ASSESSING THE INTIMATE PARTNER VIOLENCE ...

African women’s IPV

ASSESSING THE INTIMATE PARTNER VIOLENCE ASSOCIATED WITH AFRICAN WOMEN’S EDUCATIONAL LEVELS Page | 1

Submitted by: Georgette Hatcheu E-mail: [email protected] In Partial fulfillment of the Requirements for the Master of Public Health Degree Public Health and Health Services The George Washington University


I. Abstract Objective- This study aimed to examine the association between intimate partner violence (IPV) and African women’s educational levels using the couples recode data from 13 sub-Saharan Page | 2

African countries retrieved from Demographic and Health Survey (DHS) website. Methods-The study utilized data from 35773 married or living together men and woman who both declared to be married (living together) to each other and participated in the DHS survey VI/VII, recode structure VI between 2005 and 2015. The variable “women’s highest educational levels” includes the following categories:

No education, Primary, Secondary, and Higher. The

dichotomous variable “ever experienced any IPV” was constructed based on computation and recoding of the women’s responses to multiple IPV questions including any emotional, physical and sexual violence. A crosstabulation/chi-square with women’s educational levels as predictor and IPV rate as the outcome, was utilized to measure the association between the two variables, and then we ran the scatterplot and Pearson correlation to find any possible negative/positive correlation between the countries’ rate of non-educated women and the overall any IPV rate. Results- We found that the rate of any IPV experience was associated with the women’s highest educational levels; the African women with primary education level were the most affected by any intimate partner violence, and the overall country’s no education rate were negatively correlated with the growing rate of any intimate partner violence experience. Conclusion- This study indicated that women’s highest level of education influences their likelihood to experience any intimate partner violence. As women’s educational levels increase either at the individual or at the country level, the probability of experiencing any intimate partner violence increases.


II. BACKGROUND The United Nation (UN) Assembly regards violence against women as one of the most systematic forms of violations of human rights, rooted in social structures rather than in the Page | 3 individual.1 Violence can affect any women, regardless of age, socioeconomic status, education level, and region of the world, and it occurs in all societies.2 The Resolution 48/104 of the General Assembly of the UN defines violence against women as "any act of violence against the female sex, causing or capable of causing physical, sexual or psychological harm to women, including threats of such acts, coercion or arbitrary deprivation of liberty, whether in public life or private life.1 The forms and prevalence vary by county/region and depend on the environmental conditions.1 The most common forms of violence are family violence and violence by intimate partners, sexual violence (including rape), sexual harassment, and emotional or psychological abuse.3,1 Other forms include: sexual exploitation, sex trafficking, and harmful practices such as female genital mutilation / female genital cutting, forced and early marriage, "honor crimes", femicide, the selection of the fetus based on gender, female infanticide, economic exploitation, political violence, violence against older women and dowry sexual violence.3,1 Per the World Health Organization (WHO), intimate partner violence (IPV) refers to behavior by an intimate partner or ex-partner that causes physical, sexual or psychological harm, including physical aggression, sexual coercion, psychological abuse, and controlling behaviors.4 In 2013, WHO estimated that the lifetime prevalence of physical and/or sexual intimate partner violence among all ever-partnered women was 36.6% in Africa, while the global rate was 30.0%. The African region also shows the highest proportion (45.6%) of women reporting IPV and/or


non-partner sexual violence compare to Americas (36.1%), eastern Mediterranean (36.4%), Europe (27.2%), South-east Asia (40.2%) and western Pacific (27.9%).5 The multiple risk factors and consequences associated to IPV are summarized in Figure1.

Figure 1 is adapted from multiple researchers who have studied IPV in Africa.6,7,8,9,10,11,12,13 This ecological framework tests the causal pathway between individual and family factors (young age, personality disorders, acceptance of violence, employment, income, history of physical aggression of father to mother, having previous partner, alcohol, and drug use, and highest education level); the relationship factors (lack of communication, co-habiting with other family members, male dominance, intergenerational transmission, misbalance decision, conflict in

Page | 4


relationship, no respect inside the couple, have multiple partners); the community characteristics (geographical area, poverty, gender inequitable and social norms, women low socioeconomic status, weak legal sanction of IPV and lack of women civil right) and the likelihood of African Page | 5

women experiencing domestic violence as well as the health consequences.

Although very few studies are devoted to IPV among African women, there are many that examine social abuses encountered in different countries and customs. Uthman and colleagues (2009) analyzed factors associated with attitudes towards IPV in 17 sub-Saharan countries highlighted the wide acceptance of IPV as women’s transgression of gender norms.14 The traditional African patriarchal society that determines the gender power structure2 proceed differently across the region and almost terrorize women in some countries. For example, in Nigeria, polygamous marriages under customary law allows the husband to exercise exclusive sexual rights and obedience from his wives.6 A man who has paid bride price feels that the woman is his property and denies her any voice in the home.15 Even after death, some custom commands that women “drink the water that was used to bathe the corpse of their deceased husband in addition to the shaving of their hair, including pubic hair; sitting and sleeping only on the floor throughout the period of mourning and loudly crying at regular intervals as a mark of respect for the departed.”6 Many IPV prevention programs that are designed to empower women by promoting gender equity consider the gender inequality as women having lower power than man.15 And therefore, women who are more empowered educationally, economically, and socially will be the most protected against IPV.15 The analysis of the Ghana women’s approval of domestic physical violence against wives conducted by Doku and Asante (2015) showed that women’s younger ages (15-35), low education level (no education, primary and junior secondary), spiritual beliefs (Christian, Muslims and traditional) and poverty increased the likelihood to approving the IPV behavior. On the


contrary, women aged 35 and more, and highly educated were less likely to accept IPV.7 Considering that violence is often used as a strategy in conflict,7 relationships full of conflict, and especially those in which conflicts occur about women's gender role transgressions are more Page | 6

violent than peaceful relationships11 Thus, conflict rather than equality may be the result of a woman’s greater independence and may lead to an increase in the rate of separation or divorce and violence.11 The theoretical framework used to inform this research is the ecological model. This approach focuses on both population-level and individual-level determinants of health and interventions.11 The model distinguishes five different levels (intrapersonal, interpersonal, institutional, community, socio-cultural/policy) of superposing causality interacting on one another to either protect or expose to violence.16 Studying every cause in its specific socio-cultural environment may increase the effectiveness of preventive interventions.16 The Center for Disease Control uses four components of the social-ecological model to appreciate the violence and built preventive programs.2 The individual level of the model identifies biological and personal history factors that increase the likelihood of becoming a victim or perpetrator of violence (age, education, income, substance use, or history of abuse) and proposes education and life skills training to promote attitudes, beliefs and behaviors.2 The relation level examines close relationships that may increase the risk of experiencing violence as a victim or perpetrator (social circle-peers, partners, and family members) and suggests family-focused prevention programs, and mentoring and peer programs designed to reduce conflict, foster problem-solving skills, and promote healthy relationships.2 The community level explores the characteristics associated with violence in different settings (schools, workplaces, and neighborhoods). Prevention strategies at this level are typically designed to impact the social and physical environment.2 The fourth level


examines the factors that help create a climate in which violence is encouraged or inhibited. These factors include social and cultural norms that support violence as an acceptable way to resolve conflicts and policies that promote economic and social inequalities.2 Page | 7

III. SPECIFIC AIMS AND OBJECTIVES` This study aims to explore whether greater empowerment through education decreases IPV among sub-Saharan African women. The objectives of the study are the following: Objective1: To analyze the association between education level and any experience with intimate partner violence among sub-Saharan African women. Hypothesis 1: African women’s highest level of education is negatively/positively associated with any intimate partner violence experience. Descriptive statistic Independent variable: Education levels (no education, primary, secondary, higher) Dependents variables: intimate partner violence experience (dichotomous variable yes/no) Test: cross-tabulation and Chi-square Objective 2: To analyze any correlation between the countries women’s no education rate and their IPV mean rate. Hypothesis 2: countries’ rate of non-educated women appears to be a significant IPV predictor Statistics test: bivariate correlation Independent variables: women’s No education rate per country Dependent variable: Women’s IPV mean rate per country


This research contributes to the understanding of the impact of women’s highest educational levels on IPV and ultimately, be used by IPV prevention providers to increase the effectiveness and impact of IPV programs implemented in Sub-Saharan Africa. Page | 8

IV. METHODS The data used in this study are retrieved from Demographic and Health surveys (DHS) website. 17 Funded by United states Agency for International development (USAID) with contributions from other donors such as United Nations International Children's Emergency Fund (UNICEF), United Nations Population Fund (UNFPA), World Health Organization (WHO), Joint United Nations Program on HIV/AIDS (UNAIDS), the DHS program is responsible for collecting and disseminating accurate, nationally representative data for monitoring and impact evaluation indicators in the areas of population, health, and nutrition in developing countries.17 This project is implemented by ICF (Inner City Fund) International. ICF conducts vital health-focused surveys using countries national institutions and collects data that help the U.S. and global health agencies identify and monitor health risks and deliver effective interventions18. The sampling design is a multistage cluster sampling in different regions of the countries including women aged 15-49 and men aged 15-59 selected and interviewed separately. As the IPV data had been collected more than once in many African countries since 2005, we used the most recent data collected between 2005 and 2015 for each country included in the study.17 The original countries couples’ sample size range from 1148 to 8658. The 13 countries selected for this study include Burkina Faso, Mali, ivory coast, Gambia, Togo, Nigeria, Gabon, Chad, Kenya, Tanzania, Malawi, Zambia and Zimbabwe. These countries represent diverse sub-Saharan African geographical region (East, West, South and Central Africa), religion and culture.


The study used the DHS standardized questionnaire administered in the target countries. The survey Questionnaire VI or VII with the same recode structure VI was selected to allowed data comparison across the countries. The data file used is the couple’s file/couple re-coded (CR).17 “The dataset has one record for every couple. It contains data for married or living together men and woman who both declared to be married (living together) to each other and with completed individual interviews (questionnaires). The file is the result of linking the two files previously described based on whom they both declared as partners”.17 The variables selected include the women education levels, the experience of any emotional violence, the experience of any less severe physical violence, the experience of any severe physical violence, the experience of any sexual violence. The determinant variable which is the women’s highest educational level attended is a standardized variable providing a level of education in the following categories: No education, Primary, Secondary, and Higher. As stated by DHS data notes, “In some countries, the educational system does not fit naturally into this scheme. In that case, the variable is constructed as accurately as possible from the country's own scheme and the variable used for the Final Report is included as a country-specific variable”17 Outcome variables: To assess the intimate partner violence, subjects are classified as having experienced any IPV if they answered yes on at least one of the IPV-related questions (either emotional, physical or sexual violence). Inclusion criteria for the subset data studied: all respondents whose highest educational level are displayed on the dataset and who answered the survey questions (Yes /No respectively coded 1/0) associated with the above IPV-related variables17. Women’s ever experienced any emotional violence (V104) include Spouse ever humiliated her (D103A), Spouse ever threatened her with harm (D103B), Spouse ever insulted her

Page | 9


or made her feel bad (D103C, Country specific emotional abuse (D103D/E/F). Physical violence is divided into less severe and severe violence. Ever experienced any less severe violence includes (D106) include Spouse ever pushed shook or threw something (D105A), Spouse ever slapped Page | 10

(D105B), Spouse ever punched with a fist or something harmful (D105C), Spouse ever twisted her arm or pulled her hair (D105J). Ever experienced any severe violence comprise the following variables: Spouse ever kicked or dragged (D105D), Spouse ever tried to strangle or burn (D105E), Spouse ever threatened with knife/gun or another weapon (D105F) and Spouse ever attacked with knife/gun or another weapon (D105G). And the sexual violence (D108) includes Spouse ever physically forced sex when not wanted (D105H) and Spouse ever forced other sexual acts when not wanted (D105I). The data was computed and re-coded to obtain the final intimate partner violence variable used for analysis. For this purpose, we first compute ever experienced any less physical violence and severe violence to obtain the variable “ever experienced any physical violence”, further re-coded onto dichotomous variable (yes/no ever experienced any physical violence). Secondly, emotional, physical and sexual violence were computed IPVc and re-coded onto “ever experienced any intimate partner violence” (IPV).

IV. ANALYSIS PLAN Descriptive analysis was conducted to explore the association of women’s education levels with their intimate partner violence experience (emotional, physical and sexual). After examining the distribution of each of the variables, we looked for relationships among the two variables. A contingency table analysis from crosstabulation displayed a distribution of cases by their values (percentage) on the different variables. To test hypothesis 1, the joint frequency distribution was


analyzed with the chi-square statistic to determine whether the variables were statistically independent or if they were associated. This chi-square analysis was performed with women highest educational levels (No education, primary, secondary and higher) as independent variables Page | 11

and the intimate partner violence as the dependent variable. To test hypothesis 2, we ran the scatterplot and Pearson correlation using the countries women’s no education rate as predictor and the country IPV rate as outcome

V. RESULTS 1-Frequencies table Couple’ s recorde sample size

study Samp le size

5088

2

Burkina Faso Chad

3

Country

Highest educational level No educa tion (%)

Prima ry (%)

Seco ndar y (%)

High er (%)

3721

79.9

13.6

6.1

3050

2337

73.2

18.6

Gabon

1946

1502

7.4

4

Gambia

1388

758

5

Malawi

3764

6

Mali

7

Ivory cost Kenya

1

8

Intimate partner violence Experi enced any less severe violen ce (%)

Experi enced any severe violen ce (%)

.4

Exper ience d any emoti onal viole nce (%) 11.8

12.3

.7

7.7

.6

19.3

22.3

6.5

9.0

29.9

38.4

50.9

3.3

34.5

48.5

17

13.5

56.4

61.6

14.1

21

3.3

16.8

20.3

12.4

2.8

30.7

3357

17

69.7

12.7

6

23.6

19

8

15.3

35.4

3039

2059

81.1

9.7

8.5

.8

33.9

28.7

5.4

14.3

46.2

2129

1569

69.9

21.7

7.5

1.0

18.2

24.3

5.3

31.2

5265

2269

12.5

58.3

22.9

6.4

27.4

32

10.4

42.6

5.5 13.8

Exper IPV ience rate d any (%) sexua l viole nce (%) 1.4 17.7


9

Nigeria

8658

6937

42.9

20.0

29.1

8.0

18.4

13.6

5.1

5.0

24.0

10

1148

976

22.4

65

12.2

.4

28.6

29.2

9.7

12.1

38.8

11

Tanzani a Togo

2270

1747

42.1

37.0

20.1

.7

31.1

22.4

6.1

7.6

38.4

12

Zambia

7198

6064

11.3

58.0

26.8

4.0

23.2

36.4

11.7

16.1

46.3

13

Zimbab we Total

3005

2477

2.6

34.2

60.5

2.7

24.4

27.9

8.2

15.1

42.2

47948

35773

40.2

35.25

22

11.4

23.93

25.91

9.38

9.83

36.9 0

Page | 12

Means

Table 1 shows the countries, the sample size of the survey selected couples, the subset sample size of the survey’s participant who answered the intimate partner violence questionnaire, the distribution of the study participants across the countries per their highest educational levels and their various intimate partner violence components. Out of the 54 countries in Africa, only 23 IPV-related data are available at DHS website to date, giving a participation rate of 42.59%. Among the 23 countries, 13 were selected for this study, and a total of 35773 couples responded to IPV-related questionnaires. The mean rate of women highest educational levels distributed as follow: 40.2% no education, 35.25% primary education, 22% secondary education and 11.4% higher education (table1 and Fig 1a). Nigeria and Zambia have the highest sample sizes (6937 and 6064 respectively) while Gambia has the lowest sample (758). The rate of no education respondents varies across the countries. Mali (81.1%), Burkina Faso (79.9%) and Chad (73.2%) have the top three highest rate while Zimbabwe (2.6%), Gabon (11.3%) and Kenya (12.5) have the most educated population. The mean rate of any IPV experience across the 13 countries is 36.90%. (Table1). Gabon has the highest rate (56.4%), followed by Zambia (46.3%), Mali


(46.2%), Kenya (42.6%) and Zimbabwe (42.2%) while Burkina Faso (17%) and Nigeria (24%) and Chad (29.9%) are far below the mean rate. Table 2-Distribution of IPV by women’s highest educational levels (Chi-square statistics)

Country

IPV per education levels No Primary secondary higher education 17.3 20.2 19.0 00

Statistics

Pvalue

Likelihood Ratio =8.496

.037

1

Burkina Faso

2

chad

26.5

41.05

34.01

30.8

Likelihood Ratio=37.382

.000

3

Gabon

41.4

59.8

56.7

44.9

Pearson ChiSquare

.001

4

Gambia

31.5

29.9

28.3

36

Pearson ChiSquare.923a

.820

5

Malawi

32.6

36.9

31.4

28.6

6

Mali

46.8

49.2

37.9

37.5

Pearson ChiSquare=6.279a

.099

7

Ivory cost

29.8

33.8

37.3

20.0

Likelihood Ratio=4.957

.175

8

Kenya

29.3

47.4

41.6

28.3


.000

9

Nigeria

16.5

32.7

30.3

18.8


.000

10

Tanzania

33.8

44.3

20.2

00


.000

11

Togo

42.3

37.2

33.2

15.4


.006


.057

Page | 13


12

Zambia

13 Zimbabwe Means

45

48.9

43.5

46.3


.000

49.2

49.5

39.5

29.9

ChiSquare=19.529a

.000

34

40.83

34.83

25.88

Table 2 indicates the distribution of the IPV by women’s educational levels, the chi-square statistics, and the statistical significance levels. All the four variants of women educational levels have different impacts on the rate of any IPV experienced by the participants. In 10 of the 13 countries, primary education is the commonest highest educational level with the highest rate of any IPV (Fig 1b and 1c). The chi-square either Pearson or likelihood ratio P-values are statistically significant at .05 in 8 out of the 13 countries. The mean rate associated with the primary educational level across the 13 countries is equal to 40.83% compare to 34.83% for the secondary education, 34% for no education and 25.88% for the higher education. Considering the most affected educational level (primary), the five countries at the forefront include Gabon (59.8%), Zimbabwe (49.5%), Mali (49.2), Zambia (48.9%) and Kenya (47.4%) (table 2). These countries are followed by Tanzania, Chad, Togo, Malawi, Ivory coast and Nigeria with means rate oscillating between 32.7 and 44.3%. In Gambia and Burkina Faso, the primary educated have the lowest IPV rate with respectively 29.9% and 20.2% (Table2). 3- Distribution of the countries’ rate of any IPV experience compare to the countries’ rate of non-educated women

Page | 14


Page | 15

The scatter dot shows a possible negative correlation between the countries’ rate of non-educated women and the overall any IPV experience rate, as the two variables are moving in the opposite direction (see also Fig 1d). Correlations

No education (%) No education (%)

Pearson Correlation

IPV rate (%) 1

Sig. (2-tailed) IPV rate (%)

Pearson Correlation Sig. (2-tailed)

-.600* .030

-.600* .030

1


*. Correlation

is significant at the 0.05 level (2-tailed).

The Pearson correlation equal to -.600 indicates a strong negative correlation between the countries’ rate of non-educated women and country mean rate of intimate partner violence. This result is statistically significant at .05 with the P-value equal to .03.

VI. DISCUSSION Our findings in this study of 13 Sub-Saharan African countries indicated that women’s highest educational levels were associated with any experience of intimate partner violence. The relationship that emerged from table 2 and figure 2 shows that the primary educational level had the highest risk of any IPV. This result means that although increasing women’s educational level (secondary and higher education) seems to positively impact the rate of women’s IPV experience, and therefore corroborate the protective effect on women’s IPV experience19,20,15,30, having primary educational level may have a negative influence. On the contrary, the Fig 1c shows that countries with low rates of no education present higher IPV rates. This inverse relation between the countries no education rate and IPV experience opposed the protective effect when it comes to the whole community. By increasing knowledge, women’s education increases their ability to manage the household and their financial power, increases the ability to resist gender norms that undermine them, and lead to gender role transgression, which is one of the major sources of conflict and IPV risk factors19,14,11,10. The supposed positive effect observed with secondary and higher education could be explained by the fact that their increasing power and abilities could allow them to either leave the perpetrator (divorce or living separate) or serve as an incentive to prevent the husband/partner from abusing them.22,21 By perceiving women’s higher educational levels as the driver of socioeconomic position, increasing women’s educational levels can be a threat for the perpetrators who are losing control over their partners,22 and therefore increases their

Page | 16


likelihood of perpetrating violence. Exploring the association between women’s low level of education and domestic violence perpetrated by their husbands and in-laws in rural Bangladesh, researchers came up with a mixed position stating that higher level of education may increase women’s autonomy and therefore lower their level of any violence experience, although in some cases, it may indirectly intensify family violence.24 The lower IPV rate associated with no education compare to primary education can be related to the fact that non-educated women are more likely to comply with African gender norms that promote acceptance of violence against women and girls and submission to their husband.14 The study of factors associated with women acceptance of IPV conducted in 17 African countries by Uthman et al provides evidence that wealth status and education attainment are significantly associated with attitudes towards IPVAW.23 Accordingly, access to newspaper reduced the likelihood of having tolerant attitudes towards IPVAW.23 And furthermore, IPVAW was widely accepted by less-educated and poorest women,23 In Tanzania, the study of a random sample size of 5069 women indicated that having some autonomy lower women’s likelihood of tolerating IPV. Therefore, less-educated women were at higher likelihood of tolerating IPV when contrasted with peers having a postsecondary education.25 Other researchers believe that communication skills are sharpened with added levels of education and therefore, lower levels of education may be associated with violence due to the inability to properly communicate anger or frustration.27 In the United States, many research studies concerning risk factors have indicated that lack of education defined as having earned less than a high school diploma is one of the general risk factors for domestic violence.26 Compare to our study, high school diploma is earned at the end of the secondary school level in African education systems. So, earning less than high school diploma is equivalent to having a primary educational level. This result corroborated the fact that women with

Page | 17


primary educational level are more likely to experience a higher rate of any IPV than women who achieved secondary or higher levels of education. The increasing level of IPV associated with the countries decreasing no education rate does not Page | 18 agree with the previous research studies that link the women no education with an increased likelihood of experiencing violence by her partner(s).28

This study shows that incriminating

women's less education as a driver of IPV in Africa may be related to the very high rate of less educated women in Africa compared to the rate of higher educated women. this assumption is supported by the increasing rate of IPV observed in countries like Gabon and Zimbabwe where the rates of secondary and higher educated women are higher compared to those of the other studied countries. So, compare to non-educated women, educated women should increases either their coping strategies or their tactics to face their increasing rate of IPVs. Seen in this pathway, increasing women’s educational level can be IPV-protective.30 LIMITATIONS compare to the general couple’s survey sample size with 47948 interviewed, 35773 answered questions related to IPV (74.60%). This under-reporting may be related to social norms or shame, as talking about sexual behavior in Africa is a taboo and women can be stigmatized.29 This study did not consider all the contextual factors influencing the perpetrator behavior. conducting the adjustment by other factors including husband’s education, spousal education differential or employment may lead to a different impact. Finally, this paper only examined whether women experienced any IPV or not. Hence, future research needs to explore the association between women's education and frequency of IPV.

VIII. CONCLUSIONS


This study indicated that women’s highest educational levels influence their likelihood to experience any intimate partner violence. As women’s educational levels increase either at the individual or at the country level, their level of knowledge, economic power, and independence Page | 19

increases, they become more likely to transgress gender norms and get into conflict relationship with their husbands/partners, therefore increasing their probability of experiencing any intimate partner violence and leaving the perpetrators. It would be important to deepen this research by considering the frequencies contextual factors and mediating factors like husband/partner educational levels, Husband/partners educational differences and assess their contribution to this issues.

Annex 1: List of figures

educational levels percentage (%)

Fig1a:Women's highest educational levels rate per countries 90 80 70 60 50 40 30 20 10 0

Target Sub-Saharan African countries

NO education

primary

secondary

Higher


Fig1b: IPV rate per women's educational levels 70 60

IPv percentage (%)

Page | 20 50 40 30 20 10 0

Target Sub-Saharan African countries IPV No education

45

IPV Primary

IPV secondary

IPV higher

Fig 1C : Comparison of IPV rate per women's educational levels (overall means of the 13 studied countries)

IPV means rate (%)

40 35 30 25 20 15 10 5 0 No education

Primary

secondary

Educational levels

higher


Fig 1d: Comparison between Countries' no education rate and their IPV rate 90 80

Page | 21

Percentage (%)

70 60 50 40 30 20 10 0

Target Sub-SAharan African countries No education (%)

IPV rate (%)

Annex 2: Target Countries DHS survey summary Country

Year

type

Phase

Burkina Faso

May 2010 January 2011

Standard DHS-VI DHS

Chad

October 2014 - Standard DHS-VII DHS-VI April 2015 DHS

3050

Cote d’ivoire

December 2011 - May 2012

2129

Standard DHS-VI DHS

Recoded structure DHS-VI

DHS-VI

CR sample size 5088

Implementing Organization The Institut National de la Statistique et de la Démographie (INSD) Institut National de la Statistique, des Études Économique et Démographiques (INSEED) Ministère de la Santé et de la Lutte contre le Sida (MSLS),


Gabon

January 2012May 2012

Standard DHS-VI DHS

DHS-VI

1946

Gambia

February 2013 - April 2013

Standard DHS-VI DHS

DHS-VI

1388

Kenya

May 2014 October 2014

Standard DHS-VII DHS-VI DHS

5265

Mali

November 2012 February 2013

Standard DHS-VI DHS

DHS-VI

3039

Malawi

June 2010 November 2010 February 2013 - June 2013

Standard DHS-VI DHS

DHS-VI

3764

Standard DHS-VI DHS

DHS-VI

8658

December 2009 - May 2010 November 2013 - April 2014

Standard DHS-VI DHS

DHS-V

1148

Standard DHS-VI DHS

DHS-VI

2270

Nigeria

Tanzania

Togo

Institut National de la Statistique(INS) Direction Générale des Statistiques (DGS) The Gambia Bureau of Statistics (GBOS) Kenya National Bureau of Statistics (KNBS) of the Ministry of Planning and National Development INFO-STAT in collaboration with government agencies, the National Institute of Statistics (INSTAT) and the Cellule de Planification et de Statistique (CPS) National Statistics Office National Population Commission (NPC) National Bureau of Statistics General Directorate of Statistics and National

Page | 22


Zambia

August 2013 April 2014

Standard DHS-VI DHS

DHS-VI

7198

Zimbabwe September 2010 - March 2011

Standard DHS-VI DHS

DHS-VI

3005

Accounting (DGSCN) The Central Statistics Office (CSO) Central Statistical Office

Page | 23

Annex 3: Countries’ contingency table

1-Burkina Faso Women's highest educational levels

No education Primary secondary higher Total

Intimate Partner Violence (%/count) no Yes

total

82.7 (2460) 79.8 (404) 81.0 (183) 100.0 (15) 82.3 (3062)

100.0 (2974) 100.0 (506) 100.0 (226) 100.0 (15) 100.0 (3721)

17.3 (514) 20.2 (102) 19.0 (43) 0.0 (0) 17.7 (659)

Likelihood ratio 8.496 P-value .037 2-Gabon Women's highest educational levels


Intimate Partner Violence(%/count) no Yes

total

58.6 (65) 40.2 (232) 43.3 (331) 55.1 (27) 43.6 (655)

100.0 (111) 100.0 (577) 100.0 (765) 100.0 (49) 100.0 (1502)

41.4 (46) 59.8 (345) 56.7(434) 44.9 (22) 56.4 (847)

Pearson chi-square P-value .001 3-Gambia Women's highest educational levels

No education Primary secondary higher

Intimate Partner Violence(%/count) no Yes

total

68.5 (320) 70.1 (75) 71.7 (114) 64.0 (16)

100.0 (467) 100.0 (107) 100.0 (159) 100.0 (25)

31.5 (147) 29.9 (32) 28.3 (45) 36.0 (9)


Total

69.3 (525)

30.7 (233)

100.0 (758)

Pearson chi-square .923a P-value .820 Page | 24

4-Malawi Women's highest educational levels


Intimate Partner Violence(%/count) no yes

total

67.4 (384) 63.0 (1477) 68.6 (293) 8.7 (15) 64.6 (2169)

100.0 (570) 100.0 (2339) 100.0 (427) 100.0 (21) 100.0 (3357)

32.6 (186) 36.9(862) 31.4(124) 4.0 (6) 35.4 (1188)

Pearson chi-square 7.515a P-value .057 5-Mali Women's highest educational levels



total

53.2 (888) 50.8 (101) 62.1 (108) 62.5 (10) 53.8 (1107)

100.0 (1670) 100.0 (199) 100.0 (174) 100.0 (16) 100.0 (2089)

46.8 (782) 49.2 (98) 37.9 (66) 37.5 (6) 46.2 (952)

Pearson chi-square 6.279a P-value.099 6-Nigeria Women's highest educational levels



total

83.5 (2483) 67.3 (935) 69.7 (1407) 81.2 (450) 76.0 (5275)

100.0 (2973) 100.0 (1390) 100.0 (2020) 100.0 (554) 100.0 (6937)

16.5 (490) 32.7 (455) 30.3 (613) 18.8 (104) 24.0 (1662)

Pearson chi-square 203.403 P-value .000 7-Ivory coast Women's highest educational levels

No education


total

70.2 (769)

100.0 (1096)

29.8 (327)


Primary secondary higher Total

66.2 (225) 62.7 (74) 80.0 (12) 68.8 (1080)

33.8 (115) 37.3 (44) 20.0 (3) 31.2 (489)

100.0 (340) 100.0 (118) 100.0 (15) 100.0 (1569)

Likelihood ratio 4.957 P-value .175

Page | 25

8-Chad Women's highest educational levels



total

73.5 (1258) 58.5 (254) 65.9 118) 69.2 (9) 70.1 (1639)

100.0 (1711) 100.0 (434) 100.0 (179) 100.0 (13) 100.0 (2337)

26.5 (453) 41.5 (180) 34.1 (61) 30.8 (4) 29.9 (698)

Likelihood ratio 37.382 P-value .000 9-Tanzania Women's highest educational levels



total

66.2 (145) 55.7 (353) 79.8 (95) 100.0 (4) 61.2 (597)

100.0 (2019) 100.0 (634) 100.0 (119) 100.0 (4) 100.0 (976)

33.8 (74) 44.3 (281) 20.2 (24) 00.0 (0) 38.8 (379)

Likelihood ratio 33.394 P-value .000 10-Kenya Women's highest educational levels



total

70.7 (200) 52.6 (695) 58.4 (303) 71.7 (104) 57.4 (1302)

100.0 (283) 100.0 (1322) 100.0 (519) 100.0 (145) 100.0 (2269)

29.3 (83) 47.4 (627) 41.6 (216) 28.3 (41) 42.6 (967)

Pearson chi-square 45.353a P-value .000 11-Togo


total


Women's highest educational levels


57.7 (424) 62.8 (406) 66.8 (235) 84.6 (11) 61.6 (1076)

42.3 (311) 37.2 (241) 33.2 (117) 15.4 (2) 38.4 (671)

100.0 (735) 100.0 (647) 100.0 (352) 100.0 (13) Page | 26 100.0 (1747)

Likelihood ratio 12.433 P-value .006 12-Zambia Women's highest educational levels



total

55.0 (376) 51.1 (1798) 56.5 (918) 68.3 (164) 53.7 (3256)

100.0 (684) 100.0 (3516) 100.0 (1624) 100.0 (240) 100.0 (6064)

45.0 (308) 48.9 (1718) 43.5 (706) 31.7 (76) 46.3 (2808)

Pearson chi-square35.619a P-value .000 13-Zimbabwe Women's highest educational levels



total

50.8 (33) 52.5 (445) 60.5 (906) 70.1 (47) 57.8 (1431)

100.0 (65) 100.0 (847) 100.0 (1498) 100.0 (67) 100.0 (2477)

49.2 (32) 47.5 (402) 39.5 (592) 29.9 (20) 42.2 (1046)

Pearson chi-square 19.529a P-value .000

References 1. United Nations. General assembly. 48/104. Declaration on the Elimination of Violence against Women A/RES/48/104 85th plenary meeting 20 December 1993 2. Center for Disease Control (CDC). http://www.cdc.gov/violenceprevention/overview/socialecologicalmodel.html assessed on 10/27/2016 3. Dahlberg LL, Krug EG. Violence-a global public health problem. In: Krug E, Dahlberg LL,

Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1–56. 4. World Health Organization -Pan American health Organization. Understanding and


addressing the violence against women. Intimate partner violence. 2012 who.int/iris/bitstream/10665/77432/1/WHO_RHR_12.36_eng.pdf 5. World Health Organization. Global and regional estimates of violence against women:

prevalence and health effects of intimate partner violence and non-partner sexual violence. 2013. who.int/iris/bitstream/10665/85239/1/9789241564625_eng.pdf Page | 27 6. McCloskey LA, Williams C, Larsen U. Gender inequality and intimate partner violence among women in Moshi, Tanzania.Int Fam Plann Perspect. 2005;31(3):124-130. 7. Doku DT, Asante KO. Women's approval of domestic physical violence against wives: Analysis of the Ghana demographic and health survey. BMC Women's Health. 2015;15(1). 8. Ellsberg M, Arango DJ, Morton M, et al. Prevention of violence against women and girls: What does the evidence say? Lancet. 2015;385(9977):1555-1566. 9. Ellsberg M, and Heise L. Researching Violence Against Women: A Practical Guide for Researchers and Activists. Washington DC, United States: World Health Organization, PATH; 2005. 10. Garcia-Moreno C et al. WHO multi-country study on women’s health and domestic violence against women: initial results on prevalence, health outcomes, and women’s responses. Geneva: World Health Organization, 2005 11. Jewkes R. Intimate partner violence: causes and prevention. Lancet (London, England) [serial online]. April 20, 2002;359(9315):1423-1429. Available from: MEDLINE Complete, Ipswich, MA. Accessed October 27, 2016 12. Menjivar Cecilia and Salcido Olivia. Immigrant women and domestic violence common experiences in different countries. Gender & Society, Sociologists for Women in Society, Vol. 16 No. 6, December 2002 898-920 DOI: 10.1177/089124302237894? 13. Uthman OA, Lawoko S, Moradi T. Factors associated with attitudes towards intimate partner violence against women: A comparative analysis of 17 sub-Saharan countries. BMC Int Health Hum Rights. 2009;9(1). 14. Uthman OA, Lawoko S, Moradi T. Factors associated with attitudes towards intimate partner violence against women: A comparative analysis of 17 sub-Saharan countries. BMC Int Health Hum Rights. 2009;9(1). 15. Chika IS. Gender-based domestic violence in Nigeria: A socio-legal perspective. Indian J Gender Stud. 2012;19(1):137-148.

16. McLeroy K, Bibeau D, Steckler A, Glanz K. An Ecological Perspective on Health Promotion


Programs. [serial online]. 1988 p 355; Available from Minority Health Archive, Ipswich, MA. Accessed October 27, 2016.

17. Inner City Fung international (ICF). Health survey and data collection https://www.icf.com/markets/government/health/surveys-and-data-collection assessed on Page | 28 11/4/2016 18. The Demographic and Health Surveys Program http://dhsprogram.com/Who-We-Are/About-Us.cfm assessed 11/4/2016 19. Krishnan S. Do structural inequalities contribute to marital violence? Ethnographic evidence from rural South India. Violence Against Women. 2005; 11:759–775. 20. Panda P, Agarwal B. Marital violence, human development and women’s property status in India. World Dev. 2005; 33:823–850. 21. Martin SL, Moracco KE, Garro J, et al. Domestic violence across generations: findings from northern India. Int J Epidemiol. 2002; 31:560–572. 23- Uthman, O. A., Lawoko, S., & Moradi, T. (2009). Factors associated with attitudes towards

intimate partner violence against women: A comparative analysis of 17 sub-Saharan countries. BMC International Health and Human Rights, 9(1), 14. doi:10.1186/1472698X-9-14 24. Marium S. Women’s Level of Education and Its Effect on Domestic Violence in Rural Bangladesh. Development Studies, Brac Development Institute, Brac University, Bangladesh. IOSR Journal of Humanities And Social Science (IOSR-JHSS) Volume 19, Issue 5, Ver. III (May. 2014), PP 40-45 e-ISSN: 2279-0837, p-ISSN: 2279-0845 25. Kapiga S, Harvey S, Watts C, et al. Prevalence of intimate partner violence and abuse and associated factors among women enrolled into a cluster randomized trial in northwestern Tanzania. BMC Public Health [serial online]. February 14, 2017;17:1-11. Available from: Academic Search Complete, Ipswich, MA. Accessed February 25, 2017 26. http://criminal-justice.iresearchnet.com/crime/domestic-violence/education/ 27. Peedicayil, A., Sadowski, L. S., Jeyaseelan, L., Shankar, V., Jain, D., Suresh, S., Bangdiwala, S. I. and the IndiaSAFE Group (2004), Spousal physical violence against women during pregnancy. BJOG: An International Journal of Obstetrics & Gynaecology, 111: 682–687. doi:10.1111/j.1471-0528.2004. 00151.x


28. World Health Organization. Pan American health organization understanding and addressing violence against women. Intimate partner violence. http://apps.who.int/iris/bitstream/10665/77432/1/WHO_RHR_12.36_eng.pdf 29. Lawoko S. Factors associated with attitudes toward intimate partner violence: a study of women in Zambia. Violence and Victims [serial online]. October 2006;21(5):645-656. Available from: MEDLINE Complete, Ipswich, MA. Accessed February 25, 2017 30. Ackerson LK, Kawachi I, Barbeau EM, Subramanian SV. Effects of Individual and Proximate Educational Context on Intimate Partner Violence: A Population-Based Study of Women in India. American Journal of Public Health. 2008;98(3):507-514. doi:10.2105/AJPH.2007.113738.

Page | 29