Download In the landlocked country of Bolivia, the indigenous and post

In the landlocked country of Bolivia, the indigenous and post-colonial cultures have
interacted and influenced one another for hundreds of years. As these cultures have merged,
they have formed a unique set of social and cultural norms that govern the Bolivian peoples’
attitudes and behaviors towards all aspects of life. These social and cultural norms play both a
constructive and deconstructive role in creating gender roles, identities, and expectations. They
also shape what are considered appropriate and inappropriate behaviors between men and
women.
A behavior that is prevalent in Bolivia is violence against women. The 2008
Demographic and Health Survey (DHS) estimates that 47% of women, married or in common
unions, have experienced psychological, physical, or sexual violence in the past 12 months.1 In
2010, 72 women in Bolivia were victims of murder. Records show that 42% of murder victims
were between 18 and 30 years old and that their intimate partners were responsible for more than
50% of their deaths.2 In 2010, 4,515 women were admitted to hospitals due to serious violence
related injuries.3 A cross-country study led by Victor Asal and Mitchell Brown established that,
in comparison with 31 other countries, violence against women in Bolivia can be considered
widespread.4
This project aims to unpack the prevalence of violence against women in Bolivia and
identify key risk factors that are present. Due to Bolivia’s demographic diversity, it is important
Coa, Ramiro and Luis H. Ochoa. “Encuesta Nacional de Demografia y Salud (DHS) 2008.” Instituto Nacional de
Estadistica adn Ministerio de Salud y Deportes, October 2009, 240
2
“Se registraron 117 asesinatos de Mujeres.” El Diario La Paz- Bolivia. 25 November 2010.
http://enfermeria.bvsp.org.bo/cgi/sys/s2a.xic?DB=B&S2=2&S11=21319&S22=b Date of Access: 15 January 2011.
3
“En 11 meses, 4,514 mujeres llegan a hospital por violencia.” Los Tiempos 26 November 2010.
http://enfermeria.bvsp.org.bo/cgi/sys/s2a.xic?DB=B&S2=2&S11=21321&S22=b Date of Access: 15 January 2010.
4
Asal, Victor and Mitchell Brown. A Cross-National Exploration of the Conditions that Produce Interpersonal
Violence. In Politics and Policy. (38, 2) 2010, 175.
1
to analyze these factors by state. By using GIS to plot key risk factors one is better able to
present data and prescribe policy interventions.
Methodology
This project will use Lori Heise’ 1998 Violence against Women: An Integrated,
Ecological Framework, which incorporates a multilevel and multi-factor approach to analyzing
violence against women. The Ecological Framework is made up of four levels of analysis that
include the interplay between individual characteristics, relationships, community, and society.5
Within each level, Heise outlines a series of risk factors that are considered true casual factors
that affect the likelihood of an individual becoming a victim of intimate partner violence. A list
of all the factors that lead to women being victims of violence are included in Appendix A.
In this analysis, I have chosen to focus primarily on four key community-based indicators.
They are: female and male illiteracy, female and male unemployment, and poverty. Since
county or municipality data could not be found, the information is plotted at the state
administrative level. In addition, I have opted to add an indicator: the number of individuals that
identify themselves as indigenous. Since this specific data point could not be found, ‘selfidentified mother tongue’ is used as a proxy indicator. The use of this indicator is an attempt to
determine whether there is a correlation between violence against women, where people live, and
their perceived culture.
Data on the prevalence of physical, sexual, and psychological violence against women,
and the associated risk factors is derived from the Demographic Health Survey (DHS) and the
National Statistic Institute of Bolivia. The DHS data is one of the most robust collections of
Heise, Lori. “Violence against Women: An Integrated Ecological Framework.” In Violence against Women 1998
(4):262 ; Centre for Disease Control (CDC). “Sexual violence prevention: beginning the dialogue.” Centre for
Disease Control and Prevention. 2004 http://www.cdc.gov/ncipc/dvp/SVPrevention.pdf. Date of Access: 17 January
2011; Bolt (2008).
5
health-related data (the latest survey was conducted in 2008). The National Statistic Institute of
Bolivia conducted its latest population survey in 2001. The foundational maps that provide the
background to input the data come from ArcGIS. Table 1 provides a detailed description of the
data sources.
Table 1: Data Sources utilized in this study
Data
Source
Website
Violence Rates
DHS
Literacy Rates
DHS
Unemployment
Rates
Income Quintiles
DHS
Indigenous
Languages
INE
http://www.measuredhs.com/countries/country_main.cf
m?ctry_id=2&c=Bolivia
http://www.measuredhs.com/countries/country_main.cf
m?ctry_id=2&c=Bolivia
http://www.measuredhs.com/countries/country_main.cf
m?ctry_id=2&c=Bolivia
http://www.measuredhs.com/countries/country_main.cf
m?ctry_id=2&c=Bolivia
http://www.ine.gob.bo:8082/censo/make_table.jsp?query
=poblacion_15
DHS
Online
Metad
ata
N.A
Source
Scale
Year Presented
1
2008
N.A
1
2008
N.A
1
2008
N.A
1
2008
N.A
1
2001
The first step taken to prepare the data analysis was inputting the data from tabular format
into GIS. Once that data was inputted, a layer was created for each of the risk factors. Then two
foundational layers were created to highlight (1) the percentage of women that had experienced
sexual and physical violence, and (2) the percentage of women that had experienced
physiological violence. Depending on the risk factors, pie or bar graphs were created to illustrate
whether a particular state that had high percentage of violence also had a high risk factor.
Results
There is a general prevalence of violence against women across all provinces. Across the
country, between 20% and 28% of women had experienced sexual or physical abuse in the past
12 months (Map 1). The overall variance between 20% and 28% is not significant.
Map 1: Percentage of Women who have experienced
Sexual or physical assaults in the past 12 months.
Likewise, psychological violence is also prominent throughout the country with 30% to 37% of
women experiencing abuse in the past 12 months (Map 2). While the lack of a significant
variance in either one of these statistics shows that the country needs to work hard to address
violence, it does not indicate how it should go about doing so nor does it indicate how it should
prioritize strategies.
Map 2: Percentage of Women who have experienced
Psychological assaults in the past 12 months.
When correlating geography, prevalence of violence, and literacy rates one gains a greater depth
of understanding into the policy options that the government of Bolivia may have. Throughout
Bolivia, women are, on average, less educated than men. Provinces with the highest prevalence
of violence against women also have the greatest percentage of women that have not completed
primary school. Increasing women’s literacy rates and education standards is a policy option
that may reduce both poverty and the prevalence of violence against women. (Map 3 and 4)
Map 3: Percentage of level of women’s
education by state (2008)
Map 4: Percentage of level of men’s education
by state (2008)
Having an indigenous language as a mother tongue may be the most adequate proxy to
represent culture. Bolivia is a very diverse country in which multiple cultures have interacted for
centuries. There appears to be little correlation between the percentage of people who speak a
certain language and whether there is a high prevalence of violence in that state. For instance, in
the state of Potosi, more than fifty percent of the population speaks an indigenous language as
their mother tongue. However, neighboring province Oruro, which also has a high prevalence of
sexual or physical violence, the majority of people speak Spanish. More detailed studies at the
micro-administrative level need to take place to truly determine if there is a relationship or not.
What is certain is that any policy initiative will have to be communicated in a multi-lingual
format in order to reach the greatest number of households.
Map 5: Percentage of women’s mother
tongue by size of total state population
Map 6: Percentage of men’s mother tongue by
size of total state population
Poverty is another risk factor that Lori Heise identifies in her study as leading to violence
against women. In the Bolivian case, it becomes apparent that provinces with a high prevalence
of sexual and physical violence also have more than 50% of the population living in the lowest
three income quintiles. When one compares Potosi and Santa Cruz, it becomes evident that not
only does Santa Cruz have lower rates of violence against women, but wealthier people also
reside in this province. Unemployment rates for men are generally steady between 22% and
11%. However, unemployment rates for women vary extensively across provinces from 45% to
21%. Hence, another policy option for the government of Bolivia is to tackle both
unemployment and poverty.
Map 7: Income Quintiles by State and by Total
State Population
Map 8: Unemployment Rates by State
Challenges:
While there are a lot of statistics in tabular format, finding GIS related information about
Bolivia was a difficult process. Information is not always disaggregated at the municipal or
county level. This makes it difficult to prescribe policy recommendations and to conduct a more
complex analysis. For instance, by not knowing the number of police stations or their locations,
one is unable to determine whether interventions that are run from those offices are able to reach
victims of violence or not. One is also unable to use ArcGIS to answer questions related to
access to healthcare clinics or schools.
Conclusion
This project enables the visualization of very important data and allows for policy makers
to better understand the gravity of violence against women within each state. Unfortunately, this
project provides little additional insight into why there is such a concerning prevalence of
violence against women. Since the data provided within the DHS survey is at the provincial
level, there is little variance in percentage of women that experience sexual, physical or
psychological violence. More so, when analyzing the risk factors, it is unclear which one is of
more importance. To date, no researcher has aggregated risk factors or prioritized them, hence,
providing any sort of ranking is difficult.
To move forward with this project one would have to plot some of the other risk factors
that Lori Heise argues are of importance. Some of these include: whether communities sanction
violent acts, marital dissatisfaction, and use of alcohol. However, finding robust data on these is
difficult, particularly at the municipality level. Better descriptive data and GIS data, will provide
even more adequate analysis and policy prescriptions. There remains much more work to be
completed in the area of identifying risk factors and analyzing trends in violence against women.
Appendix: A Risk factors for Victims and Perpetrators of
Intimate Partner Violence and Sexual Violence
PERPETRATION BY MEN
VICTIMIZATION OF WOMEN
INDIVIDUAL LEVEL
Demographics
Demographics
Young age
Young age
Low education
Low education
Low socio-economic status/income
Low socio-economic status/income
Unemployment
Pregnancy
Separated/divorce marital status
Exposure to Child Maltreatment
Exposure to Child Maltreatment
Exposure to child maltreatment
Intra-parental violence
Exposure to intra-paternal violence
Sexual abuse
Sexual and/or physical abuse
Mental Disorder
Mental Disorder
Antisocial personality
Depression
Substance use
Substance use
Harmful use of alcohol
Harmful use of alcohol
Illicit drug use
Illicit drug use
RELATIONSHIP LEVEL
Friction over women’s empowerment/ education
disparity
Friction over women’s empowerment/ education
disparity
Multiple partners/infidelity
Large number of children
Low resistance to peer pressure
Male dominance in the family
Relationship quality
Marital dissatisfaction/discord
Gender role disputes
Marital duration
Relationship quality
Marital Dissatisfaction/discord
COMMUNITY LEVEL
Acceptance of traditional gender roles
Acceptance of traditional gender roles
Weak community sanctions
Weak community sanctions
Poverty
Poverty
Neighbourhood characteristics
Neighbourhood characteristics
High proportion of poverty
High proportion of poverty
High proportion of unemployment
High proportion of unemployment
High proportion of male illiteracy
High proportion of female illiteracy
Acceptance of violence
Acceptance of violence
High proportion of households that use corporal
punishment
High proportion of households that use corporal
punishment
Low proportion of women with high level of
autonomy
Low proportion of women with higher education
SOCIETAL LEVEL
Traditional gender norms and societal norms
supportive of violence
Traditional gender norms and societal norms
supportive of violence
Supportive of violence
Supportive of violence
Divorce regulations by government
Lack of legislation on intimate partner violence
within marriage
Protective marriage law
Adapted from the following sources: WHO (2010), Heise (1998) and CDC (2004).
Apppendix B: Annotated Bibliography
A. Coa, Ramiro and luis Ochoa. “Encuesta Nacional de Demografia y Salud ENSA 2008”
October 2009
This is the 485 page report that summarizes the data collected as part of the Demographic
Health Survey (DHS) 2008. It is the most comprehensive data on gender-based violence in
Bolivia. The data in this report can be considered representative and is dissected by province,
age, education, and income level. It becomes very useful to analyze information and construct
map layers.
B. Morales, Rolando and Ana Maria Aguilar, Alvaro Calzadilla. “Undernutrition in Bolivia:
Geography and culture matter” Inter-American Development Bank. April 2008
This paper showcases the interlink between nutrition, health, and geography in Bolivia.
My research study will try to test out the similar dynamic but including gender-based violence
instead of nutrition. This paper also provides some background information on Bolivia’s culture,
language, and religion.
C. Morrow, Betty Hearn. “Identifying and Mapping Community Vulnerabilities.” In Disasters
1999 (23,1), 1-18
This paper does not address the subject matter related to Bolivia specifically. However, it
provides a good framework to thinking about vulnerabilities (which Bolivian women have) and
which factors to consider when conducting my research on Bolivia.
D. Albers, Nathaniel. “The Geography of Domestic Violence: Assessing Reported Domestic
Violence in Missouri.” Master Thesis. University of Missouri. April 2005.
This Master Thesis is very interesting because by focusing on the state of Missouri, the
researcher is able to do some of the work that I would like to do, but the data is so hard to find
for the context of Bolivia. Nathaniel Albers uses GIS to plot incidents of violence by county
within the state of Missouri. He also analyses the known risk factors that make some women
more susceptible to becoming victims of violence. By identifying the different factors, it is easier
to make policy recommendations, while still utilizing visual tools.
E. Elizabeth Groff, David Weisburd, and Nancy Morris. “Where the Action is at Place:
Examining Spatio-Temporal Patterns of Juvenile Crime at Places using Trajectory Analysis and
GIS.” In Putting Crime in its Place 2009.
The chapter in this book is useful because it spends some time analyzing the method by
which the researchers went about to tackle the linkages between youth crime and location. They
argue that a different technique of utilizing GIS leads to different outcomes.
F. Steinberg, Michael K. and Carry Height, Rosemary Mosher, and Mathew Bampton. “Mapping
massacres: GIS and state terror in Guatemala.” In Geoforum, 2006 (37) 22 April 2003
This paper is very useful because it shows how even at the provincial level one can aggregate
information and place it on a map, utilizing GIS to further the analysis. It showed me that while
my research question may not be as ambitious as it could be, there will be value in presenting the
correlation between the different risk factors visually.