Relationships 1

RUNNING HEAD: RELATIONSHIPS

The Effect of Relationship Length, Self-Disclosure, and Relational Closeness on the Presence of
Domestic Violence within Romantic Relationships

Michael Johnson, Lauren Kordalski, Laura Roa, Stephanie Shepherd, Jesse Tessier, and Taylor
Vanasse
Bryant University

Relationships 2

Abstract
The goal of the current investigation is to examine level of disclosure, relational closeness and
length of relationships in relation to incidences of domestic violence. A sample of 90 college
students at Bryant University in Smithfield, RI completed an anonymous hard-copy survey with
questions pertaining to the above topics. Research found no correlation between length of
relationship and level of disclosure. However, data did show a positive correlation between
length of relationship and level of relational closeness. This study also investigated relational
closeness and level of disclosure as potential predictors of domestic violence. Ultimately, not
enough people reported having experienced violence within their relationships for the data to be
considered significant. The findings are further summarized and discussed in the body of the
paper.

Keywords: self-disclosure, relational closeness, domestic violence

Relationships 3

The Effect of Relationship Length, Self-Disclosure, and Relational Closeness on the Presence of
Domestic Violence within Romantic Relationships

Research on the topic of college-aged romantic relationships has been found to be quite
difficult due to the large scope of communication variables involved. Relationships are dynamic
and unique as no two partners have a connection that is completely the same. However, it is clear
that there are universal variables that affect most relationships at least one point in time. One of
the most widely studied topics in this field of research is domestic violence. According to a
national study of adolescent health, Violence between intimate partners is a significant public
health problem and is thought to be most prevalent in early adulthood, (Halpern, Oslak, Young,
Martin, & Kupper, 2001, p.1). Although so much research has been performed in this field, it is
still not clear cut as to what variables specifically contribute to this violence. The current
investigation takes a fresh approach on the topic, and analyzes new variables to determine if they
have any correlation with partner violence.
The goal of this investigation was to ultimately find relationships between four key
variables: length of relationship, self-disclosure, relational closeness, and domestic violence. It
seeks to provide more insight as to what variables may correlate with violence within romantic
relationships. It also seeks to analyze relational closeness and self-disclosure as a function of
length of relationship, as these measurements can contribute to ones overall understanding of the
complex nature of college-aged romantic relationships.

Relationships 4

Level of Disclosure
There are many studies that have been conducted in an attempt to better understand selfdisclosure within interpersonal relationships. The topic of self-disclosure has varying definitions,
but it is most often described as anything intentional or unintentional that communicates to us
what a person is like. In a study performed by Derlega, Greene, and Winstead (2006, p. 3) they
specifically defined self-disclosure as, a deliberate or voluntary activity whereby people reveal
information, thoughts and feelings about themselves to at least one other person during an
interaction, Self-disclosure has been studied within a wide variety of contexts including
romantic relationships, friendships and family relationships.
One of the first researchers to release studies on the topic of self-disclosure was Sidney
Jourard in 1964 (as cited in Greene, Derlega, and Mathews, 2006, para 3). One of the first
advocates of the importance of self-disclosure within relationships, Jourard published books on
the topic and created one of the most widely-used scales to measure disclosure among families
and partners (2006, para 3). Jourards research led to the development of many other relevant
theories and scales. For example, the next major milestone in the self-disclosure field of study
was the creation of the Social Penetration Theory by Irwin Altman and Dalmas Taylor in 1973
(as cited in Derlega, Greene and Winstead, 2006, para 8). This theory proposed that as
relationships progress, self-disclosure increases which allows for partners to penetrate new layers
of anothers personality. Before self-disclosure, partners only know each other on a superficial
level. However, as disclosure increases they know their partner on a deeper and more intimate
level (Chadwick, n.d.). This theory took the topic to a new level, as it introduced multiple
dimensions of self-disclosure. Altman and Taylor proposed that the breadth of topics disclosed,

Relationships 5

the breadth frequency, the topic time, and topic depth were all important self-disclosure
dimensions that were crucial in the development of relationships (as cited in Derlega, Greene,
and Winstead, 2006, para 9).
A major issue that has since emerged in this field of research, is attributions for and
against the level of self-disclosure in relationships. In a study conducted by Derlega, Winstead,
Mathews, and Braitman (2008, p. 4) their data showed that the most important reasons for selfdisclosure included a high level of trust, seeking help, a duty to inform, and similarity to the
significant other. Among the attributes most closely linked to non-disclosure were attempting to
protect the other, maintaining privacy, a superficial relationship, and disclosing the information
was not important to the relationship (2008). Participants in this survey were not only talking
about their romantic relationships, but friendships as well.
Like the study mentioned previously, much of the research conducted on self-disclosure
utilizes hypothetical survey questions to generate data. Less research has been performed on
people who are actually in current romantic relationships. However, in a study performed by
Magsamen-Conrad, Greene, Banarjee and Bagdasarov (2008, p. 11) only people in existing
relationships were surveyed as to reveal factors that influence the effect of self-disclosure in
relationships. Respondents were asked how close they were with their significant other before
the self-disclosure, how soon after the event they disclosed the information, the response to the
disclosure, etc. The data generated from this study demonstrated that the greater the amount of
time that elapsed between an event and disclosure, along with positive responses to disclosure,
the more positive the effect on an existing relationship (2008). While many research studies like
this have examined the effects of self-disclosure on relationships, very few studies have
examined the opposite: how length of relationship impacts level of disclosure. This is one of the

Relationships 6

reasons that the study in this research paper will further examine how romantic relationship
duration affects the amount of self-disclosure between partners.

Relational Closeness
Relational closeness is another variable that has been examined quite frequently in
studies that involve interpersonal communication and it is closely related to self-disclosure. It is
defined by most scholars as being, equivalent to the degree of interdependence between partners
in a relationship (Kelly et. al., 1983). This dependency is established when one persons actions
are likely to produce changes in the feelings or actions of another. Relational closeness can vary
based on certain factors such as rewards and benefits, costs, satisfaction, stability, and security of
any given relationship. This topic has been applied across a number of different fields to describe
relationships within families, romantic partners, and friends.
Before previous research on this topic is discussed, it is important to mention the four
different characterizations of a relationship which are included in the aforementioned definition
of relational closeness. These four dimensions were first proposed in a study performed by
Berscheid, Snyder, and Omoto in 2004. According to the study, the first component of relational
closeness is the strength of influence between partners (Berscheid, Snyder, & Omoto, 2004).
Secondly, is frequency of interaction, or how often couples spend time with their significant
others. The next dimension of relational closeness, according to this study, is the range of
different life situations in which one partner influences the other. This could include influences
exerted in a work environment, family environment, etc. The final dimension involves length of

Relationships 7

relationship. This study suggests that as two partners engage in a relationship for a longer period
of time, their degree of relational closeness is likely to increase (2004).
According to The Handbook of Communication Science and its authors Berger, Roloff,
and Roskos-Ewoldsen, The study of communication and relationship development was fueled
in the 1970s by increasingly sophisticated concepts of interpersonal communication (2010, p.
328). Recognizing a connection between relationships and the communication that takes place
within them, scholars delved further into theories of relational closeness, first focusing on
similarities between partners as a marker of compatibility and closeness. They then transitioned
into studies of how messages differ according to the closeness of the relationship.
Social penetration theory emerged from this research, explaining, the development of
personal relationships is rooted in self-disclosure, which is one aspect of relational closeness, as
discussed previously (Berger, Roloff, Roskos-Ewoldsen, 2010, p. 329). As the theory explains,
relational closeness emerges from self-disclosure, as an increase in personal sharing tends to lead
to an increase in relational intimacy. As time passes and individuals learn more about each
other, they increase the depth of their disclosure by revealing more personal information and the
breadth of disclosure by discussing a greater variety of topics (2010) leading to increased
relational closeness. This theory goes hand in hand with another popular communication theory,
Uncertainty Reduction Theory.
Uncertainty Reduction Theory suggests that, when people meet, they are motivated to
form impressions of each other in order to reduce the uncertainty they have about the other
(Antheunis, Schouten, Valkenburg, & Peter, 2012, p. 5). There are three types of uncertainty
reduction theories that people currently employ: passive strategies, active strategies and
interactive strategies. A passive strategy may include looking at someones nonverbal behaviors,

Relationships 8

and making judgments. An active strategy, on the other hand, involves actively finding out things
about another person. This could include asking people questions about them. Finally, an
interactive strategy includes actually communicating with the person to learn more about them.
All of these strategies help to lower the levels of uncertainty towards another person, and can
therefore increase the degree of closeness between the two. This closeness builds over time as
uncertainty is reduced through more frequent interactions. In one study by Theiss and Solomon,
the pair concluded that, a decrease in uncertainty was a significant predictor of intimacy
(2008, p.1), emphasizing the importance of certainty as an indicator of relational closeness.
These theories highlight an important aspect of relational closeness. As Derlega, Metts,
Petronio, and Margulis point out, relationship development is, an ongoing process as partners
negotiate a relationship definition, attempt to meet their individual goals, attempt to fit
comfortably within social and family networks, and so forth (1993, p. 9). This ongoing process
involves the relationship moving in different directions, some of which may include a destructive
relationship pattern of domestic violence, a variable in the current study.
As equity theory states, people are motivated by the rewards they receive from a
relationship. This is why many couples remain committed to each other, because as their
relational closeness builds, so do the rewards. And as the relationship develops, they are
motivated to maximize the rewards they receive and minimize their costs but they also
understand that certain norms and rules guide social relationships (2010). In certain close
relationships, these norms include domestic violence in varying forms.
The investment model attempts to explain why partners may decide to stay in these
relationships regardless of their level of destruction. The model specifies that, satisfaction is a
function of the reward value of a relationship, relative to expectations, and that commitment

Relationships 9

stems from the superiority of a relationship, compared with viable alternatives (2010). As this
model theorizes, the reward value of a relationship grows over time which begs the question,
does greater relational closeness lead to increased incidences of domestic violence?

Domestic Violence
Domestic violence within interpersonal relationships is a topic that has been widely studied,
especially within romantic relationships. The United States Department of Justice defines
domestic violence as a pattern of abusive behavior in any relationship that is used by one
partner to gain or maintain power and control over another intimate partner. [It] can be physical,
sexual, emotional, economic, or psychological actions or threats of actions that influence another
person (2014, p. 1). Although all are applicable to the field of communication, most of the
studies that have been conducted focus on the physical and emotional contexts of the subject.
Research on domestic violence and its effects on interpersonal relationships became
prevalent in 1970, when federal, state, and local legislation began to pass in favor of protecting
victims (Bettinger, 2011, p. 15). Studies conducted in recent decades have used existing models
and theories of communication to learn more about domestic violence, and have also been
focused heavily on conflict styles and the violence associated with different types. In one
investigation, Feldman and Ridley looked at the role of conflict-based communication responses
and outcomes in male domestic violence toward their female partner. It was found that the
process of conflict has 4 distinct components: conflict of interest, conflict orientations, conflict
responses, and conflict outcomes, all of which vary for males that have abusive tendencies versus
those that do not (2000, p. 552). Three years later, Feldman and Ridley conducted another study

Relationships 10

that looked at the opposite scenario of conflict-based communication in these four components in
relation to female domestic violence toward their male partners. It was discovered that in
comparison to males, females use many more verbal abuse tactics whereas males use more
physical strategies to cope with conflict, and there are many gender differences in resolution of
abusive events as well (2003, p. 157).
Another area studied specifically related to domestic violence is the effect media and its
portrayals of abuse have on romantic relationships and the violence within them. One study in
particular done by Nettleton addressed the representations seen within magazines. Women are
generally characterized as choosing the wrong men to be in relationships with, yet men are not
held responsible for their abusive actionstherefore creating a societal expectation of
responsibility between genders. By using the theories of representation and absent
representation, the study recognized four themes of domestic violence portrayals in media:
women are responsible and cannot help themselves, violence is scary to men but funny to men,
statistics are often distorted to overlook violence towards women, and the only way to prevent it
is to have the victim separate themselves from their abuser (2011, p. 148). Current studies are
using this information to look at media representation from a male perspective, and sparked the
theme of studying domestic violence and media as equal contributors.
Although the types of domestic violence, causes, differences in gender, and communication
variables related to it have been studied, it is important to be more specific in this research paper.
Rather than focusing only on the conflict styles and communication that spark abusive episodes,
this study looks at self-disclosure and relational closeness in a relationship, and if either have any
impact on the presence of domestic violence. These elements of communication are not usually

Relationships 11

associated with abuse, and it is interesting to see if they have any relationship or influence on the
prevalence of it.

Hypothesis and Research Questions

The goal of this research study is to examine level of disclosure and relational closeness
with regards to length of college-age romantic relationships and episodes of domestic violence
within them. It was predicted that the higher the level of self-disclosure and degree of relational
closeness, the less likely the partners were to engage in domestic violence. Additionally, it was
predicted that there would be higher levels of relational closeness and disclosure in relationships
that were longer. Therefore, the following hypotheses are posed:

H1: Participants who report lower levels of self-disclosure within their romantic relationship,
are more likely to experience episodes of domestic violence, perpetrated by their partner.

H2: Participants who report lower levels of relational closeness with their partner are more
likely to experience episodes of domestic violence within their romantic relationship.

H3: Participants who have been in relationships for longer periods of time, will report higher
levels of self-disclosure and relational closeness.

Relationships 12

In the absence of previous research focusing on domestic violence with regards to

relational closeness and level of disclosure, the following research questions were formed:

RQ1: Does length of relationship impact level of disclosure?

RQ2: Does length of relationship impact level of closeness?

RQ3: Are level of disclosure and relational closeness related to domestic violence?

Method
Manipulation
The most obvious and effective option for the research team was to reach out to those
close to them to take part in answering the survey. The reasoning behind this was not a matter of
commodity, instead, it was more logical to present the survey to those people that were thought
to have been in a relationship before. The outcome of this was that people in relationships
answered more surveys than those people that were not in relationship. Also, all 6 members of
the research team were currently attending Bryant University, and 5 members of the team were

Relationships 13

living on campus. Due to the convenience of this situation, the majority of the participants were
attending Bryant University at the time.

Procedure
Upon receiving approval to distribute the survey amongst the Bryant community, the
research team began presenting the survey to participants right away. Those involved in the
survey were volunteers who were recruited through friendships and acquaintances. The survey
was presented to participants in the format of a hard copy only which they filled out and then
turned in. The questions would be answered by the use of a scale from 1 through 7. The team
believed that this would render the most reliable results, therefore all of the participants took the
survey in person. The consent form was given to the participant to sign first, and once they did
they were asked to answer the questions accordingly and to the best of their knowledge. There
was no reward for completing the survey. The surveys were completed at no specific time of day,
but were taken and completed at many different times and places. The team conducting the
research was readily available as the participants took the survey to answer questions or clarify
any doubt regarding the material.

Participants
Participants ages ranged from 18 years old to 30 years old. There were 18 participants
that were 18 years of age, 15 people said they were 19 years old, 22 participants were 20 years
old, 32 participants were 21 years old, 6 participants were 22 years old and there was one
participant who was 30 years of age. The average age for the entire group of participants is 20

Relationships 14

years (20.42 to be exact). For the participants student class; 18 were freshman, 19 were
sophomores, 20 were juniors and there were 21 seniors. Out of everyone who took the survey, 79
participants were currently in a romantic relationship at the time they were taking the survey, 10
people were not. Of the 80 participants who were currently in a relationship, 5 of them have been
in their relationship for 0 to 3 months, 7 have been in their relationship for 4 to 6 months, 16
have been in a relationship for 6 months to 1 year, 19 have been in a relationship for 1 to 2 years,
and 33 people have been in a relationship for at least 2 years.
The data used for this study involved the completion of several surveys which were filled
in by Bryant University students. Participants were asked several questions regarding different
aspects of their relationship with a significant other. Participants were first asked whether or not
they were in a relationship, if they circled yes, they would continue the survey and if they circled
no, they would stop the survey and would not continue on. The participants who took part in the
study were people that knew the different members of the research group. Naturally, the research
group asked more people to take the survey that they knew were already in a relationship. The
surveys were anonymous and any participants personal information will remain confidential.

Results
Reliability
One of the first measures that was examined in the SPSS results, was the reliability of the
self-disclosure and closeness scales. The reliability was measured using Chronbachs alpha. This
measures each scale in terms of how closely related the set of items in each scale are. In regards

Relationships 15

to the first scale, which measured self-disclosure in 31 questions, the results came back with a .
79 which was acceptable. The next scale measured relational closeness in 11 questions and
resulted in a .93 reliability which is also considered acceptable.

Hypothesis and Research Questions

The first hypothesis predicted that participants who report lower levels of self-disclosure
within romantic relationships are more likely to experience episodes of domestic violence
perpetrated by their partner. In order to test this hypothesis, it was necessary to run an ANOVA
because there were three answer options. Ultimately, the data did not support the hypothesis.
This is because in order for an ANOVA to be significant, at least thirty people need to respond to
the question and this number was not reached.
The second hypothesis stated that participants who report lower levels of relational
closeness with their partner are more likely to experience episodes of domestic violence within
their romantic relationship. Once again, there was no data to support this hypothesis due to the
fact that there were not at least thirty respondents.
The third hypothesis was that participants who have been in relationships for longer
periods of time will report higher levels of self-disclosure and relational closeness. In order to
test this hypothesis, a correlation was generated between length of relationship and level of
disclosure. The reason a correlation was performed on this is because it is relating two interval
variables with each other. Although the results came back with a r-value of 0.57, indicating a
moderate relationship between length of relationship and level of disclosure, the results are not
significant because the p-value was .641, which is greater than .05 (see Table 1).

Relationships 16

Next, as demonstrated by Table 2, a correlation was run between length of relationship

and level of relational closeness. When this correlation was run, it was found that the relationship
was moderate because the r-value was .344. The results are also significant because p is less than
.05, the exact value being .002.

Violence
After using an ANOVA to analyze the results, no conclusions could be made because not
enough participants responded to the violence questions.
The following is an interpretation of the data reported in the chart that stated the results of
tests of between-subjects effects for VIO1 with Disclosure as the dependent variable (Table
3). For Type III Sum of Squares, the output value was .006. Generally, the sum of squares
measures the differences caused by effects and those caused by chance. Because the value
reported here is low, we can assume that any differences in this case were caused by chance and
not because of a measurable difference due to manipulative effects of the study. For the second
column, which reports degrees of freedom, the value was 1. This indicates that there is 1 degree
of freedom for the between-groups estimate of variance. Considering that there could either be
violence or none as options in VIO1, this makes sense due to the fact that df is calculated with N1, and in this case N=2. 2-1=1. The next column, containing the value for the mean square,
reports the value .006. This was calculated by dividing the sum of squares by the degrees of
freedom (.006/1) and this value is the estimate of variance between-groups. Because this is a low
value, one could interpret this as saying there is low variance between-groups in terms of VIO1,
however this could also be influenced by the small number of participants that answered this

Relationships 17

question versus those in the majority that didnt. The fifth column gives the F ratio. It is
calculated by dividing mean square between-groups by mean square within-groups. The F value
in this case for the data was .035. The final column gives the significance of the F ratio. This is
the p value. Keeping in mind that if the p value is less than or equal to the level, which is most
often .05, then reject the H0 that suggests all the means are equal. The value in this case for our
data was .855. Because this value is greater than the a level, there is not enough evidence to
claim that some means differ, and thus we cannot reject H0. Based on this, a fair interpretation
would say that the data suggests that disclosure doesnt impact relational violence, and any
changes that occurred are more likely attributable to chance rather than correlation.
Table 4 displays the same type of data only in this case responses from VIO2 were the
source. The type III sum of squares in this case was considerably higher than in the previous
chart, which could imply that VIO2 results were less likely attributable to chance due to this
value (.706). The value for degrees of freedom was 2, which in this case (remembering that this
is derived from N-1) shows that 2 types of violent behavior were reported. By dividing the sum
of squares value by the degrees of freedom (.706/2) one gets the value of the mean square, which
is .353. The F value here is 1.323, and the significance level is .362. Because this value is greater
than a level, .05, there is not enough evidence to claim that some means differ, and thus we
cannot reject H0. While VIO2 had more degrees of freedom and a higher sum of squares, the data
ultimately was not significant, which does not suggest any correlation between disclosure and
episodes of violence within relationships.
Table 5 uses VIO1 as a source for data, but this time the dependent variable is closeness.
The type III sum of squares was .002, with 1 degree of freedom, which again may be indicative
of too little data from VIO1 and little difference between effects, with any occurring likely being

Relationships 18

attributed to chance. The mean square and F value results are each .002 which makes sense
considering the above. The significance or p-value in this case was .967, which is drastically
high and thus suggests little significance in changes in relational violence due to level of
closeness, and that differences are more likely due to chance. Because the p-value is greater than
a of .05, the null hypothesis cannot be rejected.
Finally, Table 6 displays data from VIO2 in relation to closeness. The type III sum of
squares in this case was 1.429, which again, in reference to the other VIO2 data analysis above,
shows that there was a greater difference between effects and could suggest that this is not due
simply to chance. There are 2 degrees of freedom (N=3, N-1=2) which shows 2 types of violence
reported, and dividing 1.429 by 2 (sum of squares/df) gives the mean square of .714. The
corresponding F value is .490. The p-value for this data is .645, which is greater than .05 and
because of this, the data cannot be considered significant. Due to this, the null hypothesis cannot
be rejected.

Discussion
The findings that were reported can lead to a number of key conclusions. The first
conclusion that will be discussed, relates to length of relationship and levels of self-disclosure.
The data did not provide any statistically significant results. The results show that with a p value
of .641, 64% of the data that shows a moderate relationship between length of relationship and
level of disclosure within the study, occurred due to chance. This also means that only 36% of
the results can be explained by one impacting the other, which is not enough to suggest that
length of relationship and level of self-disclosure are in some way related to each other.

Relationships 19

In regards to the correlation that was run between length of relationship and level of
closeness, the data showed that the relationship between the two variables was significant.
Because the correlation coefficient is significant, there exists some type of relationship between
the two variables, which allows one to conclude that due to their association, as length of
relationship increases, the level of closeness within said relationship increases as well.
No conclusion was reached for the third research question which was looking at the
relationship between the level of disclosure, relational closeness, and domestic violence. Only 14
people out of the 90 people surveyed, reported that there had been instances of domestic violence
within their relationships. There was not an adequate amount of responses to run an ANOVA,
which requires at least 30 people to have answered the questions. It could potentially be assumed
that domestic violence is not very prevalent in college aged relationships, or that people were not
willing to disclose such sensitive information. The potential reasons for this outcome will be
discussed in further detail in the limitations section of the Discussion.

Significance of Findings
Many other studies have been performed in an attempt to better understand relational
closeness. The findings in this research paper have demonstrated that length of relationship is
positively related to this perceived closeness, but what other factors could be playing a role in
this relationship? Perhaps the results can be a reflection of another variable, such as perceived
level of commitment. In a 2008 study performed by Daniel Weigel, data demonstrated that as a
relationship progresses over time, partners will either increase or decrease their perceived degree
of mutuality of commitment (2008). If they sense that their partner is becoming increasingly

Relationships 20

committed to the relationship, this tends to trigger increased communication between the two.
Partners who believe that their significant other has a similar sense of commitment to the
relationship will develop an increased sense of security and will strive to keep the relationship
intact. Therefore, partners that perceive each other to have similar sentiments about the
relationship, tend to stay in relationships for longer periods of time. Thus, the degree of closeness
that people report feeling towards their partner, could be a result of many factors.
Past research has also placed an emphasis on the importance of relational satisfaction as a
function of relational closeness. Evidence has demonstrated that closeness between partners is
positively correlated with relationship satisfaction (Cropley & Reid, 2008). Like the study
previously analyzed, researchers found that this satisfaction and closeness tended to increase
when partners felt that their significant others had mutual feelings about satisfaction levels
within the relationship. These findings would make sense in light of our new findings, as mutual
feelings would lead to more closeness, which would ultimately lead to a longer relationship.
However, the findings were not all consistent with previous research with regards to the
variable self-disclosure. In this study it was found that there was not a strong enough relationship
between length of relationship and level of self-disclosure for it to be considered significant. In
one of the earliest studies performed with regards to self-disclosure, researchers Altman and
Taylor found that increased self-disclosure had positive outcomes within relationships (1973).
These findings were supported in a more recent study that surveyed people in existing
relationships. The data showed that as self-disclosure increased there were positive effects on the
relationship such as increased closeness (Magsamen, Green, Banerjee, & Bagdasarov, 2008). The
study went on to find that as people were in relationships for longer durations, episodes of selfdisclosure had less of an effect on the relationship. It was very rare that participants reported a

Relationships 21

self-disclosure as harmful to a relationship, although it was present in a few cases (2008).

Therefore, the findings in this research paper that were related to self-disclosure were not very
significant in light of what was already known about the topic. Due to the fact that most studies
saw self-disclosure as positively impacting a relationship, one would guess that self-disclosure
would positively correlate with length of relationship.
Looking at past research on the topic of domestic violence, it is surprising that this study
did not yield any significant results. A study was conducted in 2004 that involved participants in
31 universities across 16 countries. At the median university, it was reported that 29% of students
had experienced some sort of relationship violence (as cited in Murraya & Kardatzke, p. 2).
However, percentages were quite varied across the universities, and rates ranged from 17% to
45%. A similar study performed on college-aged students found that 27.5% of females had been
victims of sexual violence within their relationships (as cited in Murraya & Kardatzke, p. 2).
With all of this data suggesting that dating violence is so common across college campuses, it is
unusual that the reported rate of domestic violence at Bryant was so low. However, this can be
attributed to a problem that has come up so frequently in other studies involving violence.
College-aged students can be hesitant to report instances of violence, which has skewed the
results reported in many other research papers. Researchers DeKeseredy and Kelley performed
research on this topic in 1993 and found that reported rates are significantly underestimated (as
cited in Murraya & Kardatzke, p. 3). This can be due to subjective definitions of the term
violence and the unwillingness of victims to share this sensitive information. Therefore, the
findings in this research study can potentially be significant in that they support the notion that
rates of violence are significantly underreported.

Relationships 22

Limitations and Future Directions

The research presented was complex and strategic in regards to making connections
between length of relationship, level of disclosure, level of closeness, and presence of domestic
violence, however there are some changes that should be made in future research. First, more
people need to participate in the study by taking the survey. This study did not have enough
people to run the ANOVA test on the investigation of domestic violence episodes; making the
results inconclusive. If a larger demographic of people in relationships in could be reached, then
more responses would be recorded that could make the results statistically significant. It is
possible that due to location and the type of population that was targeted that the participants did
not feel comfortable disclosing sensitive information about violent episodes within their
relationships, or that due to the limited number of college-aged people were asked, the majority
of them are indeed all in healthy relationships. Either way, a larger number of participants in the
study would help eliminate this concern.
Next, a wider variety of participants should be asked to take the survey in order to add
another dimension to the research. Instead of asking people who are currently in relationships to
be involved with the study, it would be possible to expand that demographic to include those
who have previously been in relationships as well. This may increase the numbers of previous
abusive episodes seen, and could lead to further research on how the variables of length of
relationships, level of disclosure, and level of closeness relate to those who have gotten out of
violent relationships and how long it took, etc.
Lastly, the formatting and wording of parts of the survey that appeared to be unclear or
challenging for participants to understand, could have been changed. This includes the first

Relationships 23

question on the level of disclosure scale that uses the words when I wish, which was
misinterpreted by some participants and impacted their answers, and therefore the results. A
different scale could also be used to measure episodes of domestic violence, since the scale
seemed to be confusing and people answered incorrectly, which again falsely impacted the test
results and significance findings.
These changes and limitations aside, the current research presents opportunities for
further investigation into elements of communication and their relation to domestic violence. In a
study conducted by McLaren and Steuber, the communication privacy management theory and
appraisal theories of emotion were applied to discovering how emotions such as hurt, fear, and
anger, and communicative response theories including approach and withdrawal strategies to
boundary turbulence correspond with relational outcomes (2012, p. 606). A similarly structured
test could be conducted that uses communication privacy management theory, theories of
emotion, and communicative response theories to explain what occurs or helps explain episodes
of domestic violence within a romantic relationship. Boundary turbulence may also be another
variable or contributing factor that could be included in the study as well.
Another direction for the future that could be considered is how these elements of
communication, whether it be the variables within this study or otherwise, are affected after
abusive transgressions within a relationship. Vallade and Dillow explored extradyadic
communicative messages following relational transgressions in a relationship. They conducted a
longitudinal study that looked at transgression-minimizing and maximizing events and the
communication that led up to it, and as stated by the report, these findings support the
importance of examining negative extradyadic communication about relational difficulties in
romantic relationships (Vallade & Dillow, 2014, p. 109). A similar study could be conducted in

Relationships 24

the future using the longitudinal methodology, and looking at how communication variables
change or are affected by violent episodes in a romantic relationship. There are multiple
opportunities that this research presents for future explorations and discoveries, as well as
opportunities for improvement on what was conducted.

Relationships 25

References

Altman, I., & Taylor, D. (1973). Communication in interpersonal relationships: Social

Penetration Theory. In M. E. Roloff and G. R. Miller (Eds.), Interpersonal processes:
New directions in communication research, 257-277. Newbury Park, CA. SAGE
Publications

Antheunis, M. L., Schouten, A. P., Valkenburg, P. M., & Peter, J. (2012). Interactive Uncertainty
Reduction Strategies and Verbal Affection in Computer-Mediated Communication.
Communication Research, 39(6), 757-780.

Berger, C.R, Roloff, M.E., Roskos-Ewoldsen, D.R. (2010). The Handbook of Communication
Science 2nd Edition. Thousand Oaks, CA. SAGE Publications.

Berscheid E., Snyder M., Omoto A. (2004). Measuring closeness: The Relationship Closeness
Inventory revisited. In Mashek D. J., Aron A. (Eds.), Handbook of closeness and intimacy
(pp. 81-102). Mahwah, NJ: Erlbaum

Bettinger, C., & Natelson, R. (2011). Domestic Violence in the United States: Domestic Violence
and the Criminal Justice System, I, 15-28.

Chadwick, P. (n.d.). Social Penetration Theory. Retrieved from

http://oregonstate.edu/instruct/theory/spt.html.

Relationships 26

Cropley, C. J., & Reid, S. A. (2008). A Latent Variable Analysis of Couple Closeness,
Attributions,and Relational Satisfaction. Family Journal, 16(4), 364-374.

Derlega, V., Winstead, B., & Greene, K. (n.d.). Self-Disclosure and Starting a New Relationship
(Vol. I, pp. 153-174).

Derlega, V. J., Winstead, B. A., Mathews, A., & Braitman, A. L. (2008). Why Does Someone
Reveal Highly Personal Information? Attributions for and against Self-Disclosure in
Close Relationships. Communication Research Reports, 25(2), 115-130.

Derlega, V.J, Metts, S, Petronio, S. Margulis, S. (1993). Self-Disclosure. Newbury Park,

CA. SAGE Publications

Feldman, C. M., & Ridley, C. A. (2000). The role of conflict-based communication

responses and outcomes in male domestic violence toward female partners. Journal Of
Social & Personal Relationships, 17(4/5), 552.

Halpern, C. T., Oslak, S. G., Young, M. L., Martin, S. L., & Kupper, L. L. (2001). Partner
Violence Among Adolescents in Opposite-Sex Romantic Relationships: Findings From
the National Longitudinal Study of Adolescent Health. American Journal Of Public
Health, 91(10), 1679-1685.
Kelly H. H., Berscheid E., Christensen A., Harvey J. H., Huston T.L., & Levinger G. (1983).

Relationships 27

Close Relationships. New York, NY: Freeman.

Magsamen Conrad, K., Greene, K., Banerjee, S. C., & Bagdasarov, Z. (2008). Self-disclosure
in Existing Relationships: Response and Relational Closeness. Conference Papers -National Communication Association, 1.

McLaren, R.M., & Steuber, K.R. (2013). Emotions, communicative responses, and relational
consequences of boundary turbulence. Journal of Socail & Personal Relationships, 30(5),
606-626.

Murraya, C. E., & Kardatzke, K. N. (2007). Dating Violence Among College Students: Key
Issues for
College Counselors. Journal Of College Counseling, 10(1), 79-89.

Nettleton, P. (2011). Domestic Violence in Men's and Women's Magazines: Women Are
Guilty of Choosing the Wrong Men, Men Are Not Guilty of Hitting Women.Women's
Studies In Communication, 34(2), 139-160.

Ridley, C. A., & Feldman, C. M. (2003). Female Domestic Violence Toward Male
Partners: Exploring Conflict Responses and Outcomes. Journal Of Family Violence,
18(3), 157-

170.

Theiss, J. A., & Solomon, D. H. (2008). Parsing the Mechanisms that Increase Relational

Relationships 28

Intimacy: The Effects of Uncertainty Amount, Open Communication About

Uncertainty, and the Reduction of Uncertainty. Human Communication Research, 34(4),
625-654.

Vallade, J.I., & Dillow, M.R. (2014). An Exploration of Extradyadic Communicative Messages
Following Relational Transgressions in Romantic Relationships. Southern
Communication Journal, 79(2), 94-113.

Weigel, D. J. (2008). Mutuality and the Communication of Commitment in Romantic

Relationships. Southern Communication Journal, 73(1), 24-41.

Relationships 29

Table 1
Correlation for Length of Relationship and Level of Disclosure

Table 2
Correlation for Length of Relationship and Level of Closeness

Relationships 30

Table 3
ANOVA for VIO1 and Level of Disclosure

Table 4
ANOVA for VIO2 and Level of Disclosure

Relationships 31

Table 5
ANOVA for VIO1 and Level of Closeness

Table 6
ANOVA for VIO2 and Level of Closeness