A BASELINE RAPID ASSESSEMENT SURVEY REPORT ON THE DOMESTIC VIOLENCE ACT 2010 CONDUCTED IN KAMPALA, KAMULI AND IGANGA

Conducted By: LWANYAAGA JULIUS NANNYUNJA BARBRA

Page | i

ACKNOWLEDGEMENT The Domestic Violence Act 2010 Baseline survey was commissioned by CEDOVIP with support from UKAID_. The consultants are most grateful to both originations for the support and facilitation provided, that enabled the successful implementation of the survey. The consultants are so grateful to the staff, administration and management of CEDOVIP for the support throughout the process. More specifically, we are indebted to Tina Musuya and Diana Kagere Mugerwa who reviewed the report. Arising from the above reviews were some invaluable comments and input that led to the finalization of the report. Many thanks go to the District Community Development Officers in all districts where the study was conducted for providing invaluable information about the study communities and introducing the study team to relevant offices. Your contribution to the success of this study cannot be forgotten. The study participants were drawn from the three districts where the study was conducted. The consultant is highly indebted to all the study participants. Without their participation, it would not have been possible to successfully conduct the survey. The study also involved key informants interviews and group discussions drawn from the participating districts. The Consultants are most grateful for the input of all the key informants.

Page | i

Table of Contents
Acknowledgement ........................................................................................................................................................................................ i Table of Contents .......................................................................................................................................................................................... ii List of Figures................................................................................................................................................................................................... iii List of Tables .................................................................................................................................................................................................... iv 1.0 INTRODUCTION.......................................................................................................................................................................... 1 1.1 DOMESTIC VIOLENCE AND VIOLENCE AGAINST WOMEN .................................................................................................1 1.2 DOMESTIC VIOLENCE AND VIOLENCE AGAINST WOMAN IN UGANDA ........................................................................1 1.3 THE DOMESTIC VIOLENCE ACT 2010 ........................................................................................................................................3 1.4 THE DOMESTIC VIOLENCE ACT 2010 RAPID ASSESSMENT SURVEY ....................................................................................3 2.0 STUDY METHODOLOGY...........................................................................................................................................................................4 2.1 STUDY APPROACH, DESIGN AND DATA COLLECTION METHODS ....................................................................................4 2.2 KEY INFORMANT INTERVIEWS ......................................................................................................................................................4 2.3 FOCUS GROUP DISCUSSIONS (FGDS) ......................................................................................................................................4 2.3 STRUCTURED QUESTIONNAIRE INTERVIEWS .............................................................................................................................5 2.5 SAMPLE SIZE AND THE SELECTION OF STRUCTURED QUESTIONNAIRE RESPONDENTS ..................................................5 2.5.1 The Police ...................................................................................................................................................................... 5 2.5.2 Health Workers ............................................................................................................................................................. 6 2.5.3 Local Council Leaders ................................................................................................................................................ 6 2.5.4 Community Activists .................................................................................................................................................... 6 2.6 ELIGIBILITY CRITERIA......................................................................................................................................................... 6 3.0 PRESENTATION OF RESULTS................................................................................................................................................... 7 3.1 THE DVACT 2010 BASELINE RAPID ASSESSMENT SURVEY COMMUNITIES AND RESPONSE RATES ...........................7 3.1.0 DEMOGRAPHIC CHARACTERISTICS OF SURVEY RESPONDENTS ....................................................................................7 3.1.1 ATTITUDES TOWARDS PHYSICAL VIOLENCE ............................................................................................................... 9 3.1.3 ATTITUDES TOWARDS SEXUAL VIOLENCE ................................................................................................................. 10 3.1.3 POWER AND CONTROLLING BEHAVIORS/ EMOTIONAL VIOLENCE .................................................................... 11 3.2 THE DOMESTIC VIOLENCE LAW.............................................................................................................................................. 12 3.2.1RECOMMENDED ACTIONS UNDER THE DVAct 2010 ............................................................................................... 12 3.3 AWARENESS ON THE BENEFITS OF THE DVACT 2010 ..........................................................................................................14 3.3.0 Beneficiaries of the DVA 2010 ..............................................................................................................................................14 3.3.1 Specific benefits of the DV Act 2010 .................................................................................................................................15 3.5 AWARENESS ON THE ROLES AND JURISDICTIONS OF RELEVANT INSTITUTIONS AND DUTY BEARERS .....................16 3.5.1 Duties of Police Officers .........................................................................................................................................................16 3.5.2 Jurisdictions of the Local Council Court ................................................................................................................. 17 3.5.3 Duties of the Health care Workers (Practitioners) and the courts of law in the DVA 2010 ............................. 17 4.0 QUALITATIVE RESULTS AND DISCUSSION OF STUDY FINDINGS ........................................................................................... 19 Knowledge, attitudes and perception of VAW........................................................................................................................19 Awareness around the existence of the law.............................................................................................................................19 Awareness at the institutional/duty bearers level...................................................................................................................19 Awareness creation sessions ..........................................................................................................................................................20 4.2 CONCLUSION AND COMMENTS .............................................................................................................................................20 4.3 RECOMMENDATIONS .................................................................................................................................................................21 Page | ii

List of Figures
Figure 1: Distribution of respondents by age-groups .................................................................................................... 7 Figure 2: Distribution of Respondents Categories by Sex ............................................................................................ 8 Figure 3: Distribution of respondents by faiths ................................................................................................................ 8 Figure 4: Percentage of responses on the existence of the law on domestic violence in Uganda ................... 12 Figure 5: Percentage of the Rightful and misconceived actions under the DVAct 2010 ..................................... 13 Figure 6: Percentage of respondents identifying the rightful persons protected by the Law .............................. 14 Figure 7: Percentage of responses on the true and false benefits of the DV Act 2010.15 Figure 8: Percentage of Respondents who are aware of the duties of the Police according to the DVA 2010 ............................................................................................................................................................... 16 Figure 9: Percentage of Respondents who are aware of the duties of the LC courtsaccording to the DVA 2010 ............................................................................................................................................................... 17 Figure 10: Respondents on the duties of the HealthCare Workers and the Courts of law according to the DVA 2010 ........................................................................................................................................................................ 18

Page | iii

List of Tables
Table 1: Number of Interviews by Method of Data Source and District4 Table 2: Quantitative Survey Response Rates ......................................................................................................................... 7 Table 3: Percentage of respondents justifying wife beating ................................................................................................. 9 Table 4: Percentage of respondents who disagree witha man or woman denying a partner sex if they do not feel like ................................................................................................................................................................................ 10 Table 5: Percentage of respondents agreeing that a man has to exercise power and control over a woman ....... 11

Page | iv

ABBREVIATIONS
AIDS: CA: DV: DVB: FGD: HIV: IPV: KIs: LC: SA: UBOS: UDHS: ULRC: UN: VAW: WHO; Acquired Immune Deficiency Syndrome Community Activist Domestic Violence Domestic Violence Bill Focus Group Discussion Human Immune Deficiency Virus Intimate Partner Violence Key Informant Interviews Local Council Spouse Abuse Uganda Bureau of Statistics Uganda Demographic Health Survey Uganda Law Reform Commission United Nations Violence Against Women World Health Organization

CEDOVIP: Center for Domestic Violence Prevention DV ACT: Domestic Violence Act 2010

MOGLSD: Ministry of Gender Labour and Social Development

UWOPA: Uganda Women Parliamentary Association

Page | v

1.0 INTRODUCTION This DV Act 2010 Rapid Assessment Survey (RAS) is an evaluation conducted to assess current knowledge levels, attitudes and perceptions around the DV Act 2010. The RAS was conducted primarily among the duty bearers identified in the Act including police officers, healthcare workers, the Local Council officials (LCs) and the Community Activists (CAs). The survey adopted both qualitative and qualitative data collection approaches. Quantitative information was obtained from the duty bearers, while qualitative information was generated from individuals holding key positions among the duty bearers. The survey was conducted in ten communities administratively coded as sub-counties/divisions selected from some of the CEDOVIP intervention communities. The survey also assessed existing knowledge, attitudes and perceptions around violence against women (VAW) with particular focus on assessing attitudes towards general acceptance of VAW and types of domestic violence i.e. physical, emotional, economic and sexual violence as well as community attitude towards mens use of power over women. The baseline findings are expected to inform the intervention design and follow-up evaluation to determine the effectiveness of CEDOVIPs intervention. 1.1 DOMESTIC VIOLENCE AND VIOLENCE AGAINST WOMEN The DV Act 2010 defines domestic violence as any act or omission of a perpetrator which harms, injures, or endangers the health, safety, life, limb or wellbeing whether mental or physical, of the victim or tends to do so and includes causing physical abuse, sexual abuse, emotional, verbal and psychological abuse and economic abuse. According to the DV Act, any member of the household can be a victim of domestic violence and not just women and children. The Act provides protection for men, women, children, dependants and domestic workers. "1 The magnitude of the problem of DV is evidenced from the findings of various local, national and multi-country studies. For instance, an analysis of 10 separate domestic violence prevalence studies by the Council of Europe showed consistent findings that: 1 in 4 women experience DV over their lifetimes2. In the same trend, a study by the World Health Organisation (WHO) conducted on DV in 2003 in 10 countries found DV to be widespread in all 10 countries studied, though there was considerable variation between countries, and between cities and rural areas. While another study by Walby and Allen conducted in 2004 in London, found out that annually, there are over 13 million separate incidents of physical violence or threats of violence against women from partners or former partners. 1.2 DOMESTIC VIOLENCE AND VIOLENCE AGAINST WOMAN IN UGANDA Similarly, various studies in Uganda expose alarming DV prevalence rates across the country.The Uganda Demographic and Health Survey (UDHS) 2011 found that 56.1 % women have experienced physical violence since the age of 15, 27.7% experience sexual violence, 43% experience emotional violence while 16% experience violence while pregnant. Prior to this, the UDHS 2006 found that 60% of women had experienced violence with
1 2

Republic of Uganda, Domestic Violence Act 2010), Council of Europe (2002). Recommendation of the Committee of Ministers to member States on the protection of women against violence

Page | 1

49% reporting physical violence from their intimate partners, 36% experienced sexual violence and 49% had experienced emotional abuse3. Furthermore, the UHDS 2006 reported that rural women are more likely to experience each type of violence from their husbands than urban women are, with the eastern region having the highest statistics. These findings illustrate the trend in the prevalence of DV in rural Uganda. For the urban settings a study conducted in Kampala by Raising Voices in 2007 08 revealed that almost a half of ever-partnered women reported having ever experienced at least one act of physical or sexual violence by an intimate partner 4. This is a clear indication that DV in Uganda cuts across social classes. In terms of the victim/perpetrator ration, research showed that women are more affected. For instance, a study by Walby and Allen on VAW reveals that women are much more likely than men to be the victim of multiple incidents of abuse, and of sexual violence. The findings of the same study pointed out that 32% of women who had ever experienced DV reported that they did so four or five (or more) times, compared with 11% of the men with the same experience 5. The same study revealed that women constituted 89% of all those who had experienced four or more incidents of DV6. Locally, the UDHS 2011) pointed to men as perpetrators of Intimate Partner Violence (IPV) with forty percent of men reporting ever perpetrating physical IPV, recent episodes of IPV were also pervasive, with 34.9% of women having experienced physical violence, 24.8% sexual violence and 39% psychological violence in the last 12 months. The causes of such violence in Uganda are rooted in the unequal power relations between men and women, unequal rights to control resources within the home, biased legal framework that favors men, and patrilineal customary norms such as payment of bride price that create a sense of ownership of women by husbands The high prevalence of DV has been strongly associated with many social, economic and health offshoots. Research has proved that VAW has serious consequences on womens physical and mental health; women who have experienced abuse from their partners may suffer from or bare chronic health problems of various kinds7. Locally, the SASA baseline study (2009) findings confirmed associations between sexual risky behaviors and IPV. Both women who had experienced and men who had perpetrated IPV in the past year were more likely than those without a past year history of violence to have had multiple and/or extra-marital partners during the same time frame. Male perpetrators of IPV were also less likely to have used a condom in the last 12 months of their most recent relationship and the last time they had sex, hence the connection between DV and high HIV/AIDS infection rates. A more recent study by CEDOVIP 2012 on the Economic cost of Domestic violence found that individuals spent about UGX 19.5 billion on out of pocket expenditures; of this amount, more than 50 percent is spent on paying for police services while accessing health care providers takes 25 percent and LC courts 14 percent. Furthermore,
3 4

UBOS (2011), Uganda Demographic Health Survey Raising Voices, ( 2009), The SASA Study Baseline 5 Walby, Sylvia and Allen, Jonathan (2004) Domestic violence, sexual assault and stalking: Findings from the British Crime Survey
6 7

Walby, Sylvia (2004) The cost of domestic violence (London: Women and Equality Unit) British Medical Association (1998) Domestic Violence: A Healthcare Issue? (London: BMA)

Page | 2

service providers spend UGX 21.9 billion dealing with DV8. This means that the annual estimate of economic burden of domestic violence in Uganda stands at UGX 77.5 billion. To put the above estimated costs in perspective, they represent 0.35 percent of Ugandas GDP in 2011 (UGX 22,174 billion in 2011). This remains a huge loss to household incomes and a burden to the state. Indeed, the estimated costs borne by public providers of UGX 56 billion (i.e. health facilities, police, and judiciaries) is about 0.75% of Ugandas national budget in 2010/11 (UGX 7,376 billion). 1.3 THE DOMESTIC VIOLENCE ACT 2010 The prevalence of DV across communities in Uganda had a fertile ground since there was no legislation to criminalize DV until 2010. The absence of a specific law to address DV instigated great concern among the different Rights Activists in Uganda and the government to forge a way to outlaw DV. Consequently, in 2006 up to ten womens Rights and HIV organizations under the coordination of Center For Domestic Violence Prevention (CEDOVIP) formed a coalition to that was influential in working with the Uganda Law Reform commission in the formulation of the current DVA and lobbying for its enactment into Law in 2009. The Act was assented to by H. E the President in March 2010; hence, the Domestic Violence Act 2010 and subsequently, the regulations followed in 2011. This was the first time Uganda got a legislative tool that cast a ray of hope for the thousands of women and children who are the prime victims of domestic violence9. The DV Act 2010 provides for the protection and relief of the victims of domestic violence; the punishment of the perpetrators of domestic violence; and it provides for the procedures and guidelines to be followed by the court in relation to the protection and compensation of the victims of DV among its other functions. Thus, the line ministry together and civil society received it with delight believing that the Act will ensure fairness in dealing with violence in domestic affairs.10 1.4 THE DOMESTIC VIOLENCE ACT 2010 RAPID ASSESSMENT SURVEY (RAS) CEDOVIP and the DVA coalition embarked on a campaign of popularizing the DVAct 2010 among the communities in Central and Eastern Uganda as well as building the capacity of the Duty bearers (as defined in the DVAct 2010). In order to ease the administration of this law, CEDOVIP and the DV Act Coalition further translated the Act into eight local languages including Luganda, Runyoro- Rutooro, Runyankole-Rukiga, Rukonzo, Acholi, Akarimonjong, Alur and Ateso. In addition, CEDOVIP is planning a series of capacity building activities among the various groups of community members and duty bearers to create or increase awareness around the Act and to influence peoples attitudes positively towards the Act. Prior to the commencement of the capacity building activities CEDOVIP commissioned a baseline rapid assessment survey to assess existing knowledge levels, attitudes and perceptions towards the Act. Thus, this Rapid Assessment Survey (RAS) is part of the processes to guide programming and to assess the impact of these interventions.

8
9

CEDOVIP (2012) The economic Cost of Domestic Violence In Uganda.

Republic of Uganda, Domestic Violence Regulations 2010; Preamble 10 The Other Voice; Uganda Media Women's Association, Newsletter; Sunday January 9, 2011

Page | 3

2.0 STUDY METHODOLOGY 2.1 STUDY APPROACH, DESIGN AND DATA COLLECTION METHODS To achieve the survey objectives both quantitative and qualitative methods of data collection were employed to generate the required information. Structured interviews were used to collect quantitative data while qualitative data was gathered through the Focus Group Discussions (FGDs) and Key informant (KI) interviews. In addition to the above methods, a desk review of key related documents was conducted to compare or supplement the study finding with existing information. The use of multiple survey methods was to facilitate triangulation of the data from the different methods. 2.2 KEY INFORMANTS INTERVIEWS Key informant interviews were conducted among people who were assumed to hold comparatively higher knowledge on issues of the community gender social set up and the related legal regulations. Key informants were drawn from key positions among the duty bearers outlined in the Act. These included categories such as the District Community Development Officers (DCDO), Probation officers, District Health Officers (DHOs) and District Local councilors, and police officers from the Child and Family Protection Unit. 2.3 FOCUS GROUP DISCUSSIONS (FGDS) To supplement information from the KIs and the structured questionnaires, six FGDs were conducted in all the study districts. From each district, two FGDs were conducted; one with the female and one with the male participants (refer to table 1). The number of participants in each FGD was ten community members. The selection of FGD respondents was limited to the community members at the level of opinion leaders [communally respected and believed to be well informed about community social and legal affairs] in their respective communities. The FGD respondents were selected through the snowball sampling method guided by the local council leaders. The study supervisors moderated the FGD interview sessions. Table 1 presents a summary of sources of qualitative data. As shown, 27 qualitative research participants were recruited for both the KIs and FGDs for each study district. Altogether, 60 respondents were recruited for the FGDs while 21 respondents participated in Key informant interviews.
Table 2: Number of Interviews by Method of Data Source and District
Types of respondents DCDO DHO DPC In charge CFPU Probation Officers Magistrates Female Community Members Male Community Members Total Kampala 1 1 1 1 2 1 10 10 27 Kamuli 1 1 1 1 2 1 10 10 27 Iganga 1 1 1 1 2 1 10 10 27 Total 03 03 03 03 06 03 30 30 81 Type of information KI KI KI KI KI KI FGD FGD

Page | 4

2.3 STRUCTURED QUESTIONNAIRE INTERVIEWS In order to generate quantitative information, a descriptive cross-sectional study involving a statistically representative sample of respondents drawn from the three target districts was conducted. Individual survey respondents were interviewed using structured questionnaires. The questionnaires were translated into the popular local language spoken in the study districts. The questionnaires were pretested before the actual survey. From this exercise, all the content and translation related issues identified were addressed. All respondents were requested to provide oral consent before the interviews. The principal aim of quantitative part of the study was to quantify the levels of knowledge, attitudes and perceptions regarding VAW and the DVA. Structured interviews were administered among all categories of duty bearers. The interviews were personal in nature, conducted in a mode of one-to-one with the researchers reading the questions and recording the responses forthwith. This enabled information that was personal in nature to be explored from respondents. 2.5 SAMPLE SIZE AND THE SELECTION OF STRUCTURED QUESTIONNAIRE RESPONDENTS In determining the appropriate sample size for the survey, the most recent Uganda Bureau of Statistical Population Estimates (UBOS 2011) data was used and the following assumptions were considered in order to generate an optimal sample size for the study: 1. The main variable of interest (p) was taken to be " number of residents in the study areas " and according to UBOS (2011) p = 2,633,700 2. A confidence level of 95%, thus the standard normal variate took the value of 1.96 (two tailed) 3. The response rate was estimated at 92.8%, UDHS (2011) 4. Permissible error e was 5% of relative standard error (RSE) where RSE= Zxp An online sample size calculator developed by Creative Research Systems that uses the following statistical formula was used to determine the sample size. SS = Z 2 * (p) * (1-p) C2 Where: - Z = Z value (e.g. 1.96 for 95% confidence level) - p = percentage picking a choice, expressed as decimal (.5 used for sample size needed) - c = confidence interval, expressed as decimal (04 = 4) Substituting these values in above formula generated a sample size of 414. It was assumed that some of the respondents might not be interested in participating in the interview; therefore, some reserves were required, so the above sample size was multiplied by a factor of 1.064 that is 100/94 to get a figure of 441. Since the sampling design used for the study involved dividing the target population into categories distributed across the three districts, a design effect (DE) was factored in the calculation of the sample size. A minor value of 1 was used since there were no data on design effect for the main variable. The selection process for the particular respondents categories was as shown below 2.5.1 The Police From each of the three study districts, the selection of police officers was through simple random sampling. This involved the researcher approaching the DPC who was also to be interviewed as a KI. The research team acquired a list of police stations in the respective district from the DPC. The researcher then randomly selected eight police posts from the same list. All the officers attached to the selected posts would qualify for the study but
Page | 5

the researcher had to randomly select 4 from each police post for the interview. Special priority was given to those in the child and family protection unit (CFPU). The total number of police for the quantitative survey was thus 36 for each district. These included the 32 selected from the posts across the district and another 4 officers randomly selected from the central police station in each district. 2.5.2 Health Workers The same procedure as in the selection of the police officers was applied in the selection of health workers. However, this time the lists were obtained from the DHOs. Following the same procedure, a total of 36 health workers were selected from each of the study districts. These included the 32 selected from the eight health units across the district and four others randomly selected from the main hospital or Health Center IV in each district. 2.5.3 Local Council Leaders The selection of the LCs leaders was also random. For Kamuli and Iganga, the lists of zones/villages were obtained from the DCDOs. From these lists, the study randomly selected eight zones/villages. The LC executive officers representing each of the selected zones qualified to participate in the study. From each zone/village the researcher selected 4 executive members to participate in the study. This procedure generated 32 respondents from each district. The study also recruited four LC 2 chairpersons from each of these districts, hence the total number of 36 LCs from each district. In Kampala the list was provided by CEDOVIP, the researchers then randomly selected the eight villages/zones for the study. From each of the selected zones/villages, the researchers approached the LC 1 chairperson to help them identify the other executive members. Through snowballing, the researchers approached the two LCs executive members required from each zone/village for the interview. 2.5.4 Community Activists The selection of Community Activists (CAs) was also done randomly. In Kampala, the lists were acquired from CEDOVIP containing all CAs. In Busoga, the selection of CAs, altered from the original plan because the planned number of CAs to be selected was (36) in each district, however, the number of the CAs in Kamuli and Iganga 24 from each district. Consequently, all the 24 CAs in each of the two districts were recruited in the study.

2.6 ELIGIBILITY CRITERIA Eligibility in this survey was defined as all men and women aged between 18-69 years who were residents of the sampled zones/villages. Eligibility for the police officers, health workers and magistrates was guaranteed as long as the selected person was identified in those categories. However, the age limit set at 18-69 years applied to all respondents.

Page | 6

3.0 PRESENTATION OF RESULTS 3.1 THE DVACT 2010 BASELINE RAPID ASSESSMENT SURVEY COMMUNITIES AND RESPONSE RATES The numbers and percentages of different responses to survey questions are presented in the Table below. The results for each study district are presented in the same table indicating the numbers and percentages of each of the four categories of respondents from communities reached.
Table 2: Quantitative Survey Response Rates

Respondents categories Community Activists Health care workers Local Council officials Police officers Total

Iganga Frequency 24 36 36 36 132 Percent 28.6% 33.3% 33.3% 33.3% 91.6% 36 36 36 36 144

Kampala Frequency Percent 42.9% 33.3% 33.3% 33.3% 100% 24 36 36 36 132

Kamuli Frequency Percent 28.6% 33.3% 33.3% 33.3% 91.6%

Overall, the response rate for the survey was very high across all categories rated at 94.4%. Relatively lower response rates were among the CAs. However, the drop in the CA response rates was due to the actual number of CAs being lower than the planned sample for this category. The earlier plan was to engage 36 CAs from each district; however, the number of CAs in Busoga region was slightly lower than estimated. Each one of the two Busoga districts had twenty-four CAs representing 28.6 percent CAs from Iganga and Kamuli districts respectively. 3.1.0 DEMOGRAPHIC CHARACTERISTICS OF SURVEY RESPONDENTS Figure 1: Distribution of respondents by age groups Majority of respondents were between 30 39 years constituting 30 percent of the entire sample. Close to this range were the respondents in the 40 49 age range constituting 29%. Other significantly represented age groups were 18 29, 50 59 and 60 69 constituting 26%, 12% and 3% respondents respectively. The mean age of respondents was 38 years. The age distribution in the survey was representative of the National population age distribution in the civil service and community leadership.
Page | 7

The sex distribution of respondents in the overall sample was almost equal, with 51.2 percent of respondents being female and 48.8 percent being male as seen in Figure 2. Figure 2: Distribution of Respondents Categories by Sex Overall, there were more female respondents representing 51% of the study population compared to 49% male respondents. The proportion of male respondents was high among the health workers, police officers and among the CAs representing 63%, 53% and 50% of respondents in each of the three categories. On the other hand female respondents were high among the Health workers representing 73% of the respondents in this category.

Figure 3: Religious distribution of respondents Distribution by religion was: Christians 76%, Muslims 23% and the Traditionalists 1%. The distribution resonates with other national surveys with more Christians followed by Muslims and lesser traditionalists.

Page | 8

3. 1.1 ATTITUDES TOWARDS PHYSICAL VIOLENCE Domestic violence includes a range of sexually, psychologically and physically coercive acts used against adult and adolescent woman/man by current or former intimate partners11. However, physical violence remains the most visible due to the injuries sustained and the most common form of domestic violence reported. To assess existing awareness and attitudes towards physical violence, the respondents were asked if a husband is justified to hit his wife for any of the options in Table 3. Only respondents that agreed with the provided options are shown below: Table 3: Percentage of respondents justifying wife beating Action Community Health Activists workers Local Council Officials n=108 55.7% 52.0% 55.7% 65.7% 58.3% 64.8% 61.1% 59.04% Police officers Percentage who agree with any specific reason n=408 47.25% 46.13% 42.40% 53.33% 47.60% 63.43% 57.00% 51.05%

Number of Respondents She denies him sex She answers back to him She demands for his money She disobeys him She disrespects his relatives He suspects that she is unfaithful Percentage per respondent Category

n=84 31.2% 31.2% 30.0% 35.2% 33.6% 64.8% 39.60%

n=108 42.8% 44.8% 30.2% 52.2% 41.1% 66.7% 62.0% 48.54%

n=108 59.3% 56.5% 53.7% 60.2% 57.4% 57.4% 53.7% 57%

She neglects taking care of the Children 51.2%

Overall, more than half of the respondents (51%) believed that at least one reason was sufficient justification for wife beating. The results indicated that wife beating was acceptable for any of the reasons indicated in table 3. The most widely accepted reason for beating a wife was suspected unfaithfulness mentioned by 63.4% survey respondents. The other justifications for wife beating according to the respondents included neglecting to take care of the children; disobeying the husband; denying the husband sex and disrespecting husbands relatives identified by 57%, 53%, 47.3% and 47.6% survey respondents respectively. However, analysis points to the social acceptance of power and control by men over women and belief in rigid gender roles, thus, the justification for any means used to enforce mens use of power over women.

11

Department of Population and Family Health Sciences, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, USA. Correspondence should be addressed to Dr. M.A. Koenig (email:[email protected]).

Page | 9

Close to three quarters of the LCs and Police representing 59% and 57% respondents in each category respectively held the view that wife beating is justifiable for any reason in table 3. Ironically, profession and service mandate the Police and LCs to enforce laws yet many of them still justify wife battery. In the same trend, the rates among the Health workers and the CAs were significantly high, with 40% CAs and 49% health workers supporting wife beating for any of the options in table 3. A comparison of the study findings with the UDHS 2011 survey findings on attitude towards wife beating, points to an almost similar conclusion. In the UDHS 2011 58% women, justified wife beating for any single reasons while among the men it was 44% with the same perception. The current results indicate that there is generally high acceptance of VAW and the right of a man to control his wifes behavior through violence. 3. 1.3 ATTITUDES TOWARDS SEXUAL VIOLENCE The right for adult men and women to engage in sexual activity at free will makes the activity equally beneficial to both. In the survey, this was measured when women and men were asked if it is okay for a man or a woman to deny a partner sex if they did not feel like. The other option was if the wife had right to insist on condom use. Table 4 presents the responses of the respondents who said No to all options.
Table 4: Percentage of respondents who disagree with a man or woman denying a partner sex if they do not feel like
Community Activists Health workers Local Council officials n=108 53(49.1%) 57(52.8%) 60(55.6%) 42(38.9%) 49.10% Police Officer Percentage who disagree with any specific condition n=408 42% 41% 46% 37% 41%

n=84 Is it okay for a woman to deny her husband sex if 29(34.5%) she does not feel like it? Is it acceptable if a married woman refuses to 24(28.6%) have sex with her husband? Is it okay if a wife insists on using a condom even 34(40.5%) if her husband does not want to? Is it okay for a man to deny his wife sex if he does 23(27.4%) not feel like it? Percentage per respondent category 32.70%

n=108 44(40.7%) 44(40.7%) 43(39.8%) 47(43.5%) 41.18%

n=108 46(42.6%) 45(41.7%) 50(46.3%) 40(37.0%) 41.9%

Overall results indicated that there was high objection to womens control over their bodies, sexuality and reproductive health. There was high objection (43%) to all the three options on womens sexual rights. More specifically, 42% respondents objected to a woman denying her partner sex if she does not feel like it, 46% objected to a womans right to insist on condom and 41% objected to a married womans right to deny her husband sex if she does not feel like. On the contrary, the results indicated that the men are allowed

Page | 10

relatively higher sexual freedom compared to their female counterparts with only 37% respondents objecting to a mans right to refuse to have sex with his partner if he does not feel like it. Attitudes towards womens sexual freedom were very negative across all categories of respondents with more LCs (49.1%) objecting to denial of sex to a partner for any reason and insisting on condom use. The other respondents categories also highly shared the same opinion with 41.18% Health Workers, 41.9 Police officers and 32.7% CAs objecting to all options in Table 4.
3.1.3 POWER AND CONTROLLING BEHAVIORS/ EMOTIONAL VIOLENCE

Misuse of power and control of one party over another is the root cause of violence among intimate partners. The survey investigated the attitudes towards emotional violence in forms of; possessiveness, disciplining, exercise of authority and power through threats and control over the right to work. Table 5: Percentage of respondents agreeing that a man has to exercise power and control over a woman
CAs Health workers LCs Police Officer Percentage who disagree with any specific condition n=408 25% 25% 58% 45% 46% 40%

n=84 Is it okay for man to discipline his wife through shouting if she has done something wrong? Are women sometimes to blame for violence against them? Is it true that possessiveness and jealousness is a way in which a man shows love to his wife/partner? It is a husbands role to decide whether his wife can work outside their home? Is it true that a man needs to show he is the head of the house by using threats? Percentage per category of respondents 20.3% 16.5% 64.3% 46.0% 38.3% 37%

n=108 27.0% 28.0% 56.5% 49.3% 42.0% 41%

n=108 24.3% 28.8% 60.2% 32.8% 47.4% 39%

n=108 28.4% 26.7% 50.9% 41.9% 55.7% 41.0%

Close to three quarters (58%) of respondents thought that, controlling behaviours such as possessive and jealousness are ways in which a man shows love to his wife/partner. In the same trend, 46% respondents indicated that it is okay for a man to prove that he is the head of the house using threats and intimidation and 45% indicated that it is a husbands role to decide whether his wife can work outside their home. There was relatively lower but significant percentage of respondents (25%) indicating that it is okay for man to discipline his wife through shouting if she has done something wrong and that women are sometimes to blame for violence against them. Overall, the levels of acceptance of controlling behaviours and mens use of power over women were high across all categories of respondents with close to half of respondents (40%) approving all the options in Table 5.
Page | 11

i.e. for the Health Workers and the Police the percentage was 41% for each category. While the LCs and CAs were 39% and 37% respondents respectively. 3.2 THE DOMESTIC VIOLENCE LAW The DV Law came into force in April 2010. In order to assess awareness of the existence of the DV Act 2010, the respondents were directly asked if Uganda has a law on DV and the responses were limited to only three options Yes, No and Not sure as seen in Figure 4 below.

Figure 4: Percentage of responses on the existence of the law on domestic violence in Uganda

Overall, awareness of the existence of the law on DV was low across all respondents categories. Higher levels of awareness were among the police with 43% police answering Yes. However, among other categories, it was 28%, 23% and 18% Health care workers, CAs and LCs respectively answering Yes . Lower awareness level on the existence of the DV law was among the CAs with (46%) answering No. There was also a significant percentage of CAs 31% indicating that they were Not sure if Uganda had a law on DV. The percentage of respondents answering Not sure was almost uniform across all the four categories ranging from 30% among health care workers as the lowest to 32% among the LCs. 3.2.1RECOMMENDED ACTIONS UNDER THE DV Act 2010 The DVA 2010 outlines several remedies to DV to ensure justice to the survivors and perpetrator accountability. Question 8 (According to the DV Act 2010, a person experiencing violence can) was used to assess the respondents knowledge on the remedies for survivors and perpetrators, procedures and responsibilities of the duty bearers defined by the Act and several items requiring either of the following responses yes/no or not sure were presented. The list of items had both false and right options. Figure 5 shows the rightful and misconceived Actions under the Act, only percentages of yes and not sure are presented.
Page | 12

Figure 5: Percentage of the Rightful and misconceived actions under the DV Act 2010

In Figure 5, the results indicated that there was high misconception of the actions provided under the law with close to three quarters of the respondents (57%) sharing at least one misconception. The most highly misconceived actions included forgiving and forgetting, fighting back in the event of DV and reporting to religious leaders all tying at 62%. Additionally, a significant percentage of respondents (23%) indicated that they were not sure if the stated options were legally approved actions or not. On the other hand, results on the legally approved actions in the event or suspicion of violence also indicated limited awareness. The options that were rightly identified by a significant percentage of respondents were; taking any action under the Act 47%, reporting the matter to Police 44%, and filing a case against the offender 43%. The average percentage of respondents saying not sure to all the options in Figure 5 was 31% The limited knowledge on the provisions within the Act pauses a serious threat as it is likely to result into victim blame and further victimization, perpetuation of silence among survivors because the response points of reference are not supportive to the survivors needs. In the process, such unsupportive environment may increase community revenge and mob justice.

Page | 13

3.3

AWARENESS ON THE BENEFITS OF THE DVACT 2010

3.3.0 Beneficiaries of the DVA 2010

Section 3 of the DVA 2010 defines Domestic Relations as a family relationship; a relationship similar to a family relationship or a domestic setting that exists or existed between a victim and a perpetrator. The DVA 2010 stresses that every individual in a domestic setting including children, the wife, husband [intimate partners], domestic workers, person in a past relationship, person sharing the same residence and any one in a relationship determined by the court has a right to protection under the law. Question 9 (Who of the following is protected under the DV law?) was aimed at assessing respondents awareness in this area. Figure 6 indicates the combinations of persons protected by the law identified by the respondents.

Figure 6: Percentages of respondents identifying the persons protected by the Law

There was very limited awareness around the persons protected by the DV law. Only 3% respondents indicated that the law protects the women, men, children and domestic workers. Majority of the respondents 38% reported that the DVA protects only the women. The options of men only and children only were each selected by 2% respondents respectively. The other combinations identified included those who reported that the Act only protects the children and woman 11%; 14% said it protects children men and women while 3% said that it protects children, domestic worker and women only.

Page | 14

3.3.1 Specific benefits of the DV Act 2010

To get respondents knowledge levels on the specific benefits of the DVA. Several false and right items were set and the respondents were required to answer, either yes, no or not sure. Figure 7 illustrates the respondents who answered yes and Not sure to the selected options.
Figure 7: Percentage of responses on the true and false benefits of the DV Act 2010

The results indicated low awareness of the benefits of the Act among all respondents with a siginificant percentage of respondents indicating that they were not sure and an overall percentage below average of respondents saying yes to the true benefits of the Act. For the true benefits of the Act the most highly sellected option was Empowering the victim, neighbors, and friends to report domestic violence cases at 42% followed by protecting all survivors from abuse 40%, mother can keep her children with her at 34% and outlaws all acts and forms of domestic violence by 33% respondents. The overall mean percentage of respondents saying not sure to the true benefits of the Act was 36%, with majority 43% saying they were not sure if the Act outlaws all acts and forms of DV. On the other hand, 43% identified forgiving the offender as a benefit of the Act. The other options like allowing man to discipline wife, punishing the victim, and only police can handle DV cases were selected by 34%, 32% and 28% respondents respectively. The options that registered a significant number of respondents indicating that they were not sure included the 31% saying not sure to only police can handle DV cases and 30% saying not sure to forgiving the offender. In the same trend, 29% and 28% respondents respectively answered not sure to punishing the offender and allowing man to discipline the wife. This indicates that there is low awareness around the benefits of the Act across all groups of respondents.
Page | 15

3.5 AWARENESS ON THE ROLES AND JURISDICTIONS OF RELEVANT INSTITUTIONS AND DUTY BEARERS The DVA 2010 mandates and defines roles for the Local council court, the police, the magistrates court, and the Family and Childrens court and the health practitioners to handle DV cases. To assess the respondents knowledge levels on the roles and jurisdictions of the above institutions and duty bearers a list of roles was provided for respondents to select the right combination of at least three options for each role on the list. The section below presents the responses.

3.5.1 Duties of Police Officers

Section 7 of the DV Act 2010 outlines the duties of a police officer. In addition to the traditional role of investigating complaints, a police officer are supposed to give assistance or advice in obtaining shelter for the victim; for sexual or physical violence cases to ensure that the victim undergoes medical examination and receives treatment. The police also, advise the victim on the right to apply for relief; as well as offering guideline to ensure that the victim, the witness or the victims representatives are safe. Lastly, the police are supposed to record the victims statement or their representative on the nature of the DV. The results are indicated in Figure 8.
Figure 8: Percentage of Respondents who are Aware of the Duties of the Police according to the DVA 2010

Figure 8 above indicates that there was average awareness around some of the duties of the police. However, higher awareness was around the traditional roles of the police including arresting and detaining suspect (60%); taking victims statement (59%) investigating complaint and apprehending suspect identified by 53% and 33% respondents respectively. The respondents that identified wrong roles for the police included the 52% said that the police document the visits of the victim, 46% said that the police are responsible for compensation and 45% said the police are responsible for reconciliation. The results indicate that there is limited awareness on the roles of the police in handling DV cases as stipulated in the Act.
Page | 16

3.5.2 Jurisdictions of the Local Council Court Sub-section 5 of section 6 of the DV Act 2010 indicates that in addition to recording relevant documentation on the availed case of DV, the LC court may in exercise of its powers make the following orders for the victim or against the perpetrator: caution, apology to the victim, counseling, community service, a fine, compensation and reconciliation. Several options true and false item were provided and respondents were supposed to select the rightful options from a list. The results are shown in Figure 9
Figure 9: Percentage of Respondents who are aware of the duties of the LC Courts according to the DVA 2010

According to figure 9 above, there was limited awareness around the duties of the LC court, only 39.7% identified fining the perpetrator; 36% identified cautioning, 33.3% identified reconciliation while 8.2% respondents identified compensation. However, more than half of the respondents identified several wrong options. 3.5.3 Duties of the Health care Workers (Practitioners) and the courts of law in the DVA 2010 Sections 8 and 9 of the DVA 2010, elaborate the jurisdictions of the health care worker (practitioner) and magistrates court respectively. According to the Act, a practitioner who suspects that the person under his/her care is a victim of domestic violence is supposed to: Offer the requisite medical assistance to the victim; Accurately document the visits of the victim; Inform the victim of the options available in the judicial system and to make him/herself available to testify in court in case there is need. Regarding the magistrates courts, the Act states three major duties including: hearing any case of DV presented; issuing a protection order for the victim and conduct all its duties in the same manner as the child and family court.

Page | 17

The respondents were interviewed about the duties of the health care workers and the magistrates court. Several true and false options were given for respondents to choose the most appropriate. Figure 10 combines the findings on both questions.
Figure 10: Respondents on the duties of the HealthCare Workers and the Courts of law in the DVA 2010

The results in figure 10 above indicate that many respondents (50%) were aware of the health care workers traditional role of providing medical assistance. However, awareness around the other duties of health care workers was far below average. Only 11% respondents were able to identify making him/herself available to testify in court and 9% identified documenting visits of the victim as duties of the health care workers. There was a significantly high percentage of respondents who selected wrong options as duties of the health care workers, these included the 13% who mentioned compensation, an almost equal percentage of 10% respondents said that the health practitioners are responsible for investigating complaints and advising victim to apply for relief. On the other hand, the duties of the magistrates were highly misunderstood with respondents mentioning the duties of other bearers in the place of the magistrates court. There was no one that mentioned any of the three duties of this court as mentioned in the preamble to this section. Overall, awareness level around the various duties of the health workers was generally low amongst all respondents while the duties of the magistrates court were completely unknown to all respondents.

Page | 18

4.0 QUALITATIVE RESULTS AND DISCUSSION OF STUDY FINDINGS Knowledge, attitudes and perception of VAW The results indicated low awareness around VAW and high levels of acceptance of several behaviours and practices that perpetuate VAW. The results illustrated a general acceptance of all forms of violence across all the study communities. The general belief was that any form of violence could be justified as long as the husband is perceived to be disciplining his wife. Information gathered through KIs attributed this to negative attitudes towards womens low status in society and rights and lack of awareness around what constitutes violence. This comes out clearly from the quote that, Indeed all forms of violence are common here but the truth is that majority of the men do it out of ignorance. For example, one will excitedly tell you that, they do not beat a wife but just discipline her by denying the supplies at home. PO Kamuli TC. In addition, sexual violence was widely accepted by both men and women in the community. However, gender inequality points to the favor of the men over the women with the former having no right to object to sex with a formal partner for any reason because that is the reason they are married. In the same way, a large number of the participants denounced the use of condoms among couples. To this, several FGD respondents wondered, A condom, with your husband! How do you begin suggesting it? That can only work where there is proof from the doctors that one of you is positive and the other one is negative. FGD respondent in Iganga Awareness around the existence of the law The results also indicated low awareness around the existence of the DV law across all study communities. Majority of the FGD respondents did not know the existence of the law. However, since the study coincided with the public debate on the Marriage and Divorce Bill, many community members said that the law was just underway but not yet in place. Limited awareness of the existence of the law was across all social classes and levels of education. To this, the PO Kamuli district noted that, Awareness of the existence of the DV Act 2010 could only be among the elite in our community and I think across the country. And you will find education does not count in this because even our partners in CFPU, I bet if more than twenty percent of the officers know its existence. Awareness at the institutional/duty bearers level The results also revealed limited awareness of content of the DVA across all the relevant institutions. In Kampala and Iganga districts, the police in the CFPU heads sounded a little more aware of the existence of the Act. However, most officers only knew that the law exists but were not familiar with the content. A big number of the officers said they had never seen or read the law or even attended a session on the law. One police officer noted that, You will find that many officers are green, they have not seen that Act. For us we use the compendium of the law most frequently but the one we have here is the 2008 version and you are talking about 2010, where do you think people can read about it. This also revealed the problem of limited reading materials in relation to the law and discriminative awareness creation approaches that only
Page | 19

targeted the people in big position, which were found to be ineffective in feeding their colleagues with new skills. Awareness among the heath workers was very low as well. Many health workers do not know what the law requires them to do in the event of a case of DV. Many heath workers sounded detached from the prescriptions of the law and felt it was police work. To this end the Administrator Kamuli Hospital noted that, there is completely no knowledge around that Act among the medics, you even tell from the way they document DV cases, they only write things like cut, wound, with nothing explaining the origin of the wound in relation to DV. How do you expect to investigate such a case? The results also indicated lack of collaboration between the health workers and the police with the former accusing the later of obstructing justice. To this the OC in charge CFPU Iganga noted that, There is need to engage the health workers to collaborate with the police, for them to fill in form 3 they need 20000/= if not a police officer has to physically accompany the survivor to the health center, yet we do not have enough staff to divide like that. This limits the progress of investigations around DV cases. The same was with the LCs Awareness creation sessions There was very limited mention of any engagement sessions aimed awareness creation around the DV Act 2010 by data collection time. There was much concern among the Probation Officers based at the sub counties saying all awareness creation activities target the district-based officers who never share the information. That is why the DCDO Kamuli noted that, Even the Probation Officers (POs) are not aware of the content of the Act. Most of my POs have never even had a chance to look at the Act yet with DV even the police refer to us here. In Kamuli there was mention of a training by UWONET but it targeted only the high ranking officers in the health and police departments of the district. This raised a lot of concern among the junior officers who directly deal with the cases but are never trained. 4.2 CONCLUSION AND COMMENTS From the data above, below are the general conclusions; 1. There is general acceptance of violence against women in Uganda basing on the attitudes towards violence in the society. 2. There are low levels of knowledge about the Domestic Violence Law and its contents among the communities, and the duty bearers. 3. Duty bearers including police, health care workers, and Lc are not aware of their roles and jurisdiction as mandated by the DV Act thereby putting access to justice for survivors in a balance. 4. The negative attitude on VAW clearly indicates a very high ground or level for victim blame and their denial to justice. 5. This is also a clear indication that if the situation is not addressed (awareness rising) violence will continue with impunity.
Page | 20

6. The law enforcement institutions dont seem to have done much to enable their officers to implement the new laws, there are no significant actions taken by them. 4.3 RECOMMENDATIONS 1. There is urgent need to train duty bearers on the content and application of the DVA 2010 2. CEDOVIP needs to check in with the leadership of the law enforcing bodies to know what their implementation plan of the DVA is 3. There is urgent need to create the publics awareness of the DVA, its contents and benefits of the Act 4. There is need to provide copies of the law, related literature or other IEC materials especially to the law enforcement institutions addressed specifically to the institution not to individuals because in most cases they are affected by the transfers 5. A lot of emphasis should be put on the CAs because unlike the duty bearers, the CAs are more accepted in the communities because of their voluntary approach to DV related cases 6. An aggressive campaign is urgently needed to influence the attitudes of the community members positively towards the law 7. A rigorous campaign is urgently needed to engage the health care practitioners on a range of activities including documenting DV cases, being more receptive towards DV survivors and a refresher on the urgency of voluntarism in serving DV survivors. 8. There was concern over the misleading information from some media houses around the law creating bias and negative attitudes based on misinformation. There is thus need for some sessions for the media reporters since they are held significantly in the communities 9. There is need to work around building cooperation among all the duty bearers so as they can hold the same attitude and approach towards the law

Page | 21

REFERENCES Action for Development; Sexual and Gender Based Violence in Uganda: Experiences of Sexual Violence among Women and Girls in Pallisa and Kisoro Districts, 2009 Asiimwe-Okiror, G., et al., Change in sexual behavior and decline in HIV infection among young pregnant women in urban Uganda. AIDS. Biraro, S., et al., Is sexual risk taking behaviour changing in rural south-west Uganda? Behaviour trends in a rural population cohort 1993-2006. Sex Transm Infect, 2009. Boyce C and Neale P.: Conducting in-depth interviews: A guide for designing and conducting in-depth interviews for evaluation input. May 2006. Pathfinder International Tool Services. Monitoring and Evaluation 2. British Medical Association (1998) Domestic Violence: A Healthcare Issue? (London: BMA) Center For Domestic Violence Prevention; The economic Cost of Domestic Violence In Uganda; 2012, Kampala Uganda Council of Europe (2002); Recommendation of the Committee of Ministers to member States on the protection of women against violence; Adopted on 30 April 2002 and Explanatory Memorandum. (Strasbourg, France Council of Europe) Koenig, M., et al., Domestic violence in rural Uganda: evidence from a community-based study. Bulletin of the World Health Organization, 2003. 81(1): p. 53-60. Raising Voices; The SASA Study Baseline Survey Report, 2009 The Other Voice; Uganda Media Women's Association, Newsletter; Sunday January 9, 2011 The Republic Of Uganda; The Local Council Courts Act 2006, Kampala, Uganda The Republic Of Uganda; The Domestic Violence Act 2010, Kampala, Uganda (UBOS), U.B.o.S. and M.I. Inc., Uganda Demographic and Health Survey. 2007, UBOS and Macro International Inc.: Calverton, Maryland, USA. (UBOS), U.B.o.S. and M.I. Inc., Uganda Demographic and Health Survey 2006. 2007, UBOS and Macro International Inc: Calverton, Maryland, USA Walby, Sylvia (2004) The cost of domestic violence (London: Women and Equality Unity Walby, Sylvia and Allen, Jonathan (2004) Domestic violence, sexual assault and stalking: Findings from the British Crime Survey (London: Home Office Research, Development and Statistics Directorate)
Page | 22