671-687.
Lewis, M., Feiring, C., & Rosenthal, S. (2000). Attachment overtime. Child
Development, 71(3), 707-720.
Lusterman, D. (2005). Helping children and adults cope with parental infidelity. Journal
of Clinical Psychology.Special Issue: Treating Infidelity, 61(11), 1439-1451.
Martin, L. R., Friedman, H. S., Clark, K. M., & Tucker, J. S. (2005). Longevity following
the experience of parental divorce. Social Science & Medicine, 61(10), 2177-2189.
Mechanic, D., & Hansell, S. (1989). Divorce, family conflict, and adolescents' well-being.
Journal of Health and Social Behavior, 30(1), 105-116.
Molnar, D. S., Sadava, S. W., DeCourville, N. H., & Perrier, C. P. K. (2010). Attachment,
motivations, and alcohol: Testing a dual-path model of high-risk drinking and
adverse consequences in transitional clinical and student samples. Canadian
Journal of Behavioural Science/Revue Canadienne Des Sciences Du
Comportement, 42(1), 1-13.
National Center for Health Statistics (2002). Cohabitation, marriage, divorce, and
remarriage in the United States. Vital and Health Statistics, 23(22).
41


National Center for Health Statistics (2008). Health, United States, with chartbook.
Hyattsville, MD. Retrieved November 9, 2009 from
http://www.cdc.gov/nchs/data/hus/hus08.pdf.
Ryan, S., Franzetta, K., Schelar, E., & Manlove, J. (2009). Family structure history:
Links to relationship formation behaviors in young adulthood. Journal of
Marriage & the Family, 77(4), 935-953.
Sandberg, D. A., Suess, E. A., & Heaton, J. L. (2010). Attachment anxiety as a mediator
of the relationship between interpersonal trauma and posttraumatic
symptomatology among college women. Journal of Interpersonal Violence, 25(1),
33-49.
Schwartz, J. E., Friedman, H. S., Tucker, J. S., & Tomlinson-Keasey, C. (1995).
Sociodemographic and psychosocial factors in childhood as predictors of adult
mortality. American Journal of Public Health, 85(9), 1237-1245.
Schwinn, T. M., Schinke, S. P., & Trent, D. N. (2010). Substance use among late
adolescent urban youths: Mental health and gender influences. Addictive
Behaviors, 35(1), 30-34.
Sori, C. F. (2007). "An affair to remember": Infidelity and its impact on children. In P. R.
Peluso (Ed.), Infidelity: A practitioner's guide to working with couples in crisis.
(pp. 247-276). New York, NY, US: Routledge/Taylor & Francis Group.
Spruijt, E., & Duindam, V. (2005). Problem behavior of boys and young men after
parental divorce in the netherlands. Journal of Divorce & Remarriage, 43(3-4),
141-156.
Teachman, J. D. (2002). Childhood living arrangements and the intergenerational
transmission of divorce. Journal of Marriage and Family, 64(3), 717-729.
Wallerstein, J. S., & Kelly, J. B. (1996). Surviving the breakup: How children and
parents cope with divorce. Basic Books.
Wolfinger, N. H. (2000). Beyond the intergenerational transmission of divorce: Do
people replicate the patterns of martial instability they grew up with? Journal of
Family Issues, 27(8), 1061-1086.
Zill, N., Morrison, D. R., & Coiro, M. J. (1993). Long-term effects of parental divorce on
parent-child relationships, adjustment, and achievement in young adulthood.
Journal of Family Psychology, 7(1), 91-103.
42


APPENDIX A
ANNOUNCEMENT FOR STUDY PARTICIPATION
We are inviting you to participate in a research study entitled Relationships and Health
being conducted by Beth Allen, Ph.D., a faculty member in the Department of
Psychology at the University of Colorado at Denver (Elizabeth.allen@cudenver.edu).
This research has been reviewed and approved by the UCD Institutional Review Board.
The purpose of the research is to learn more about why and how people make decisions
about monogamy. For your participation, you will be asked to complete a fairly lengthy
questionnaire (it should take about 1 hour) which includes questions about some personal
issues such as sexual behaviors, infidelity, drug and alcohol use, and sexually transmitted
infections. Participation is open to all, regardless of experiences with these issues.
Because these questions are so personal, we are taking several steps to keep your answers
anonymous, so that no one can connect your answers to you specifically. Students are
being recruited from several psychology classes, with a goal of recruiting a minimum of
400 participants.
If you participate for credit, the study coordinator will write down your name so we can
tell your teacher you participated, but your name wont be connected to your
questionnaire. We think you might learn some interesting things about the research
process from your participation, but also are aware that some people find the experience
of answering personal questions very upsetting. We will let you know about free or low
cost counseling options, and also who to talk to if you have questions, concerns, or
complaints about the research experience.
If you are interested in participating, please contact the study coordinator,
_______________. She will schedule a time with you to take the questionnaire.
43


Full Text

PAGE 1

! i ADULT CHILDREN'S ACCOUNTS OF PARENTAL INFIDELITY AND DIVORCE: ASSOCIATIONS WITH OWN INFIDELITY, RISKY BEHAVIORS, AND ATTACHMENT by Aaron Michael Spence B.A., Bowling Green State University, 2009 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Masters of Arts Clinical Psychology Program 2012

PAGE 2

! ii This thesis for the Master of Arts degree by Aaron Michael Spence has been approved for the Clinical Psychology Program by Peter Kaplan, PhD, Chair Elizabeth Allen, PhD, Advisor Michael Zinser, PhD Joan Bihun, PhD Date: 11/05/2011

PAGE 3

! iii Spence, Aaron Michael (M.A., Clinical Psychology Program) Adult Children's Accounts of Parental Infidelity and Divorce: Associati ons with Own Infidelity, Risky Behaviors, and Attachment Thesis directed by Assistant Professor Elizabeth Allen, Ph.D. ABSTRACT Although there is a great deal of research that examines the effect that divorce has on children, there are very few studies that explore the impact that parental infidelity may have on this population. In this study, we examined undergraduates who report ed information about parental infidelity and divorce and measured their own infidelity behaviors, attachment, and risk taking. We hypothesized that participants with parents who were known to have committed infidelity and were divorced would have the high est rates of their own infidelity, more risk taking behavior, and more insecure attachment styles. Data was analyzed using ANOVAs and regression. Results partially supported the initial hypotheses. Specifically, participants whose parents engaged in inf idelity and remained married had the highest rates of their own infidelity. Additionally, participants with divorced parents had the highest rates of risk taking. Parental infidelity may serve as a model promoting children to also engage in this behavior in their own relationships. However, more work needs to be conducted in this area in order to establish causal connections. The form and content of this abstract are approved. I recommend its publication. Approved: Elizabeth Allen, Ph.D.

PAGE 4

! iv TABLE OF CON TENTS CHAPTER I. INTRODUCTION..................................................................................................... 1 Impacts of Divorce.................................................................................................. 1 Health Compromising and Risky Behaviors and Divorce.......................... 4 Attachment Styles and Divorce....................................... ........................... 7 Dissolution of Marriage among Adult Children of Divorce.................... 10 Impact of Infidelity............................................................................................... 11 Risky Behaviors and Infid elity................................................................. 12 Attachment and Infidelity......................................................................... 14 Parental Infidelity and Adult Children's Own Infidelity..................... ..... 16 II. METHODS.......... ................................................................................................... 13 Participants............................................................................................................ 19 Procedures............................................................................................................. 20 Measures............................................................................................................... 21 Parental Infidelity and Divorce................................................................. 21 Attachment Style............. .......................................................................... 21 Risky Behaviors........................................................................................ 22 Own Infidelity..................................................... ...................................... 23 Hypotheses............................................................................................................ 23 Overview of Data Analytic Approach........................................................ .......... 24 III. RESULTS................................................................................... ............................. 26 Parental Status and Own Infidelity...................................................................... 26

PAGE 5

! v Parental S tatus and Risky Behaviors................................................................... 28 Parental Status and Attachment........................................................................... 30 Parental Infidelity and Gender Interactions........ ................................................. 32 IV. DISCUSSION ............................................................................. ............................ 34 REFERENCES................................................................................................................. 39 APPENDIXES............ ........................... .............................................. ............................. 43 A Announcement for Study Participation ........................................................... 43

PAGE 6

! vi LIST OF TABLES Table II.1 Recoded values of open ended item responses into 0 5 scale score ....................... 22 II I 1 Cell sizes for parent and participant genders regarding infidelity behavior..... ........ 33

PAGE 7

! vii LIST OF FIGURES Figure III.1 Relationship between Gender of Parent and Child with Infidelity ........................... 34

PAGE 8

1 CHAPTER I INTRODUCTION Divorce may have significant impacts on the adults involved in the divorce, as well as important intergenerational effects on the parents' offspring. Related to this, parental infidelity, which may or may not end in divorce, may have important additional impacts on children. There are major gaps in the literature on this topic, however, but some noteworthy results have still been found. This paper will seek to discuss and integrate research findings regarding import ant intergenerational effects of divorce on children, factors that may influence these effects, and intergenerational impacts of infidelity. Impacts of Divorce Divorce is extremely prevalent in the United States today, and it is estimated that 50% of child ren in America will experience parental divorce (Lansford, 2009). This overwhelming statistic may have far reaching implications, as marital dissolution may impact children in a variety of ways. Some researchers have hypothesized that divorce may have im portant psychosocial effects on children, resulting in insecure attachment styles, behavioral problems, health compromising behaviors, cognitive and social deficits, psychological distress, poor academic achievement, and poor self concepts (Crowell, Trebou x, & Brockmeyer, 2009; Huure, Junkkari, & Aro, 2006; Lansford, 2009; Zill, Morrison, & Coiro, 1993). Divorce may also impact children throughout adulthood, and may lead to higher divorce rates among adult children whose parents have divorced.

PAGE 9

! 2 A variety of studies throughout the years have provided empirical support for the idea that divorce has a variety of impacts on children (Duncan & Hoffman, 1985; Guidubaldi, Cleminshaw, Perry, & Mcloughlin, 1983; Lansford, 2009; Zill, Morrison, & Coiro, 1993). These studies have also found that the impacts that children face are not always readily apparent. Children may appear quite resilient to the impacts of divorce, but may be suffering major emotional problems beneath the surface (Laumann Billings & Emery, 2000). Laumann Billings and Emery (2000) also provided evidence that the distress felt by children from divorced families may be considerable decades later; some children even speculated that they might be a different person today if their parents had not divor ced. Some caution needs to be used in interpreting these findings, however, as Clarke Stewart, Vandell, McCartney, Owen, and Booth (2000) hypothesized that parental variables (i.e. income, depression, education, ethnicity, and childrearing beliefs) may me diate the effects between divorce and the consequences felt by children. Other studies that have examined the impacts of divorce on children failed to find any significant results (Baydar, 1988; Mechanic & Hansell, 1989). These findings raise noteworthy c oncerns regarding the divorce literature; some of the studies were overpowered with very large sample sizes, and, yet, few significant differences were found (Baydar, 1988). Further, the significant differences that were found between children from divorc ed and intact families had negligible effect sizes. Results from these studies also showed that neither recent nor early divorce had an impact on health outcomes (Mechanic & Hansell, 1989). To examine these contradicting findings, Amato and Keith (1991a) performed a meta analysis on over 13,000 children from preschool to college age. This technique

PAGE 10

! 3 allowed Amato and Keith to look at effect sizes and account for differing design features of 92 studies. The meta analysis revealed that children from divorce d families generally experience lower academic achievement, poorer psychological adjustment, more behavioral problems, more negative self concepts, increased social difficulties, and more relationship problems with their mothers and fathers (Amato & Keith, 1991a). These results were echoed in multiple reports by the National Center for Health Statistics (2002; 2008), that stated that children from single parent households do worse in terms of academic achievement, depression, and behavioral problems than c hildren in two parent households. Amato and Keith (1991b) carried out an additional second meta analysis that analyzed over 80,000 adults to see if the impacts of divorce carry into adulthood. This analysis revealed that adult children who experienced par ental divorce had more behavioral problems, less education, lower job status, impaired psychological well being, a lower standard of living, lower marital satisfaction, a heightened risk of divorce, and worse physical health (Amato & Keith, 1991b). These findings illustrate major intergenerational implications, and it is interesting to note how children from divorced families are more likely to follow in their parents' footsteps and have relationships that end in divorce as well. It is also worth noting t hat Amato (2001) followed up these meta analyses a decade later and incorporated 67 studies conducted in the 1990s and found similar results. This study, as well as others (e.g. Lansford's review (2009) that examined the effects of divorce in more recent years), helped demonstrate that impacts of divorce seem to be consistent and stable throughout time.

PAGE 11

! 4 Amato (1994) urges not to be overly concerned with these findings, however, as the decline in functioning is small, and there is a great deal of overlap in well being between children of divorced and intact families. Also, psychosocial impacts on children as a result of divorce seem to be the exception rather than the rule, as most children do not experience significant negative outcomes. Following this, a nother, more recent study, which followed individuals over 15 years after their parents divorced also found that divorce carried effects into adulthood (€ngarne Lindberg & Wadsby, 2009). This study demonstrated that adult children of divorce experienced m ore negative life events, which resulted in difficult adjustment. However, these findings and effect sizes were small, with minimal between group differences. The small differences found in these studies are still important to note, nonetheless, so that clinicians know where to focus and can work to try to make differences between children from divorced and intact families as diminutive as possible. Although Amato's (1994) review gives hope to children of divorce (the overall difference in well being between children whose parents have divorced and children from intact families is small), there may be other variables that show a stronger impact on children of divorce. Three particular variables of interest that have received attention in the divorce l iterature are health compromising and risky behaviors, attachment styles, and subsequent divorce of adult children, all of which seem to be impacted by parental divorce. Each variable will be reviewed below. Health Compromising and Risky Behaviors and Div orce Health compromising and risky behaviors encompass a wide range of behaviors that may lead to many poor health outcomes and early morbidity and mortality.

PAGE 12

! 5 Examples of these behaviors include smoking; heavy drinking; using illicit drugs; and engaging i n risky sexual behaviors such as having unprotected sex, having sex under the influence of alcohol or drugs, or engaging in sexual activities with multiple partners; the psychosocial implications of all of these behaviors have been documented thoroughly in the literature. Currently, there have been a handful of studies conducted that have examined the link between parental divorce and children's involvement in risky health behaviors. Huure, Junkkari, and Aro (2006) conducted a 16 year longitudinal study th at examined the impacts of divorce on a variety of outcomes, including health compromising behaviors. The study examined 1,471 participants between the ages of 16 to 32 whose parents either divorced or did not. Results showed that daily smoking and hazar dous alcohol consumption were more prevalent in the children from divorced families than those from intact families for both males and females. The findings of this study also help provide evidence for the long standing impacts that divorce can have well into adulthood. Another study conducted by Schwartz, Friedman, Tucker, and Tomlinson Keasey (1995) looked at archival data in order to examine mortality differences between children from divorced and intact families. It was found that children from divorc ed families had decreased longevity, and, on average, died four years earlier than those from intact families; a finding that may be a result of increased engagement in health compromising behaviors. Martin, Friedman, Clark, and Tucker (2005) followed up this study to assess potential mediators and moderators of the decreased longevity among children of divorce finding, and found that smoking was the strongest mediator of this relationship; a finding

PAGE 13

! 6 that provides further evidence that children of divorce are at an increased likelihood of engaging in risky health behaviors. Further evidence that children from divorced families engage in more risky health behaviors than children from intact families comes from the work of Spruijt and Duindam (2005). The aut hors used data gathered from the longitudinal study Utrecht Study of Adolescent Development, which is a representative random sample of adolescents and young adults (age range from 12 30) to examine differences between males and females from divorced and i ntact families. The results showed that males from divorced families engaged in more risky behaviors (defined as smoking, drinking, and using soft drugs), and had a higher number of sexual partners. There are a variety of theories as to why children of di vorce may engage in health compromising and risky behaviors more so than children from intact families. One theory that attempts to explain the link between divorce and health compromising behaviors focuses on the disparities between children from divorce d and intact families as a result of parental divorce. It has been found that children from divorced families have lower education, higher unemployment, and higher prevalence of depression and minor psychiatric disturbance (Huure et al., 2006), all of whi ch are predictive of health compromising behaviors (Bachman, et al., 2008; Downing Matibag & Geisinger, 2009; Schwinn, Schinke, & Trent, 2010). These findings suggest that divorce may increase an adult child's propensity to engage in such behaviors throug h the mediating factors of lower education, unemployment, and impaired mental health status. Another popular theory as to why children of divorce may engage in more risky health behaviors than their counterparts from intact families stems from a lack of pa rental

PAGE 14

! 7 supervision and involvement. Children from divorce are often under the primary custody of one parent and do not receive as much parental regulation as children who have two parents to share the burden. Resources may also be strained after divorce, and adequate child care services are not available. This lack of sufficient supervision may give children the opportunity to engage in risky behaviors and form bonds with peers who engage in these behaviors as well. An additional theory states that child ren may engage in risky behaviors in order to cope with parental separation. Parental divorce creates a major shift in family dynamics, and children may feel quite distressed from the change. Children may engage in poor coping techniques, such as using d rugs or engaging in risky sex, to deal with the loss and displace any depression they may feel. It has also been widely documented that adolescents may also engage in risky behaviors through acting out (Allison, & Furstenberg, 1989; Lansford, 2009; Zill e t al., 1993); a behavior that may serve as another means to cope with parental separation. Attachment Styles and Divorce Attachment, broadly defined, is the way in which individuals cognitively conceptualize their internal working models of self and other s in interpersonal relationships (Carranza, Kilmann, & Vendemia, 2009). Attachment theory was first conceptualized by Bowlby (1969) in the context of infant/caretaker relationships, and was based on the manner in which the infant interacted with and witho ut his or her primary caretaker. Attachment theory has also been extended to adult intimate relationships (Hazan & Shaver, 1987) in order to assess how individuals function in and conceptualize their romantic relationships.

PAGE 15

! 8 The concept of attachment flows from two continuum of anxiety and avoidance regarding interpersonal relationships. Anxiety is indicative of a desire for extreme closeness in relationships, as well as a reliance of self worth that is completely maintained by others. At the same time, a nxious attachment is colored with worries about abandonment (Allen & Baucom, 2004). Conversely, individuals with avoidant attachment are overly concerned with their autonomy and are uncomfortable with being dependent or too close to others (Allen & Baucom 2004). Put another way, individuals with high anxiety are said to have a negative view of self, while individuals with high avoidance are said to have a negative view of others (Brennan, Clark, & Shaver, 1998). Stemming from the two dimensions of anxiet y and avoidance are four separate attachment categories that an individual can fall into. An individual that is high on the anxious continuum and low on the avoidant dimension is said to have a preoccupied attachment style, while an individual who is high on the avoidant continuum and low on the anxious continuum is said to have a dismissive attachment style (Brennan et al., 1998). Individuals who rank highly on both scales have a fearful attachment style; a style theoretically based on a negative view of both others and self, so that extreme closeness is desired but feared and avoided due to concerns regarding rejection (Fraley, Waller, & Brennan, 2000). Finally, an individual who rates low on each scale is said to have a secure attachment style with a p ositive view of self and others. The measurement of attachment has changed and improved throughout the years. Attachment has been measured in a variety of ways, including the use of observational and interview techniques. However, it has become a quite p opular and accepted method to measure attachment from self report scales (Brennan et al., 1998). Two of the

PAGE 16

! 9 commonly used self report measures in studies that assess attachment are the Relationship Styles Questionnaire (RSQ) and the Experiences in Close R elationships Inventory (ECRI; Fraley et al., 2000). A myriad of studies have been conducted throughout the years that have assessed how attachment may impact various outcomes, ranging from high risk drinking (Molnar, Sadava, DeCourville, & Perrier, 2010) t o developing posttraumatic symptomatology after a traumatic event (Sandberg, Suess, & Heaton, 2010) to infidelity (Allen & Baucom, 2004). These studies have primarily found that insecure attachment styles (preoccupied, dismissive, and fearful) lead to poo rer outcomes. Attachment styles that are developed in childhood are also thought to influence adult attachment, and are believed to be relatively stable throughout life. However, significant disruptions, changes, or events in an individual's life may imp act his or her attachment style and cause it to change (Hamilton, 2000). For this reason, many researchers have been interested in how divorce impacts children's attachment styles, especially because it has been tied to many adverse outcomes. Further, it is easy to theorize how divorce may impact attachment styles by disrupting parent/child relationships, and by potentially causing children to view relationships with feelings of anxiety and avoidance. Many studies have found that divorce may play an impor tant role in affecting individuals' attachment styles, with divorce leading to more insecure attachment styles among children (Beckwith, Cohen, & Hamilton, 1999; Crowell, Treboux, & Brockmeyer, 2009; Lewis, Feiring, & Rosenthal, 2000). In a study of 202 c ouples consisting of 404 individuals between the ages of 19 and 35, Jacquet and Surra (2001) also found that women from divorced families had more distrust, dissatisfaction, and ambivalence in

PAGE 17

! 10 their own relationships; all of which are characteristics of in secure attachment styles. Men did not seem to be as impacted, however, and only seemed to demonstrate more distrust and ambivalence if their partner's parents were divorced; gender disparities of this study and other findings will be discussed later. It seems clear in the literature that the major shift in parental caretaking and family dynamics that characterize divorce is a phenomenon that readily impacts children's attachment styles. Dissolution of Marriage among Adult Children of Divorce There are mu ltiple theories that speculate why adult children from divorced families may be more likely to divorce themselves. One theory posits that modeling may offer the potential explanation (Crowell et al., 2009). Parents who divorce may illustrate maladaptive behaviors to maintain a committed marriage such as poor communication, negative emotion, and withdrawal from conflict resolution (Crowell et al., 2009). Subsequently, their children may model these behaviors in their own marriages, characteristics that ar e likely to lead to their own divorce. Another hypothesis states that parents may convey, intentionally or unintentionally, positive beliefs about divorce and a shallow commitment to marriage (Crowell et al., 2009). Children may foster these beliefs, and see divorce as an acceptable alternative in their own lives. This hypothesis was supported in a large study of 2,033 married individuals by Amato and DeBoer (2001); they found that it was divorce itself that predicted the intergenerational risk of divorc e, apart from any conflict in the parent's relationship. The authors concluded that individuals from divorced families have a weaker commitment to the norm of lifetime marriage than children from intact families (Amato & DeBoer, 2001). Children of divorc e have also been shown to

PAGE 18

! 11 understand that life goes on after divorce (Greenberg & Nay, 1982), are more accepting of alternatives to marriage (Amato, 1988), and they also hold less optimistic views of marriage (Franklin, Janoff Bulman, & Roberts, 1990). Th ere is a plethora of research to support these hypotheses, and the intergenerational transmission of divorce has been widely documented (Amato & DeBoer, 2001; Amato & Keith, 1991b; Teachman, 2002; Wolfinger, 2000). The consequence of parental divorce also has a potent effect on the intergenerational transmission of divorce; Amato and DeBoer (2001) note that parental divorce doubles the chances that their adult children will also divorce. Multiple family disruptions may further exacerbate the problems, as children who undergo multiple parental divorces and transitions may be at an increased risk for divorce, behavioral problems, poor educational outcomes, and unstable relationships (Cavanagh, Crissey, & Raley, 2008; Fomby & Cherlin, 2007; Ryan, Franzetta, S chelar, & Manlove, 2009; Wolfinger, 2000). Overall, it seems safe to conclude that parental divorce is a strong predictor of adult children's own divorce. Impacts of Infidelity Infidelity, also known as extradyadic involvement (EDI), can be operationalize d as having sexual encounters with an individual outside of one's primary relationship in which monogamy is expected. EDI is a prevalent phenomenon in romantic relationships, and when discovered, can elicit a great deal of distress and conflict in both pa rtners (Allen, Atkins, Baucom, Snyder, Gordon, & Glass, 2005). Infidelity is also a common precipitant of divorce (Amato & Rogers, 1997), and it is important to hypothesize how EDI may have significant intergenerational impacts on children. However, ther e have been very few empirical studies that have examined the intergenerational impacts of

PAGE 19

! 12 infidelity, and, therefore, it is important to utilize past research from a related topic. Drawing from the divorce literature, then, it would seem important to exa mine the impacts that infidelity has on a variety of outcomes, namely children's risky behaviors, attachment styles, and own infidelity. Evidence for infidelity's impact on each outcome, and theories of why these variables may be affected, will be conside red in turn. Risky Behaviors and Infidelity There is currently no empirical evidence to support the link between parental infidelity and children's risky behaviors. However, a variety of sources have speculated about infidelity's impact in this area. Wal lerstein and Kelly (1996) theorized that adolescents who find out about parental infidelity may regress and be extremely anxious in the wake of their sexual and aggressive impulses; effects that could potentially lead to acting out and serious behavior cha nges. Wallerstein and Kelly provided a case example of a girl named Jean. Jean became sexually active at the age of 14, which coincided with her discovery of her father's affair. With her mother, Jean formed strategies for punishing her father and fanta sized about his sexual involvements. After her parent's divorce, she continually flirted with her older male teachers and soon began abusing alcohol and drugs. Sori (2007) also commented on the idea that infidelity may impact children's risky behaviors. She stated that older children may react to infidelity through externalizing behaviors, and, for adolescents or young adults, this may mean using alcohol, drugs, or sex to act out. Using sex as an outlet may especially be pronounced in older children as they have a heightened sense of their own sexuality; increased acting out, linked with insufficient parent/child communication about safe sex practices as a

PAGE 20

! 13 result of family disruption, may further lead to increased sexual risk behaviors such as not using condoms (Hadley et al., 2009). Further, a clinical case study illustrated by Lusterman (2005) illustrated the impact that infidelity had on a teenage girl. In the c ase, the girl found out about her father's infidelity, but kept it a secret at his request. However, she began acting out and dressing provocatively; signs that she may be engaging in risky behavior. Caution must be given in interpreting this and other c ase examples, as generalizability may be limitied; however, these case studies and speculations provide important direction and insight on how infidelity may impact children's risky behaviors. Empirical studies are needed to see if these results generaliz e and are a common finding among children who experience parental infidelity. There have been multiple theories developed to hypothesize why children may engage in risky behaviors following parental infidelity. One hypothesis states that infidelity causes a "shattered superego" in the child. The case study of Priscilla (Wallerstein & Kelly, 1996) illuminated this idea vividly with her statement, "The beliefs which gave me the ability to effectively deal with life were blown apart for me by my dad's [infid elity]". Parental infidelity may undermine many of the ideals that the child was taught by the parent, with the result that the child does not know what is right or wrong anymore. This may lead the child to engage in behaviors that were previously consid ered bad, such as drinking, smoking, doing drugs, or having sex. Another hypothesis that links parental infidelity to risky behaviors in children states that an affair may destroy a child's sense of security and trust, and they may no longer see the world as safe or predictable (Sori, 2007). This shattered trust may lead a child to seek security elsewhere in order to cope and feel safe. However, this search may

PAGE 21

! 14 lead to risky behaviors such as drug use or sex. Additionally, adolescents may return this sha ttered sense of trust with anger and hostility towards the parent(s) (Sori, 2007; Wallerstein & Kelly, 1997). This may lead adolescents to engage in risky behaviors in order to cope or to lash out at parents. One additional theory states that children may engage in risky behaviors, namely risky sexual behaviors, due to modeling. Children may believe that their parent(s) are engaging in unsafe sexual practices, and model this behavior in their own relationships. They may believe that this behavior is acce ptable if their parent(s) do it and follows in their footsteps. Attachment and Infidelity Theories regarding the impact of infidelity on attachment are relatively straightforward. Generally, they state that infidelity may be a major disruption or shift i n an individual's life and can precipitate a change in the child's attachment style; usually with the end result being an insecure attachment style. This may occur because infidelity may diminish an individual's self esteem and shatter trust and conceptio ns of intimacy (Sori, 2007). Following this, Lusterman (2005) points out that the daughter's trust in boys was severely impacted by the discovery of her father's infidelity, and all of her subsequent relationships were affected. Further, children may no longer hold the same beliefs about the importance of intimate relationships or the stability of partnerships (Sori, 2007). Therefore, seeing a parent's infidelity may lead the child to develop insecure attachment for a couple of reasons. Anxious attachme nt may occur due to a desperate need for closeness, while maintaining an acute fear of abandonment. Conversely, parental infidelity may lead to avoidant attachment, as an individual may take on a

PAGE 22

! 15 negative view of others and avoid closeness. Although the literature relating to the intergenerational impacts of infidelity is sparse, there have been two empirical studies conducted that have assessed parental infidelity's impact on adult children's attachment. Walker and Ehrenberg (1998) conducted an advent st udy that examined the intergenerational impacts of infidelity. This study sought to examine if a child's perceptions of parental divorce had an impact on his or her attachment style. Walker and Ehrenberg (1998) found that children who attributed their pa rents' divorce to infidelity were significantly more likely to have an insecure attachment style, which may have important implications for their own intimate relationships later in life. Another important study conducted by Platt, Nalbone, Casanova, and W etchler (2008) examined if parental conflict and infidelity could serve as predictors for adult children's attachment style and own infidelity. Platt et al. (2008) only found partial evidence that parental conflict predicted attachment style, and they fou nd no evidence that parental infidelity predicted attachment. This seems to run contrary to Walker and Ehrenberg's (1998) findings, but parental conflict may have been a confounding variable in Walker and Ehrenberg's study. The two studies also used diff erent measures to determine attachment style, which may explain the difference. Both of the studies conducted by Walker and Ehrenberg (1998) and Platt et al. (2008) have important limitations that need to be noted. Both studies had relatively small sample sizes and the participants consisted only of undergraduate students. This means that smaller effects may not be able to be detected and that the samples may have been unrepresentative. Another major limitation of these studies is the failure to include participants whose parent(s) have committed infidelity, but remain married.

PAGE 23

! 16 Parental Infidelity and Adult Children's Own Infidelity The intergenerational impact of parental EDI on adult children's subsequent infidelity is theoretically tied to the intergen erational effects of divorce. Parental infidelity may lead children to develop positive views of infidelity and a shallow commitment to being faithful to their partner. Subsequently, children may believe that infidelity is not a big deal, and is a common and acceptable practice. Other theories regarding the intergenerational transmission of infidelity are similar to the impacts of parental EDI on risky behaviors. This theoretical link may be tied to the idea that infidelity is a form of risky behavior. Thus, children may act out and engage in risky sexual practices, including having multiple partners or committing infidelity. The intergenerational transmission of infidelity may also be tied to theories of risky behavior through modeling. Children may o bserve parental infidelity and take on the belief that "if my parents do it, it must be ok." Thus, children will see their parents engaging in EDI and follow in their parent's footsteps and actions. Although few empirical studies have been conducted, ther e has been some evidence concerning the intergenerational transmission of infidelity. Platt et al. (2008) found that children that know their father has had EDI are more likely to engage in infidelity themselves. When further examined, it was found that this relationship only remained significant for boys. This may illustrate a modeling effect where boys know of their fathers EDI, and subsequently find the behavior to be more acceptable. However, this link needs to be studied much further in depth befor e conclusions can be made. The impacts of parental divorce and conflict on children have been well documented in a plethora of studies throughout the years. Research that focuses on the

PAGE 24

! 17 impacts that infidelity has on children is in its infancy, however, a nd there is a paltry amount of literature on this issue. The studies conducted by Walker and Ehrenberg (1998) and Platt et al. (2008), as well as the case studies by Lusterman (2005) and Wallerstein and Kelly (1997) provide helpful directions to future re searchers interested in exploring this new research topic. The findings regarding the intergenerational effects of divorce should also be utilized, and the child's gender is important to include as a potential moderator on the impacts of infidelity. Stud ies of divorce have also pinpointed that children are generally affected in terms of risky behaviors, attachment, and subsequent divorce, and infidelity research should focus on these outcomes as well. Further, the divorce literature seems to agree that t he impacts of divorce carry well into adulthood, a finding that should be explored in the infidelity literature as well. A great deal of research needs to be conducted to fully explain the intergenerational impacts of infidelity, but with time and effort it can eventually be well understood. A study that I propose to conduct would build off of the research already conducted regarding the intergenerational impacts of infidelity. Currently, there is a discrepancy in the research regarding the impacts of inf idelity on attachment style, little investigation regarding an adult child's subsequent infidelity with knowledge of parental infidelity, a gap in the literature regarding the intergenerational impacts of infidelity on children engaging in risky behaviors, and a lack of studies that use samples that include children whose parents committed infidelity but remain together. Therefore, I propose to conduct a study with four main purposes: (1) given the conflicting findings in the literature, to further investi gate the impact of parental infidelity on an adult child's attachment style, (2) to explore if parental infidelity results in higher rates of subsequent infidelity

PAGE 25

! 18 behaviors from adult children, (3) to examine whether or not parental infidelity leads to ch ildren engaging in more risky behaviors, and (4) to assess the potential moderating impact that the child's gender has on these outcomes. For all these research questions, I will use the adult children's self report, thus, these links between parental inf idelity and current functioning will only relate to parental infidelity which is known by the respondent.

PAGE 26

! 19 CHAPTER II METHODS In order to conduct this study, I conducted secondary analyses using a data set collected by Dr. Elizabeth Allen. An undergraduate sample (described below) was used to test the hypotheses. Participants Participants in Dr. Allen's study consisted of undergraduate students from a local university campus ( N = 484). The undergraduate students were recruited from psychology courses, and were given the option to participate in the study for credit. The age of the students ranged between 18 to 67 years ( M = 21.53). Males constituted 29% of the sample, while females composed the other 71%. The sample contained African Americ ans (5.9%, N = 26), American Indian or Alaskan Natives (.5%, N = 2), Pacific Islanders (.9%, N = 4), whites (65.8%, N = 291), Asians (15.9%, N = 70), and individuals who identify with another racial group (5.7%, N = 25) or more than one racial group (4.5%, N = 22). Further, 12.2% ( N = 55) of the undergraduate sample identified as having a Hispanic ethnicity, seven of which identified as white Hispanic. There were 44 participants who did not select any race category. In order to participate in the larger u ndergraduate study, students had to be over 18 years of age and a student in the given course. For the current study, participants' data were not included in the study if their parents were never married, there was missing data regarding parental divorce o r infidelity, or if it was determined that their responses to the survey were invalid. These inclusion and exclusion criteria left a total of 435 participants with data that could be analyzed.

PAGE 27

! 20 Procedure Participants were recruited from psychology courses from the Auraria campus. The participants' instructor or the investigator announced the research opportunity to the class. Instructors either offered the opportunity to volunteers without any course incentives or in exchange for class credit; the instruc tor had the option of choosing which method to use for recruitment. Students were told that the survey was about experiences with infidelity or monogamy and contained personal and sensitive questions regarding sexual practices and drug use (see appendix A for full announcement for study participation). This was done to help minimize potential psychological risks for the students that were thinking about participating. Individuals who decided to participate in the study were asked to review a consent form, complete a questionnaire, and then given a debriefing pamphlet. The informed consent form again clarified the nature of the questions in the survey, and that the participants could skip any questions they found too personal. Participants completed the q uestionnaires in a proctored on campus location, which was usually a classroom. Several times were scheduled when participants were able to take the survey. Separate seating was enforced in order to protect participants' answers from being read or seen b y others. The debriefing forms addressed issues that may have been raised by the questionnaire, and listed personal and relationship counseling resources for those who needed it. All questionnaires were kept confidential, and names were not attached to an y survey. The questionnaires did contain demographic variables, but no identifying codes were assigned to the surveys. After completing the study, participants placed their

PAGE 28

! 21 questionnaires in an envelope, sealed it, and returned it to the study coordinato r. The first page that contained the demographic data was then removed and kept separate from the survey; this was done so that participants would not be able to be identified during data entry due to any unusual demographic combinations. Measures Parenta l Infidelity and Divorce Parental divorce was measured by the question "If your parents were married, did they divorce (yes/no)". Parental infidelity was assessed by asking the participants to check boxes indicating who they have known that have engaged in infidelity. The check boxes included "My father (biological, adoptive, step)", "At least one of my mother's boyfriends", "My mother (biological, adoptive, step)", "At least one of my father's girlfriends", "Other rela tives (uncles, aunts, grandfather, grandmother, etc.)", and "close friends". The gender of the parent who committed infidelity was determined by the participant's selection of "My father", "My mother", or both on the checklist. Attachment style The Experi ences in Close Relationships Inventory (ECRI; Hazan & Shaver, 1988) was used to measure the attachment of participants. The scale consists of two subscales Avoidance and Anxiety with each subscale containing 18 items. Individuals who score highly on the avoidant subscale are overly concerned with their autonomy, and are uncomfortable being dependent or too close to others. Those who score highly on the anxiety subscale are said to have a desire for extreme closeness in relationships, as well as a rel iance of self worth that is based on the approval of others. For the current sample, scores on the Avoidance Subscale averaged = 2.85 ( SD = .97) out of a possible range of 1

PAGE 29

! 22 to 7. The internal consistency was excellent (! = .92). The current sample score d an average of 3.59 on the Anxiety Subscale ( SD = 1.11) out of a possible range of 1 to 7. This scale also had excellent internal consistency (! = .92). Risky Behaviors. A novel scale that contained items about various risky sexual behaviors, drinking, and drug use was used to assess participants' risky behaviors. The scale consisted of five questions on a 0 5 Likert scale. This items included "In the past year, how often have you had sexual intercourse without using a condom (scale 0 5)", "In the pas t year, how often were you under the influence of alcohol when having sex (scale 0 5)", "Please estimate the number of sexual partners you have had in the past year (open ended)", "Over the last year, about how many alcoholic beverages do you drink a week (open ended)", and "Please estimate the number of times over the last year you have used recreational drugs (open ended)". Items on the 0 5 scale were presented as 1 = never, 2 = almost never, 3 = sometimes, 4 = almost always, 5 = always, and 0 = I have n ot had sexual intercourse in the past year. Open ended questions were restricted to a 0 5 response in order to form the scale. The risky behavior scale had an average score of 1.77 ( SD = 1.12), and demonstrated adequate internal consistency (! = .75). T he scale's alpha only marginally increased to .76 if the question "In the past year, how often have you had sexual intercourse without using a condom" was removed, so all items were included. Table II.1 Recoded values of open ended item responses into 0 5 scale score Scale Score Please estimate the number of sexual partners you have had in the past year Over the last year, about how many alcoholic beverages do you drink a week Please estimate the number of times over the last year you have used recreation al drugs 0 0 0 0

PAGE 30

! 23 Table II.1 (con't.) 1 1 <1 1 2 2 1 1.99 2 4 3 3 2 3.99 5 10 4 4 5 4 9.99 11 45 5 6+ 10+ 46+ Own Infidelity. One's own infidelity behavior was measured by the participant's response to "Total number of other people that you have had sexual contact with while in a serious and steady relationship". The average number of encounters was .88 ( SD = 1.40). Due to th e skewed distribution of this variable (data range from 0 to 40 encounters), the responses were restricted to a 6 point scale. The restricted infidelity scale was recoded as follows: 0 on the scale meant that the individual had no infidelity encounters, 1 on the scale meant that the individual had one previous infidelity encounter, 2 on the scale meant that the individual had two previous infidelity encounters, 3 on the scale meant that the individual had three previous infidelity encounters, 4 on the scal e meant that the individual had four or five previous infidelity encounters, 5 on the scale meant that the individual had six or more previous infidelity encounters. Hypotheses I hypothesized that individuals' whose parents have committed infidelity would have more insecure attachment styles, engage in more risky behaviors, and would be more likely to have infidelity encounters themselves than those whose parents have not engaged in infidelity. Further, I hypothesized that divorce would compound these

PAGE 31

! 24 psy chosocial effects; therefore, individuals whose parents are both divorced and have committed infidelity would have the highest levels of insecure attachment, risky behavior, and own infidelity. I also hypothesized that there would be an interaction between the gender of the parent who committed infidelity and the gender of the participant. Specifically, I predicted that males whose fathers have engaged in infidelity would have more extradyadic involvements than those whose mother has committed infidelity. Conversely, females whose mothers have engaged in infidelity would have more extradyadic involvements than females whose father has engaged in infidelity. Males and females who have parents that have both engaged in infidelity would have the highest rate s of extradyadic involvements. Overview of Data Analytic Approach Using the questions described above, participants whose parents have not committed infidelity nor divorced were in the no infidelity/no divorce group (NI/ND, N = 267). Checking father or mo ther on the infidelity question without the parents being divorced placed the participant in the infidelity/no divorce group (I/ND, N = 35). Participants who answered yes to the divorce question, but whose parent(s) have not engaged in infidelity were pla ced in the divorce/no infidelity group (D/NI, N = 74). Finally, participants whose parent(s) have committed infidelity and divorced were in the infidelity/divorce group (I/D, N = 59). These were the four groups that served as the independent variable in a series of ANOVAs. The dependent variable for the first ANOVA was Avoidance as measured by the ECRI. Anxiety was the dependent variable for the second ANOVA. The third ANOVA

PAGE 32

! 25 had the risky behavior scale as the dependent variable. Finally, one's own infidelity was the dependent variable for the fourth ANOVA. Post hoc analyses allowed for direct comparisons of each group or main effect regarding relative levels of insecure attachment styles, risky behaviors, and infidelity encounters. In order to test the hypotheses about the interaction between the gender of the parent who committed infidelity and the gender of the participant, a 2 x 3 ANOVA was used with participant gender and gender of the parent who committed infidelity as the independent variables Gender of the participant was male or female, and gender of the parent who committed infidelity was father only, mother only, or both parents. I predicted that there would be an interaction between gender of the participant and gender of the parent who committed infidelity such that participants whose same sex parent committed infidelity would have the highest levels of own infidelity, while opposite sex parental infidelity would result in lower levels of own infidelity. During the proposal defense, com mittee members were interested in alternate ways to statistically test the findings in order to demonstrate competence in this area and to compare and contrast findings using different statistical procedures. Thus, as an additional major goal of this proj ect, multiple analyses were run for each research question. Factorial ANOVAs and regressions were conducted to help illustrate any significant findings of the study. Any discrepancies between these analyses and the proposed analyses were evaluated and di scussed.

PAGE 33

! 26 CHAPTER III RESULTS Parental Status and Own Infidelity Initially, a One Way ANOVA was run with parental status as the independent variable (NI/ND, I/ND, NI/D, and I/D) and the participant's own infidelity encounters as the dependent variable. Results demonstrated that parental status was associated with one's own infidelity, F (3, 417) = 3.41, p < .05. Post hoc analysis indicated that individuals whose parents remained married but had engaged in infidelity (I/ND) had the highest rates of own infidelity ( M = 1.47, SD = 1.83). This group was significantly diffe rent from individuals whose parents remained married without infidelity (NI/ND; M = .76, SD = 1.26). However, there were no significant differences between the I/ND group and the NI/D group ( M = .90, SD = 1.39) and the I/D group ( M = 1.12, SD = 1.45). Fol lowing up the initial ANOVA, a 2 x 2 ANOVA was run with parental divorce status and parental infidelity status as the independent variables. A significant main effect of the model was found for infidelity, F (1, 416) = 7.22, p < .05, partial 2 = .01. Thu s, the results demonstrated that individuals whose parent(s) had engaged in infidelity had significantly more infidelity encounters themselves. However, there were no other significant main effects or interactions in the model; neither parental divorce st atus nor the interaction between parental divorce and parental infidelity were significant. Next, a regression was run with dummy coded parental status variables (NI/D, I/ND, and I/D) as the independent variables. The NI/ND group was used as the referenc e group for this analysis. Results demonstrated that the overall model predicted a

PAGE 34

! 27 significant, yet small amount, of an individual's own infidelity encounters, F (3, 416) = 3.41, p < .05, R 2 = .02. Consistent with the one way ANOVA results, the only signi ficant predictor in the model was the I/ND group, Std. = .14, p < .05. This demonstrated that the only group with significantly more infidelity encounters than participants in the NI/ND group was participants whose parent(s) had engaged in infidelity, b ut did not divorce. Another regression was conducted with parental infidelity status (0 = no infidelity, 1 = infidelity) and parental divorce status (0 = no divorce, 1 = divorced) as predictors. The overall regression model was also significant, but had a small effect size, F (2, 417) = 4.08, p < .05, R 2 = .02. Consistent with the 2 X 2 ANOVA, parental infidelity status was the only significant predictor in the model, ( = .14, p < .05). This indicated that individuals whose parent(s) engaged in infideli ty were more likely to have more infidelity encounters themselves. Finally, due to the fact that the distribution of own infidelity encounters was significantly skewed, nonparametric tests were run on both the scaled infidelity encounters (responses confi ned to 0 5) and the nonscaled infidelity encounters (responses were used as given, 0 to 40 range). A Kruskal Wallis test was conducted with parental status group (NI/ND, I/ND, NI/D, and I/D) as the independent variable. The test demonstrated that for bot h the scaled and nonscaled infidelity encounters, parental status significantly impacted participants' own infidelity (scaled H (3) = 10.07, p <.05; nonscaled H (3) = 9.68, p < .05). This provided further evidence that parental divorce and infidelity status affected participants' infidelity behavior.

PAGE 35

! 28 Mann Whitney tests were used to further explore this finding on the scaled infidelity encounters to determine where the differenc es between groups occurred. Consistent with the one way ANOVA, it was found that children in the NI/ND group and children in the I/ND group significantly differed, with the I/ND group having more infidelity encounters, U = 3340.5, z = 2.55, p < .05, r = .15. Additionally, children in the I/D group had significantly higher infidelity encounters than children in the NI/ND group, U = 6285.5, z = 2.31, p < .05, r = .13. In summary, the I/ND group had the highest rates of own infidelity. Following this wa s the I/D group, which had the second highest rates of infidelity. Next, the NI/D group had the third highest rates. Finally, the NI/ND group had the lowest rates. Thus, when comparing groups, it is found that the I/ND group is significantly higher than the NI/ND group. Additionally, interaction effects are not found. Rather, there is a main effect of infidelity when collapsing the groups across the levels of parental status (infidelity and divorce). Parental Status and Risky Behaviors Initially, a On e Way ANOVA was run with parental status as the independent variable (NI/ND, I/ND, NI/D, and I/D) and the participant's risky behaviors as the dependent variable. Results demonstrated that parental status significantly affected risk taking behavior, F (3, 429) = 3.34, p < .05. Specifically, individuals whose parents divorced and had not engaged in infidelity (NI/D) had the highest rates of risky behaviors ( M = 2.08, SD = 1.18). This group was significantly different from individuals whose parents remained married without infidelity (NI/ND; M = 1.68, SD = 1.15). However,

PAGE 36

! 29 there were no significant differences between the NI/D group and the I/ND group ( M = 1.79, SD = 1.00) and the I/D group ( M = 2.04, SD = .99). Exactly as had been done with the infidelity o utcome variable, the same analyses were run with risky behaviors as an outcome variable. Following the first ANOVA, a 2 x 2 ANOVA was run with parental divorce status and parental infidelity status as the independent variables. A significant main effect of the model was found for divorce, F (1, 429) = 5.08, p < .05 partial 2 = .01. Results suggested that individuals whose parents had divorced had significantly higher levels of risk taking behavior than individuals whose parents did not divorce. There we re no other significant main effects or interactions in the model; neither parental infidelity status nor the interaction between parental divorce and parental infidelity significantly impacted risky behaviors. Next, a regression was run with dummy coded parental status variables (NI/D, I/ND, and I/D) as the independent variables. The NI/ND group was used as the reference group for this analysis. Results demonstrated that the overall model predicted a significant, yet small amount, of an individual's ris k taking behaviors, F (3, 429) = 3.34, p < .05, R 2 = .02. There were two significant predictors in the model, the NI/D group, Std. = .13, p < .05 and the I/D group, Std. = .11, p < .05. This showed that individuals whose parents had divorced, with or without infidelity, had higher levels of risk taking compared to individuals whose parents had neither divorced nor engaged in infidelity. Another regression was conducted with parental infidelity status (0 = no infide lity, 1 = infidelity) and parental divorce status (0 = no divorce, 1 = divorced) as predictors. This overall model was also significant, but had a small effect size, F (2, 430) = 4.88, p < .05, R 2 = .02. Parental divorce status was the only significant pr edictor in the model, (

PAGE 37

! 30 = .14, p < .05). This result demonstrated that parental divorce status significantly predicted one's risk taking behavior, as parental divorce was associated with an increase in risky behaviors. Parental Status and Attachment Anx ious attachment. Initially, a One Way ANOVA was run with parental status as the independent variable (NI/ND, I/ND, NI/D, and I/D) and the participant's anxious attachment score as the dependent variable. Results demonstrated that parental status did not significantly affect anxious attachment, F (3, 410) = 1.74, p > .05. There were no differences between individuals in the NI/ND group ( M = 3.51, SD = 1.07), individuals in the NI/D group ( M = 3.58, SD = 1.12), individuals in the I/ND group ( M = 3.84, SD = 1.10), and individuals in the I/D group ( M = 3.82, SD = 1.27). Exactly as had been done with the previous outcome variables, the same analyses were run with anxious attachment as an outcome variable. Following up the first ANOVA, a 2 x 2 ANOVA was run wit h parental divorce status and parental infidelity status as the independent variables. Neither of the groups significantly impacted anxious attachment. However, there was a trend for a main effect of infidelity, F (1, 410) = 3.83, p = .05. Individuals wh ose parent(s) engaged in infidelity tended to have more anxious attachment. Next, a regression was run with dummy coded parental status variables (NI/D, I/ND, and I/D) as the independent variables. The NI/ND group was used as the reference group for this analysis. The overall model was not significant, F (3, 410) = 1.74, p > .05, R 2 = .01. However, the I/D dummy coded group showed a trend, = .10, p = .06.

PAGE 38

! 31 Another regression was conducted with parental infidelity status (0 = no infidelity, 1 = infidelity) and parental divorce status (0 = no divorce, 1 = divorced) as predictors. This model also did not achieve significance, F (2, 411) = 2.57, p > .05, R 2 = .0 1. However, parental infidelity status demonstrated a trend towards significance, = .10, p = .05. This result demonstrated a tendency for parental infidelity to be associated with higher levels of anxious attachment. Thus, while there was a trend show n for a main effect of infidelity in both the ANOVA and the regression, when comparing the four groups there was not one single group that emerged as highest in anxious attachment. Avoidant attachment. Initially, a One Way ANOVA was run with parental st atus as the independent variable (NI/ND, I/ND, NI/D, and I/D) and the participant's avoidant attachment score as the dependent variable. Results demonstrated that parental status did not significantly affect avoidant attachment, F (3, 411) = 1.90, p > .05. There were no differences between individuals in the NI/ND group ( M = 2.84, SD = 1.04), individuals in the NI/D group ( M = 2.68, SD = .83), individuals in the I/ND group ( M = 3.05, SD = .98), and individuals in the I/D group ( M = 2.62, SD = .81). Exactly as had been done with the previous outcome variables, the same analyses were run with avoidant attachment as an outcome variable. Following up the first ANOVA, a 2 x 2 ANOVA was run with parental divorce status and parental infidelity status as the indep endent variables. A significant main effect of the model was found for divorce, F (1, 411) = 5.31, p < .05 partial 2 = .01. Contrary to expectations, it was found that individuals whose parents had not divorced had significantly higher ratings of avoidan t attachment. There were no other significant main effects or interactions in the

PAGE 39

! 32 model; neither parental infidelity status nor the interaction between parental divorce and parental infidelity significantly impacted risky behaviors. Next, a regression wi th dummy coded parental status variables (NI/D, I/ND, and I/D) as the independent variables was run. The NI/ND group was used as the reference group for this analysis. The overall model was not significant, F (3, 411) = 1.90, p > .05, R 2 = .01. Another r egression was conducted with parental infidelity status (0 = no infidelity, 1 = infidelity) and parental divorce status (0 = no divorce, 1 = divorced) as predictors. This model overall also did not achieve significance, F (2, 412) = 2.22, p > .05, R 2 = .01. However, parental divorce status was a significant predictor, = .11, p < .05. This result indicated that individuals whose parents never divorced had higher levels of avoidant attachment. Parental Infidelity and Gender Interactions In order to test the relationship between the gender of the parent who engaged in infidelity and the gender of the participant, a 2 x 3 ANOVA would be run with the gender of the participant (male or female) and gender of the parent who engaged in infidelity (father, mother, or both) as independent variables, and the participant's own infidelity as the dependent variable. However, cell sizes for this analysis were quite small which would severely limit statistical conclusions to be made. Thus, the ANOVA was not run a s the power of detecting an interaction for the model was only .24. However, see Figure 1 for an illustration of the group differences. The figure illustrates exactly what would have been expected to happen, as males with father infidelity had the highes t rates of own

PAGE 40

! 33 infidelity, and it appears that there is an interaction between the gender of the parent who engaged in infidelity and the gender of the participant. Table II I 1 Cell sizes for parent and participant genders regarding infidelity behavior "# $%&'&(#)%!*+),+$ ! "#$+)%!-./!0/11&%%+,! 2)3&,+4&%5 6#4+ 7+1#4+ 7#%.+$!8)45 9: ;; 6/%.+$!8)45 : 9< =/%.!"#$+)%> 99 ?? Figure III.1 Relationship between Gender of Parent and Child with Infidelity "! "#$! "#%! "#&! '#(! '#)! '#*! +,-.! /.0,-.! /,12.3!45-6! +712.3!45-6! 8712!9,3.51:!

PAGE 41

! 34 CHAPTER IV DISCUSSION Using a large undergraduate sample, the effects of parental infidelity and divorce were examined on adult children's attachment styles, risky behaviors, and own infidelity. Results demonstrated that both divorce and parental infidelity may impact children in a variety of ways, even into adulthood. Namely, children whose parents were divorced had elevated levels of risk taking behaviors relative to those whose parents remained together. Parental infidelity did not seem to contribute to higher risk taking. Divorce may lead to higher rates of risky behaviors, as parental resources are generally lessened in single parent homes; single parents cannot dedicate as many resources to supervising their children and checking in on what behaviors they are engaging i n. Furthermore, single parent families are often more strained in terms of financial resources compared to two parent homes. This may lead to a lack of extracurricular opportunities for children that may be incompatible with, and, thus, prevent the engag ement in, risky behaviors. Another explanation as to why divorce may lead to risk taking may be that the children of divorce are acting out their frustrations from the shift in family dynamics. Children of divorce may begin engaging in maladaptive coping techniques in order to deal with the loss of a parent, and these coping strategies may become habitual patterns of behavior throughout the individual's development. Conversely, infidelity did not seem to impact the degree to which individuals engaged in r isky behaviors. This study was the first quantitative attempt to examine the relationship between parental infidelity and risk taking, and it does not appear to support the link speculated about in some case studies. The absence of this relationship may be due to the fact that children can see the damaging

PAGE 42

! 35 effects that infidelity (a risky behavior in itself) can have, and choose not to engage in behaviors that may have similar consequences. Another major finding of the current study was that individuals w hose parents committed infidelity had higher rates of infidelity than individuals whose parents did not engage in infidelity. Furthermore, adult children whose parents remained married after infidelity had the highest rates of own infidelity. This may be a result of a modeling process in which children see their parents engage in infidelity, but remain married, and thus there may be greater perceptions that infidelity is an acceptable behavior for them to do as well, or they may simply have more prolonged exposure to the unfaithful parental model. Hypotheses regarding the impact that parental divorce and infidelity would have on adult children's attachment styles were generally not supported. Neither divorce nor infidelity significantly affected participa nts anxious attachment styles; however, individuals whose parents committed infidelity showed a trend towards having more anxious attachments. Even though this effect was quite small, individuals who see their parents engage in infidelity may develop a mo re anxious attachment towards their own significant others. They may desire to be extremely close to their romantic partner, but fear rejection (possibly through an infidelity encounter). It would be interesting to see if individuals whose parents commit ted infidelity were more afraid of their own partners engaging in infidelity than controls; however, more research needs to be conducted on this topic. Few effects were found regarding parental divorce and infidelity status on avoidant attachment as well. One finding that did arise was that individuals whose

PAGE 43

! 36 parents did not divorce had higher levels of avoidance. This finding was quite unexpected, and may suggest that individuals who have parents that remain married have an increased fear of getting close to others. One possible interpretation of this finding is that individuals at a young age may avoid getting too close to others if the relationship does not seem as strong as their parent's. These individuals may avoid getting too close in relationships until they are sure that their partner is "the one". However, this finding has not been established in previous literature, was a small effect, and in fact may be a spurious, Type I, effect. The overall lack of findings regarding parental effects on atta chment was surprising. It seems that children may not fear that their own relationships are bound to end up like their parents, and so they do not obtain insecure attachment styles due to parental behaviors. However, the time at which parents get divorce d or engage in infidelity may have differential effects on their children. If children are younger during these events, they may experience stronger impacts in terms of attachment styles. Unfortunately, this study was not able to control for the age at w hich the infidelity or divorce occurred. Future research should focus on any age influences that may play a role. The hypothesis regarding the interaction of the parent(s)' gender who committed infidelity and the gender of the participant was partially su pported. That is, it appeared that male participants whose father engaged in infidelity had higher rates of infidelity than if only the participants' mother committed infidelity. Additionally, it appeared that females whose mother committed infidelity ha d higher rates of own infidelity than if only their fathers committed infidelity. These patterns suggest a modeling process in which

PAGE 44

! 37 participants look to their same gendered parent for acceptable behaviors to engage in. Males who see their father engage in infidelity, or females who see their mother engage in infidelity, may model the same behavior later in life. It would be interesting to see if this link is upheld by the gender commonality between child and parent, or if it is more closely related to w hich parent the child feels closest to (which presumably occurs more often with same gender dyads). More research with larger sample sizes should be conducted in order to statistically examine this relationship. It is also important to note that the effec t sizes of the significant findings were very small, and accounted for a minimal amount of variance. Thus, although it seems that parental infidelity and divorce may play a role in shaping their adult children's behaviors, there seem to be other factors t hat carry more weight. One variable that has consistently been found to carry stronger effects is relationship conflict between the parents. Parents who have a high conflict relationship and get divorced or engage in infidelity may have a stronger impact on their children. Thus, it would be important to look at the divorce or infidelity situations more closely, including factors such as parental conflict. There were several limitations to the current study. First, the sample was comprised of only undergraduates from a local university. It is unclear how these results would generalize to the overall young adult population and to older community populations. Second, only known parental infidelity was assessed. Participants had to be aware of their parent(s) infidelity, and families in which the children know of an infidelity may have different characteristics than families with unknown parental

PAGE 45

! 38 infidelity. It would interesting to see whether these results would remain consistent for families in wh ich there is parental infidelity, but the children do not know about it. Finally, it is unclear if it was the participants' primary caretaker who engaged in infidelity based on how the question was asked. If participants endorsed mother or father infideli ty, it could have been their biological, adoptive, or stepparent. Thus, a participant whose parents divorced when he or she was very young may have had a biological parent engage in infidelity, but most of his or her experiences were with the stepparent. This may impact some of the modeling conclusions that were developed; however, the amount of participants that this discrepancy would be applicable for would presumably be low. More research needs to be conducted in this area to strengthen the findings o f the current study and address the limitations noted here.

PAGE 46

! 39 REFERENCES Allen, E. S., Atkins, D. C., Baucom, D. H., Snyder, D. K., Gordon, K. C., & Glass, S. P. (2005). Intrapersonal, interpersonal, and contextual factors in engaging in and responding to extramarital involvement. Clinical Psychology: Science and Practice, 12 (2), 101 130. Allen, E. S., & Baucom, D. H. (2004). Adult attachment and patterns of extradyadic involvement. Family Process Special Issue: Meaningful Voices, Old and New, 43 (4), 467 48 8. Allison, P. D., & Furstenberg, F. F. (1989). How marital dissolution affects children: Variations by age and sex. Developmental Psychology, 25 (4), 540 549. Amato, P. R. (1988). Parental divorce and attitudes toward marriage and family life. Journal of M arriage & the Family, 50 (2), 453 461. Amato, P. R. (1994). Life span adjustment of children to their parents' divorce. The Future of Children, 4 (1), 143 164. Amato, P. R. (2001). Children of divorce in the 1990s: An update of the Amato and Keith (1991) met a analysis. Journal of Family Psychology, 15 (3), 355 370. Amato, P. R., & DeBoer, D. D. (2001). The transmission of marital instability across generations: Relationship skills or commitment to marriage? Journal of Marriage & the Family, 63 (4), 1038 1051. A mato, P. R., & Keith, B. (1991a). Parental divorce and the well being of children: A meta analysis. Psychological Bulletin, 110 (1), 26 46. Amato, P. R., & Keith, B. (1991b). Parental divorce and adult well being: A meta analysis. Journal of Marriage & the Family, 53 (1), 43 58. Amato, P. R., & Rogers, S. J. (1997). A longitudinal study of marital problems and subsequent divorce. Journal of Marriage & the Family, 59 (3), 612 624. €ngarne Lindberg, T., & Wadsby, M. (2009). Fifteen years after parental divorce: Mental health and experienced life events. Nordic Journal of Psychiatry, 63 (1), 32 43. Bachman, J. G., O'Malley, P. M., Schulenberg, J. E., Johnston, L. D., Freedman Doan, P., & Messersmith, E. E. (2008). The education drug use connection: How successes and failures in school relate to adolescent smoking, drinking, drug use, and delinquency New York, NY: Taylor & Francis Group/Lawrence Erlbaum Associates. Baydar, N. (1988). Effects of parental separation and reentry into union on the emotional well being of children. Journal of Marriage & the Family, 50 (4), 967 981.

PAGE 47

! 40 Beckwith, L., Cohen, S. E., & Hamilton, C. E. (1999). Maternal sensitivity during infancy and subsequent life events relate to attachment representation at early adulthood. Developmental Psych ology, 35 (3), 693 700. Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment New York: Basic Books. Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self report measurement of adult attachment: An integrative overview. In J. A. Simpson, & W. S. Rholes (Eds.), Attachment theory and close relationships. (pp. 46 76). New York, NY, US: Guilford Press. Carranza, L. V., Kilmann, P. R., & Vendemia, J. M. C. (2009). Links between parent characteristics and attachment variables for college students of par ental divorce. Adolescence, 44 (174), 253 271. Cavanagh, S. E., Crissey, S. R., & Raley, R. K. (2008). Family structure history and adolescent romance. Journal of Marriage and Family, 70 (3), 698 714. Clarke Stewart, K. A., Vandell, D. L., McCartney, K., Owe n, M. T., & Booth, C. (2000). Effects of parental separation and divorce on very young children. Journal of Family Psychology, 14 (2), 304 326. Crowell, J. A., Treboux, D., & Brockmeyer, S. (2009). Parental divorce and adult children's attachment representa tions and marital status. Attachment & Human Development, 11 (1), 87 101. Downing Matibag, T. M., & Geisinger, B. (2009). Hooking up and sexual risk taking among college students: A health belief model perspective. Qualitative Health Research, 19 (9), 1196 1 209. Duncan, G. J., & Hoffman, S. D. (1985). Economic consequences of marital instability. In Horizontal equity, uncertainty, and economic well being M. David and T. Smeeding, eds. Chicago: University of Chicago Press. Emery, R. E. (1982). Interparental conflict and the children of discord and divorce. Psychological Bulletin, 92 (2), 310 330. Fomby, P., & Cherlin, A. J. (2007). Family instability and child well being. American Sociological Review, 72 (2), 181 204. Fraley, R. C., Waller, N. G., & Brennan, K A. (2000). An item response theory analysis of self report measures of adult attachment. Journal of Personality and Social Psychology, 78 (2), 350 365. Franklin, K. M., Janoff Bulman, R., & Roberts, J. E. (1990). Long term impact of parental divorce on op timism and trust: Changes in general assumptions or narrow beliefs? Journal of Personality and Social Psychology, 59 (4), 743 755. Greenberg, E. F., & Nay, W. R. (1982). The intergenerational transmission of marital instability reconsidered. Journal of Marr iage & the Family, 44 (2), 335 347.

PAGE 48

! 41 Guidubaldi, J., Cleminshaw, H. K., Perry, J. D., & Mcloughlin, C. S. (1983). The impact of parental divorce on children: Report of the nationwide NASP study. School Psychology Review, 12 (3), 300 323. Hadley W., Brown L. K., Lescano C. M., Kell, H., Spalding, K., DiClemente, R., Donenberg, G., (2009). Parent adolescent sexual communication: Associations of condom use with condom discussions. AIDS and Behavior 13 (5), 997 1004. Hamilton, C. E. (2000). Continuity and disco ntinuity of attachment from infancy through adolescence. Child Development, 71 (3), 690 694. Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology 52 511 524. Huurre, T., Junkka ri, H., & Aro, H. (2006). Long term psychosocial effects of parental divorce: A follow up study from adolescence to adulthood. European Archives of Psychiatry and Clinical Neuroscience, 256 (4), 256 263. Jacquet, S. E., & Surra, C. A. (2001). Parental divor ce and premarital couples: Commitment and other relationship characteristics. Journal of Marriage & the Family, 63 (3), 627 638. Lansford, J. E. (2009). Parental divorce and children's adjustment. Perspectives on Psychological Science, 4 (2), 140 152. Laumann Billings, L., & Emery, R. E. (2000). Distress among young adults from divorced families. Journal of Family Psychology, 14 (4), 671 687. Lewis, M., Feiring, C., & Rosenthal, S. (2000). Attachment over time. Child Development, 71 (3), 707 720. Lusterma n, D. (2005). Helping children and adults cope with parental infidelity. Journal of Clinical Psychology.Special Issue: Treating Infidelity, 61 (11), 1439 1451. Martin, L. R., Friedman, H. S., Clark, K. M., & Tucker, J. S. (2005). Longevity following the exp erience of parental divorce. Social Science & Medicine, 61 (10), 2177 2189. Mechanic, D., & Hansell, S. (1989). Divorce, family conflict, and adolescents' well being. Journal of Health and Social Behavior, 30 (1), 105 116. Molnar, D. S., Sadava, S. W., DeCou rville, N. H., & Perrier, C. P. K. (2010). Attachment, motivations, and alcohol: Testing a dual path model of high risk drinking and adverse consequences in transitional clinical and student samples. Canadian Journal of Behavioural Science/Revue Canadienne Des Sciences Du Comportement, 42 (1), 1 13. National Center for Health Statistics (2002). Cohabitation, marriage, divorce, and remarriage in the United States. Vital and Health Statistics, 23 (22).

PAGE 49

! 42 National Center for Health Statistics (2008). Health, Unit ed States, with chartbook. Hyattsville, MD. Retrieved November 9, 2009 from http://www.cdc.gov/nchs/data/hus/hus08.pdf. Ryan, S., Franzetta, K., Schelar, E., & Manlove, J. (2009). Family structure history: Links to relationship formation behaviors in young adulthood. Journal of Marriage & the Family, 71 (4), 935 953. Sandberg, D. A., Suess, E. A., & Heaton, J. L. (2010). Attachment anxiety as a mediator of the relationship between interpersonal trauma and posttraumatic symptomatology among college women. Jou rnal of Interpersonal Violence, 25 (1), 33 49. Schwartz, J. E., Friedman, H. S., Tucker, J. S., & Tomlinson Keasey, C. (1995). Sociodemographic and psychosocial factors in childhood as predictors of adult mortality. American Journal of Public Health, 85 (9), 1237 1245. Schwinn, T. M., Schinke, S. P., & Trent, D. N. (2010). Substance use among late adolescent urban youths: Mental health and gender influences. Addictive Behaviors, 35 (1), 30 34. Sori, C. F. (2007). "An affair to remember": Infidelity and its impact on children. In P. R. Peluso (Ed.), Infidelity: A practitioner's guide to working with couples in crisis. (pp. 247 276). New York, NY, US: Routledge/Taylor & Francis Group. Spruijt, E., & Duindam, V. (200 5). Problem behavior of boys and young men after parental divorce in the netherlands. Journal of Divorce & Remarriage, 43 (3 4), 141 156. Teachman, J. D. (2002). Childhood living arrangements and the intergenerational transmission of divorce. Journal of Mar riage and Family, 64 (3), 717 729. Wallerstein, J. S., & Kelly, J. B. (1996). Surviving the breakup: How children and parents cope with divorce Basic Books. Wolfinger, N. H. (2000). Beyond the intergenerational transmission of divorce: Do people replicate the patterns of martial instability they grew up with? Journal of Family Issues, 21 (8), 1061 1086. Zill, N., Morrison, D. R., & Coiro, M. J. (1993). Long term effects of parental divorce on parent child relationships, adjustment, and achievement in young a dulthood. Journal of Family Psychology, 7 (1), 91 103.

PAGE 50

! 43 APPENDIX A ANNOUNCEMENT FOR STUDY PARTICIPATION We are inviting you to participate in a research study entitled "Relationships and Health" being conducted by Beth Allen, Ph.D., a faculty member in the Department of Psychology at the University of Colorado at Denver (Elizabeth.allen@cudenver.edu). This research has been reviewed and approved by the UCD Institutional Review Board. The purpose of the research is to learn more about why and how peopl e make decisions about monogamy. For your participation, you will be asked to complete a fairly lengthy questionnaire (it should take about 1 hour) which includes questions about some personal issues such as sexual behaviors, infidelity, drug and alcohol use, and sexually transmitted infections. Participation is open to all, regardless of experiences with these issues. Because these questions are so personal, we are taking several steps to keep your answers anonymous, so that no one can connect your answ ers to you specifically. Students are being recruited from several psychology classes, with a goal of recruiting a minimum of 400 participants. If you participate for credit, the study coordinator will write down your name so we can tell your teacher you participated, but your name won't be connected to your questionnaire. We think you might learn some interesting things about the research process from your participation, but also are aware that some people find the experience of answering personal quest ions very upsetting. We will let you know about free or low cost counseling options, and also who to talk to if you have questions, concerns, or complaints about the research experience. If you are interested in participating, please contact the study co ordinator, _____________. She will schedule a time with you to take the questionnaire.