Landgrave, Carly

Domestic Violence and Marital Homogamy

Faculty Mentor: Renata Forste, Sociology

Introduction:
Domestic violence is one of the most pressing global human rights issues of the 21st century. It
is estimated that roughly half of the women killed globally in 2012 were killed by intimate
partners or family members (U.N. Facts and Figures, 2012). In 2014, three in ten women in
Egypt reported that they had experienced domestic violence during their lifetime and almost onefifth
reported that they were the target of an episode of spousal violence in the past twelve
months. More than one third of these women experiencing spousal physical or sexual violence
were injured as a result of the violence, and in 2014, 7 percent had serious injuries. Of those who
experienced violence, only a third sought help (Ministry of Health and Population, 2015).
As such figures publicize the circumstance of women in Egypt and other countries, key global
institutions and country leaders call for worldwide reform. In response, researchers aim to
identify factors which serve as inhibitors of domestic violence. Much of the current research
concludes that an increase in a wife’s status decreases the occurrence of domestic violence
(Gibson, 2005; Schuiler, 1996; Monozea, 2005). To further address this issue, I examine how a
woman’s relative status (in comparison to her husband’s status) is associated with domestic
violence in Egypt – a country with high rates of family abuse.

Methodology
Data are from the Demographic Health Surveys (DHS) conducted in Egypt in 2014. The
sampling method was a multistage cluster sample based on the 2014 census. The census included
a survey of ever-married women age 15-49 years of age. The survey is unique in that it asks
detailed questions concerning domestic violence. I limit my analyses to women in a relationship;
thus my final sample size is 20,418 women.
To evaluate the data, I use logistic regression. The outcome variable is a binary indicator of
whether or not a participant has ever experienced domestic violence and is estimated using
binary logistic regression. The results are presented as odds ratios, which represent the increase
or decrease in the odds of experiencing violence associated with a unit or category change in the
independent variables.

Results
Each year increase in a respondent’s education is associated with a four percent reduction in the
odds of ever experiencing domestic violence. Each year increase in the gap between a respondent
and her husband’s education is associated with a two percent reduction in the odds of ever
experiencing domestic violence. Each year increase in a respondent’s age is also associated with
a two percent reduction of the odds of experiencing domestic violence.
Making decisions about visiting family members is associated with a reduction in the odds of
ever experiencing domestic violence. Women who have a say in visiting their families are 22
percent less likely to report ever having experienced violence. An increase in employment is
associated with an increased odd of ever experiencing domestic violence, as is an increase in the
number of children a respondent has. Women who are employed are 20 percent more likely to
have experienced domestic violence. Additionally, for every additional child, women are nine
percent more likely to have experienced domestic violence.

Discussion
Based on the literature, I had hypothesized that women would face the least amount of domestic
violence when their status was most similar to their husband’s status; however, my results do not
support my hypothesis. In the review of current literature, I find that women are less likely to
face domestic violence when they are of similar educational status to their husbands (Akmatov,
2008). In contrast, my data demonstrate that women are less likely to ever have experienced
domestic violence when their husbands have more education than they do.
The literature is divided on the topic of occupational homogamy and finds both that women who
work are more likely to face domestic violence, and that women who work are less likely to
experience domestic violence, depending on the study (United Nations Statistical Commission,
2009; Franklin, 2012; Flake, 2005). Based on the variables, I was not able to directly compare
respondent to husband occupational status. However, I was able to measure the relationship
between women who are currently working, and chances of ever having experienced domestic
violence. In concurrence with research that demonstrates that women who work are at greater
risk of violence, my research indicates that there is a positive relationship between being
employed and ever having experienced domestic violence.
In evaluating age homogamy, there is no significance in the relationship between comparative
age and experience with domestic violence. Thus my study does not support the literature which
concludes that women who are older than their husbands are less likely to face domestic violence
(Chakwana, 2004).
In each homogamy study, the relationships remain even after controlling for urban/rural living,
socioeconomic status, children ever born, and autonomy. My control variables also offer insight
into various other factors predictive of domestic violence. The literature documents that
increasing a woman’s education, regardless of her spouse’s education, decreases her chances of
experiencing domestic violence (Monozea, 2005; United Nations Statistical Commision, 2009).
In controlling for individual education, my findings concur with these studies, indicating that an
increase in a respondent’s education is related to a decrease of ever having experienced domestic
violence. Similar to previous research, my research also indicates that women with more children
are more likely to have ever experienced domestic violence (Yount and Carrera, 2006; Demaris
et al, 2003).

Finally, my study indicates that women who have more autonomy of movement are at lower risk
of ever experiencing domestic violence, and that autonomy over visitation of family members is
the only significant autonomy measure in reducing chances of ever have experienced violence.
This suggests that women who are unable to visit their family members may be more at risk of
violence, which supports the current literature (Yount, 2005).

Conclusion
This study is limited in that the data are point in time data. For this reason, the results cannot
establish causation in any of the relationships. These findings highlight the strength of education,
in decreasing domestic violence, and the importance of having autonomy over movement. The
results add to the dialogue concerning the relationship between female employment and domestic
violence. They call for further investigation into the influence of familial relationships after
marriage on domestic violence.