Vol.:(0123456789) ORIGINAL PAPER
Female intimate partner violence in Zimbabwe: does empowerment matter?
Marifa Muchemwa1 · Clifford Odimegwu1
Received: 23 July 2020 / Accepted: 26 July 2021 / Published online: 1 September 2021
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2021
Abstract
Globally, studies on gender-based violence and domestic violence have advocated for female empowerment to counter violence. Zimbabwe is among countries that have called for female empowerment as a remedy; however, the widespread violence against females persists largely perpetrated by males. This study examines whether female empowerment is associated with increased or decreased levels of intimate partner violence (IPV) in Zimbabwe. A cross-sectional study using data obtained from the Zimbabwe Demographic and Health Survey (ZDHS) of 2015 was con- ducted. The population of interest are females in a union (4917) and not in a union (2306). The main independent variable is female empowerment measured by the following proxies: full participation in decision-making, educational level, owner- ship of assets, and employment status. The dependent variable is IPV in two forms:
physical and sexual violence. A multivariate analysis using binary logistic regres- sion was used to identify the association between measures of female empower- ment and intimate partner violence among females in a union and not in a union.
The females who are in a union and empowered were significantly more likely to experience physical and sexual violence (Odds ratio: 1.38 and 1.82) than those not empowered. Despite being empowered, the married and single females experienced intimate partner violence. Effective interventions should note that empowerment of females might be viewed as undermining the power of males over females, resulting in males resorting to violence. Hence, attention should be given to patriarchy, soci- etal norms and values which promote a male dominant environment.
Keywords Intimate partner violence · Domestic violence · Female empowerment · Zimbabwe
* Marifa Muchemwa mmuchemwa@cartafrica.org
1 Demography and Population Studies Programme, Schools of Public Health and Social Sciences, University of the Witwatersrand, Johannesburg, South Africa
Abbreviations
DHS Demographic and health survey
ZDHS Zimbabwe demographic and health survey
Introduction
Intimate partner violence is prevalent worldwide, with almost a third of females who have been in a relationship having experienced physical or sexual violence (García- Moreno et al. 2013). Intimate partner violence refers to any form of violence by an intimate partner that causes physical, psychological or sexual harm to those in the relationship (Jewkes 2002). The forms of intimate partner violence include physi- cal, sexual and emotional violence (Nyamayemombe et al. 2010). Globally, intimate partner violence against females has been labelled as a public health concern and a human rights problem linked to different health, family, social and economic effects (United Nations Women 2014; Wekwete et al. 2014; Fidan and Bui 2016).
The prevalence of intimate partner violence in Zimbabwe is very high. A study found that 47% of females had experienced intimate partner violence with 28%
reporting physical violence, 29% emotional violence and 18% sexual violence (Mashiri 2013). Another study in Zimbabwe showed similar rates as 48% of females were reported to have experienced intimate partner violence with 26% experiencing spousal sexual violence, 25% spousal emotional violence and 27% physical violence (Wekwete et al. 2014). It is stated that two in three females in Zimbabwe have expe- rienced intimate partner violence in their lifetime (United Nations Women 2014).
The patriarchal society in Zimbabwe has been described as responsible for intimate partner violence (Mashiri 2013; Fidan and Bui 2016). The existence of patriarchy installs power in males over females and this allows them to dominate and control females (Mashiri 2013; Fidan and Bui 2016).
In Zimbabwe, the Domestic Violence Act of 2006 makes provision for the pro- tection and relief of victims of domestic violence (Domestic Violence Act [Chap- ter 5:16]). The Act was introduced to address various forms of violence such as physical, sexual, economic and emotional abuse. Prior to the Act, no law in Zimba- bwe addressed violence against women (Osirim 2003). Any form of domestic vio- lence which occurred was treated as a common assault with intent to cause grievous bodily harm or indecent assault (Osirim 2003). The Act aims to protect women and minimise any acts of domestic violence even those derived from any cultural prac- tices which discriminate against women, including virginity testing, female genital mutilation, pledging of women to appease spirits, forced wife inheritance, child mar- riage and forced marriages. This also includes sexual intercourse between fathers in law and newly married daughters in law as well as abuse perpetrated because of vic- tims’ age, or physical and mental incapacity or illness and disability. The Domestic Violence Act does not specifically address issues to do with the empowerment of women though it has some aspects which can be identified as speaking into empow- erment of women. This includes treating as a criminal offence, depriving women of economic or financial resources and denying them the right to seek employment or any activities aimed at generating income. In 1988 a non-governmental organisation,
Musasa Project, was established to deal with the awareness of the illegal nature of domestic violence in the country (Stewart 1995). Despite the Domestic Violence Act and the Musasa project’s awareness of violence, females in Zimbabwe continue to experience violence (Chireshe 2015). A baseline study on the prevalence of inti- mate partner violence in Zimbabwe unmasked that 50% of the females knew about the domestic violence Act, but chose not to report (United Nations Women 2014).
Studies argue that lack of female empowerment results in the females not being able to make decisions and many are economically dependent on their male part- ners (Mashiri 2013; United Nations Women 2014). However, other literature chal- lenges the assumption that greater empowerment is linked to lower levels of inti- mate partner violence (Rahman et al. 2011; Pambe et al. 2014; Green et al. 2015a, b). Female empowerment refers to the ability or power to make decisions, put into effect choices and also encompasses a shift from powerlessness to an empowered state (Kabeer 1999; United Nations Women 2014). This economic dependence has seen many females being exposed to physical and sexual violence (United Nations Women 2014). This has led to studies in Zimbabwe on gender-based violence, domestic violence and intimate partner violence advocating for the empowerment of females as a remedy to violence and also as a tool to equip females to be in a position to leave violent and abusive relationships (Tom et al. 2013; Hove and Gwa- zane 2011). Since female empowerment is recommended to be a panacea, this study examines the association between female empowerment and experiences of intimate partner violence among females in Zimbabwe.
In terms of research, this study will contribute immensely to the existing body of knowledge since studies in Zimbabwe have limitedly focussed on female empow- erment and intimate partner violence. Only one study has been conducted in Zim- babwe on the association between gender-based violence and female empowerment (Wekwete et al. 2014). The study resembles the current study; however, some of the female empowerment indicators used and the study population differ from this study. The study focused on females who were currently in a union and excluded females who were not in a union, widows and divorcees. The study further pointed to a need for further research which includes females who are not in a union and divorcees (Wekwete et al. 2014). In response to the suggestion, this study includes all females in Zimbabwe regardless of their marital status. By focussing on all the females, it is possible to identify the differences in experiences of intimate partner violence between the females in a union and those who are not.
Empowerment is a highly contested concept that can mean different things to dif- ferent people and it is context specific (Mahmud et al. 2012). It is not once-off, but rather a process that is dynamic and aimed at change for those who cannot make choices, to acquire such ability (Kabeer 1999). The concept of female empower- ment can be traced to the 1980s and 1990s. It was a radical approach aimed at trans- forming the existing power relations through addressing female rights, social justice and the restructuring of the economic, social and political structures (Kabeer 2005;
Mukhopadhyay 2007). In theory female empowerment should reduce violence, but other studies elsewhere have shown that the involvement of females in labour markets has led to tension and increased risk of violence between intimate partners (Castro et al. 2008). Sometimes the empowerment of females increases violence
because it challenges the status of gender power (Presser and Sen 2000). Apart from increasing females risk of violence, the empowerment of females has seen a reduc- tion in levels of intimate partner violence in some settings.
Methods
Study design and sample
The study utilised secondary data obtained from the Zimbabwe Demographic and Health Survey (ZDHS) which was conducted in 2015. The study population are females in Zimbabwe of age group 15–49 years old who answered questions on domestic violence. The ages have been selected because females between the ages of 15–49 are likely to be involved in intimate relationships and to experience intimate partner violence. All the females within all types of relationships; married, never married and those not married but living with a partner were included.
The ZDHS constitute a representative sample of 9 955 females in Zimbabwe (Zimbabwe National Statistics Agency and ICF International 2012). The study excluded females who did not respond to domestic violence questions and this led to a final sample of 7223 females who responded to questions pertaining to domes- tic violence. To ensure national representativeness of the data, weighting was done to correct oversampling and undersampling. The weighted sample of females was 7247. Data weighting refers to the tabulation of results characterised by re-balanc- ing the data to accurately reflect the population (Biemer and Christ 2008). The data were weighted using the recommended weights by the Demographic and health sur- vey (available from https:// www. dhspr ogram. com/ data/ Using- Datas ets- for- Analy sis.
cfm).
Measures
The dependent variable is intimate partner violence in two forms: physical and sex- ual violence. Physical violence refers to any of the following acts: pushed, slapped, punched with a fist, kicked, strangled, threatened by knife and arm twisted and all these variables had four categories. The respondents did not necessarily have to experience all the seven forms of physical violence to be labelled as having expe- rienced physical violence. Experiencing any of the seven acts of physical violence qualified the respondent to be categorised as having experienced physical violence.
Sexual violence refers to any of the following acts: ever been physically forced into unwanted sex and forced into unwanted sexual acts. As mentioned previously, the respondents did not have to report all acts of sexual violence to be categorised as having experienced sexual violence. The study explores whether or not increased female empowerment is associated with increased or decreased levels of intimate partner violence.
The main independent variable was female empowerment measured by participa- tion in decision-making, ownership of assets, education level and employment status.
The aforementioned variables were selected to measure female empowerment as they have been used in various studies on female economic empowerment (Castro et al.
2008; Dalal 2011; Kwagala et al. 2013; Fidan and Bui 2016). The females were asked questions about their background characteristics (education level, place of residence, religion, region, employment status) and on women empowerment (participation in household decision-making and who they owned assets with (land or house). Princi- pal component analysis was performed for all the four participation in decision-making variables to reduce the variables to a small set. Based on the factor scores, two cat- egories were generated, the first category was coded 0, and it represented not empow- ered females. The second category was coded 1, and it denoted empowered females.
Other independent variables were used to provide the background of the female par- ticipants are age, marital status, religion, type of place of residence and region. These variables were selected as control variables since various studies have shown that the variables are determinants of intimate partner violence. Thereafter, the variable owner- ship of assets was generated by creating an index where the variables, ownership of a house and ownership of land were summed and the scores used to rank the order of the variables. Assets are used to determine empowerment in women hence ownership of assets was utilised as a measure of female empowerment in the study (Resko 2007).
The scores ranged from 0 to 4; the score 0 represented those who had no assets and was coded as 0. The scores 1–2 depicted those who owned assets alone and were coded 1 and the scores 3–4 denoted females who jointly owned assets with their partners and it was coded as 2.
Data analysis
The univariate analysis involved describing the background characteristics of the respondents by using a series of frequencies and percentage distributions. The back- ground characteristics presented information of the respondents on the following:
educational level, employment status, participation in decision-making, ownership of assets, marital status, religion, age, place of residence and region. The bivariate analy- sis was done to present the profile of the sample according to the females’ experiences of intimate partner violence. It involved the cross-tabulation of the outcome variables, physical and sexual violence with all the independent variables using Pearson’s chi- square test. This was done to measure the association between each independent vari- able and the outcome variable. The multivariate analysis included a stepwise forward binary logistic regression, which is used to identify the association between measures of female empowerment and intimate partner violence while controlling for other inde- pendent variables.
Ethical consideration
To obtain data from the ZDHS, permission was sought through registering on the DHS website. As part of the registration procedure, brief information on why the
data set was being requested was required and this was provided. The permission to download the data set was granted.
Results Descriptive data
The females not in a union had higher reports of physical violence (37.94%) and sexual violence (16.53%) compared to the married females, of whom (27.88%) reported experiencing physical violence and (8.56%) reported experiencing sexual violence (Fig. 1).
Table 1 indicates the background characteristics of the females in the study. The majority of the married females lived in rural areas (66.27%), while (57.57%) who were not in a union lived in rural areas. In urban areas, the proportion of females not in a union was (42.45%) and in a union was (33.73%). Based on region, Harare province constituted a greater proportion of females in a union (15.58%) and the least reports less than (5%) were in Matabeleland North, Matabeleland South and Bulawayo. Of all the provinces, the highest reports of females not in a union were in Harare (18.83%) compared to other provinces.
The majority of the females in a union were of the apostolic sect religious denom- ination (47.45%), Pentecostal religion had (22.11%) and the protestant constituted (13.76%). Religious denominations such as Roman Catholic reported (5.56%) and
Fig. 1 Levels of female physical and sexual violence in Zimbabwe
Table 1 Background characteristics of the sample based on weighted frequencies from ZDHS 2015
Characteristics Females not in a
union (2306) Females in union (4917) Place of residence
Urban 42.45 33.73
Rural 57.57 66.27
Region
Manicaland 11.91 14.02
Mashonaland Central 6.17 10.70
Mashonaland East 9.30 10.42
Mashonaland West 10.11 13.13
Matabeleland North 4.92 4.49
Matabeleland South 6.28 3.43
Midlands 11.89 12.39
Masvingo 12.17 11.94
Harare 18.83 15.58
Bulawayo 8.40 3.90
Religion
Roman catholic 7.88 5.56
Protestant 19.51 13.76
Pentecostal 26.82 22.11
Apostolic sect 36.27 47.45
Other christian 4.61 4.10
None 4.90 7.02
Participation in decision-making
Not empowered 73.78
Empowered 26.22
educational level
Primary or no education 22.56 31.57
Secondary or higher education 77.44 68.43 Employment status
No 59.54 56.69
Yes 40.46 43.31
Ownership of assets
None 84.44 38.87
Alone 12.96 26.40
Jointly only 2.60 34.73
Age
15–19 41.38 5.90
20–24 15.03 17.53
25–29 11.65 21.52
30–34 8.81 22.92
35–39 7.87 15.26
40 + 15.26 16.86
other Christians had the least reports (4.10%) of females in a union. Females in a union who were from the non-religious denomination were (7.02%). A greater pro- portion of females in a union were not empowered (73.78%), while those empow- ered were (26.22%).
Table 1 further shows that females not in a union with secondary or higher edu- cation constituted a greater proportion (77.44%) than those in union (68.43%). The females not in a union had the least reports of primary or no education (22.56%) in relation to females in a union (31.57%). The reports of unemployed females not in a union were greater (59.54%) than those of females in a union (56.69%). The females in a union had higher reports of being employed (43.31%) in relation to females not in a union (40.46%). The greatest proportion of females with no assets was reported by females not in a union (84.44%), while those in a union constituted (38.87%). Females in a union who jointly owned assets with partners were (34.73%) and (26.40%) of the females owned the assets alone. The majority of the females aged 15–19 were not in a union (41.36%), while 5.90% from the same age group were in a union. A greater proportion of females aged between 25 and 29 and 30–34 were in a union.
Bivariate analysis
Table 2 shows the bivariate analysis of the association between female empower- ment with physical and sexual violence. The females in a union who are empowered had higher reports of physical violence (30.75%) and sexual violence (11.61%) than not empowered females who reported (26.79%) physical and (7.39%) sexual vio- lence. The females not in a union with primary or no education had high reports of both physical (40.07%) and sexual (18.73%) violence compared to females in a union who reported physical (29.83%) and sexual (8.99%). Females not in a union with secondary or higher education constituted a greater proportion of physical (37.01%) and sexual (15.58%) violence than females in a union who had (27.08%) reports of physical violence and (8.39%) sexual violence. Females who were not in a union and not employed had higher reports of physical (41.07%) and sexual (18.81%) violence compared to females in a union who reported (26.24%) physical and sexual (7.59%) violence. The females not in a union and currently employed constituted a greater proportion of reports on physical (36.17%) and sexual (15.25%) violence. For those in a union and currently employed the reports were (30.09%) for physical violence and (9.87%) for sexual violence.
The females not in a union and had no assets had very high reports of physical (40.23%) and sexual violence (17.26%) compared to females in a union who had no assets who reported (27.39%) physical violence and (7.61%) sexual violence.
Female in a union who jointly owned assets with partners had lower reports of phys- ical (27.99%) and sexual (9.01%) violence, while those not in a union and jointly owned assets had higher reports of physical (35.29%) and sexual (11.76%) violence.
The females who were in a union and owned assets alone had lower reports of physi- cal (28.56%) and sexual (9.57%) violence compared to those not in a union who reported (32.34%) physical and (15.32%) sexual violence.
Table 3 below presents the results of the crude or unadjusted odds ratios of physi- cal and sexual violence for females not in a union and those who are in a union.
Overall female empowerment was significantly associated with experiences of inti- mate partner violence among females in a union and insignificantly associated with experiences of intimate partner violence among females not in a union. The females in a union who are empowered were significantly more likely to experience physi- cal (Odds: 1.25) and sexual (Odds: 1.73) violence in relation to females who are not empowered. Females in a union were significantly less likely to experience physical violence. The females in a union who were employed were significantly more likely to experience physical (Odds: 1.21) and sexual violence (Odds: 1.26) in relation to those who are unemployed.
In Mashonaland West province, the females who were not in a union were signifi- cantly more likely to experience physical (Odds: 2.41) and sexual violence (Odds:
3.24) than females in Manicaland Province. In Mashonaland Central, the females in a union were significantly less likely to experience physical violence compared to females in Manicaland. In Masvingo, Matabeleland North and South Province, the females in a union were significantly less likely to experience physical and sexual violence compared to females in Manicaland province. The females in a union from Bulawayo were significantly less likely to experience sexual violence in relation to females in Manicaland.
Among females in a union, those from Pentecostal, Apostolic sect and none reli- gious denomination had significantly increased odds of experiencing physical vio- lence. Females not in a union aged from 20 to 40 and older were significantly less likely to experience sexual violence than females aged between 15 and 19. Among females in a union only those aged 40 and older were significantly less likely to experience physical violence.
Table 2 Percentage distribution of physical and sexual violence by background characteristics represent- ing female empowerment
Characteristics Females not in a union Females in a union
Physical Sexual Physical Sexual
Participation in decision-making
Not empowered 26.79 7.39
Empowered 30.75 11.61
Educational level
Primary or no education 40.07 18.73 29.83 8.99
Secondary or higher 37.01 15.58 27.08 8.39
Employment status
No 41.07 18.81 26.24 7.59
Yes 36.17 15.25 30.09 9.87
Ownership of assets
None 40.23 17.26 27.39 7.61
Alone only 32.34 15.32 28.56 9.57
Jointly only 35.29 11.76 27.99 9.01
Table 3 Unadjusted odds ratios for the binary logistic regression of the association between the back- ground characteristics with physical and sexual violence
Characteristics Females not in a union Females in a union All females Physical Sexual Physical Sexual Physical Sexual Participation in decision-making
Not empowered 1.00 1.00 1.00 1.00 1.00 1.00
Empowered 1.25*** 1.73*** 1.25*** 1.73***
Educational level
Primary or no education 1.00 1.00 1.00 1.00 1.00 1.00
Secondary or higher 0.96 0.82 0.86** 0.99 0.87** 0.94
Employment status
No 1.00 1.00 1.00 1.00 1.00 1.00
Yes 0.87 0.75 1.21*** 1.26** 1.20*** 1.23**
Ownership of assets
None 1.00 1.00 1.00 1.00 1.00 1.00
Alone only 0.73 0.79 1.00 1.26 0.91 1.03
Jointly only 0.71 0.63 1.01 1.17 0.89 0.88
Place of residence
Urban (RC) 1.00 1.00 1.00 1.00 1.00 1.00
Rural 0.88 0.94 1.10 1.06 0.66** 0.99
Region
Manicaland (RC) 1.00 1.00 1.00 1.00 1.00 1.00
Mashonaland Central 1.31 1.70 0.65*** 0.80 0.71** 0.89
Mashonaland East 1.55 1.28 1.22 0.67 1.27 0.76
Mashonaland West 2.41*** 3.24*** 1.11 0.99 1.23 1.24
Matabeleland North 0.92 0.33 0.37*** 0.45*** 0.44*** 0.43***
Matabeleland South 0.93 0.61 0.55*** 0.42*** 0.61*** 0.46***
Midlands 1.36 1.13 0.89 0.70 0.94 0.77
Masvingo 0.95 0.88 0.69** 0.59** 0.73** 0.64
Harare 1.40 1.55 0.80 0.79 0.89 0.94
Bulawayo 1.40 1.08 0.74 0.57** 0.85 0.68
Religion
Roman Catholic (RC) 1.00 1.00 1.00 1.00 1.00 1.00
Protestant 1.04 0.70 1.28 1.11 1.18 0.91
Pentecostal 0.98 0.87 1.51* 1.20 1.33 1.04
Apostolic sect 0.92 0.77 1.72* 1.61 1.45** 1.21
Other christian 1.25 0.97 1.40 0.88 1.27 0.81
None 0.49 0.57 2.50* 1.94 1.84** 1.34
Age 15–19 (RC)
20–24 0.75 0.18*** 1.08 0.99 1.01 0.67
25–29 0.61 0.19*** 1.22 0.90 1.11 0.63**
30–34 0.42** 0.21*** 0.98 0.73 0.87 0.54***
35–39 0.50 0.24*** 0.83 1.01 0.80 0.76
40 + 0.29*** 0.15*** 0.65** 0.66 0.62*** 0.53***
**p < 0.05
***p < 0.01
Multivariate analysis
Table 4 presents the results of the adjusted odds ratios of females not in a union and in a union regarding experiences of physical and sexual violence. Similar to the unadjusted odds ratio, female empowerment was significantly associated with experiences of intimate partner violence, while for females not in a union it was not. Females who are in a union who are empowered were significantly more likely to experience physical and sexual violence than those not empowered.
Interestingly, the females in a union who were employed were significantly more likely to experience physical violence (Odds: 1.23) than unemployed females.
The females not in a union who did not belong to any religious denomina- tion were significantly less likely to experience sexual violence (Odds: 0.37) than Catholic females. The females in a union from the apostolic sect, other Christian, and those who are not of any religious group were significantly more likely to experience physical violence and insignificantly more likely to experience sex- ual violence than Catholic females. Females in Mashonaland West not in a union were more likely to experience physical (Odds: 2.74) and sexual violence (Odds:
4.11) than their counterparts in Manicaland province. Similar to the unadjusted odds, the adjusted odds results indicated that females not in a union aged from 20 to 40 and older were significantly less likely to experience sexual violence.
Discussion
This study revealed that experiences of physical violence were higher than those of sexual violence. The empowered females in this study referred to females who made decisions on their own, while not empowered females referred to those who did not make decisions. Both the unadjusted and adjusted odds ratios showed that the empowered females in a union were at an increased risk of experiencing physical and sexual violence in relation to those who are not empowered. These findings are in support of literature which argues that females, who participate in decision-making were more vulnerable to intimate partner violence (Rahman et al. 2011; Wekwete et al. 2014). Some scholars argue that the response of using violence by males is a sign of feeling powerlessness when females participate in decision-making and the males restore their power by using violence (Rahman et al. 2011). It can be argued that being an empowered female within a union threatens males leading to the use of violence as a means to regain power and control over females. This is in support of a study in Burkina Faso which identi- fied the indicators of female empowerment as responsible for triggering domestic violence among females currently married or living with a partner (Pambe et al.
2014).
Based on the findings being not empowered is protective against intimate part- ner violence. This could be possibly attributed to the tendency of females who are not empowered to accept decisions made by their partners without questioning or opposing the decisions. It can be argued that females who are not empowered are
Table 4 Adjusted odds ratios for the binary logistic regression of the association between all the inde- pendent variables with physical and sexual violence
Characteristics Females not in a union Females in a union All females Physical Sexual Physical Sexual Physical Sexual Participation in decision-making
Not empowered 1.00 1.00
Empowered 1.39*** 1.82***
Educational level
Primary or no education 1.00 1.00 1.00 1.00 1.00 1.00
Secondary or higher 0.83 0.75 0.87 1.12 0.87 1.12
Employment status
No 1.00 1.00 1.00 1.00 1.00 1.00
Yes 0.86 0.76 1.23*** 1.29 1.23*** 1.28
Ownership of assets
None 1.00 1.00 1.00 1.00 1.00 1.00
Alone only 0.95 0.93 1.07 1.29 1.06 1.29
Jointly only 1.04 0.93 1.15 1.30 1.15 1.30
Place of residence
Urban (RC) 1.00 1.00 1.00 1.00 1.00 1.00
Rural 0.95 1.01 1.06 1.03 1.06 1.03
Region
Manicaland (RC) 1.00 1.00 1.00 1.00 1.00 1.00
Mashonaland Central 1.36 1.86 0.58*** 0.72 0.58*** 0.72
Mashonaland East 1.71 1.37 1.28 0.67 1.28 0.67
Mashonaland West 2.74*** 4.11*** 1.08 0.93 1.08 0.93
Matabeleland North 0.89 0.34 0.34*** 0.43*** 0.34*** 0.43***
Matabeleland South 1.02 0.79 0.54*** 0.43*** 0.54*** 0.43***
Midlands 1.30 1.27 0.88 0.70 0.88 0.70
Masvingo 0.99 0.95 0.69** 0.60 0.69** 0.60
Harare 1.50 1.91 0.88 0.84 0.88 0.84
Bulawayo 1.37 1.25 0.84 0.62 0.84 0.62
Religion
Roman catholic 1.00 1.00 1.00 1.00 1.00 1.00
Protestant 1.05 0.81 1.25 1.14 1.25 1.14
Pentecostal 0.90 0.93 1.41 1.20 1.41 1.20
Apostolic sect 0.89 0.72 1.53** 1.60 1.53** 1.60
Other christian 1.20 1.01 1.64** 1.04 1.64** 1.04
None 0.37** 0.48 2.25*** 2.09 2.25*** 2.09
Age
15–19 (RC) 1.00 1.00 1.00 1.00 1.00 1.00
20–24 0.74 0.18*** 1.11 0.96 1.11 0.96
25–29 0.65 0.21*** 1.20 0.81 1.20 0.81
30–34 0.41** 0.20*** 0.96 0.65 0.96 0.65
35–39 0.46 0.22*** 0.80 0.89 0.80 0.89
40 + 0.30*** 0.15*** 0.61*** 0.57 0.61*** 0.57
**p < 0.05
***p < 0.01
generally tolerant of their partners and are prepared to save their marriages and relationships since some would be financially dependent on the males. However, in some contexts, females who participate in decision-making have reduced reports on violence, evident in a study conducted in Zambia which illustrated that females in a union who participated in making decisions concerning health issues had a lower likelihood of experiencing intimate partner violence compared to females who did not participate in decision-making (Okenwa and Lawoko 2010).
The findings indicate that females who embody characteristics of empowerment are the same group who are at risk of physical and sexual violence. These sugges- tions are supported by some scholars who contend that empowerment of females does not spare them from risks of intimate partner violence unless their male part- ners are involved in supporting the movement to shift the power dynamics (Rah- man et al. 2011; Green et al. 2015a, b). This can be attributed to males’ failure to admit that females can also be breadwinners and dominant role players in the home.
Instead, to males, the empowerment of females diminishes their role of being bread- winners and to dominate over females.
It is believed that education empowers females to be able to have a say or partici- pate in decision-making (Lawoko 2006). Some studies state that it reduces violence and protect females from intimate partner violence (Okenwa and Lawoko 2010;
Wekwete et al. 2014; Fidan and Bui 2016). This is in line with the findings of this study which showed that both females who are in a union and not with secondary or higher education were less likely to experience physical and sexual violence in relation to those who had primary or no education. However, both the unadjusted and adjusted odds ratios indicated that the relationship was statistically insignificant.
However, some studies argue that the empowerment of females through improved education has served as a tool for accelerating violence instead of reducing it (Jew- kes, 2002; Castro et al. 2008; Abramsky et al. 2011; Zacarias et al. 2012).
The employment of females has been recommended by different development poli- cies as a means to facilitate empowerment (Bedi et al. 2011). Some scholars state that females who work while their partners were not, had high risks of experiencing vio- lence and including in relationships where both were not working (Abramsky et al.
2011). The findings of this study show that the females in a union who were employed were significantly more likely to experience physical and sexual violence in relation to those who are unemployed. Meanwhile, the females who were not in a union and not employed were insignificantly less likely to experience physical and sexual violence.
The adjusted odds ratios indicated that the females in a union who were employed were significantly more likely to experience physical violence and insignificantly more likely to experience sexual violence than unemployed females.
These findings are in support of studies conducted in Mexico, India and Bangladesh which discovered that working females in a union were at huge risks of intimate partner violence (Castro et al. 2008; Dalal 2011; Rahman et al. 2011). This can be a result of females who are employed developing a sense of autonomy or independence in mak- ing household decisions which might undermine the role of their partners in making decisions. This also accentuates females’ productive roles in the household economy making some of the females’ key breadwinners in the family. This is because working
might destruct fulfilment of domestic roles creating tensions among couples and part- ner’s jealousy was identified as part of the reasons (Castro et al. 2008).
The unadjusted and adjusted odds ratios indicated that there was no relationship between ownership of assets and physical or sexual violence within females not in a union and those in a union. However, some scholars argue that the ownership of assets such as a house or land by females offers security in times of conflict in relationships and makes them have the power to decide to end abusive relationships (Bedi et al.
2011; Oduro et al. 2015). A study conducted in Zimbabwe highlighted that females who jointly owned land with partners had a higher likelihood of experiencing physical violence compared to those who did not own land (Wekwete et al. 2014). The same study found no association between ownership of house or land with sexual violence and these findings resembles those of the current study (Wekwete et al. 2014). Despite ownership of assets being identified as not associated with physical and sexual vio- lence, other studies point out that ownership of assets in relation to being employed among females provides them with security and autonomy to leave abusive relation- ships (Panda and Agarwal 2005).
Conclusion
The study examined the association between female empowerment and intimate partner violence among females in Zimbabwe. It distinguished experiences of intimate partner violence between females in a union and those who are not.
Based on the findings, being empowered and in a union does not protect females from intimate partner violence. The findings indicate that females who are in marital union and socially and economically empowered were at a greater risk of experiencing intimate partner violence than females not in a union.
This could be because the women in a union live with their husbands mak- ing them more vulnerable to experiencing intimate partner violence in cases where there is tension between them and their husbands. As shown in the findings when females in a union are employed, have secondary or higher education and own assets alone or jointly with a partner they are at an increased likelihood of experiencing intimate partner violence. Meanwhile, not being in a union exposes females to a lower likelihood of experiencing intimate partner violence. These findings are synonymous with other studies conducted in Latin America, Asia and Sub-Saharan Africa.
Looking at the findings of this study, it can be argued that a lot needs to be done besides the empowering of females to address intimate partner violence.
As shown in this study, the empowerment of females is not enough to address females’ vulnerability to violence. It can be argued that societal norms and val- ues place barriers on females such that rising against such barriers through the empowerment of females can be portrayed as going against set norms and values in some societies. The societal norms and values define the gender roles and what is expected of females and this is installed in the females through the socialisation process. According to some societies, females should be submissive, tolerant and honour their husbands. Such notions can be used to explain why the findings of
this study indicated that not empowered females in a union had a reduced risk of intimate partner violence compared to their counterparts, the empowered females.
This is because not empowered females tend to be submissive to their partners as they look up to them. Above all, gender inequalities in Zimbabwe are rooted in the patriarchal society which portrays females as subjects to be controlled by males.
Strength and limitations
The Zimbabwe and Demographic Health Survey is nationally representative of all the women in the country. However, the study is without limitations, the cross- sectional nature of the study does not allow to infer causality. A causal relation- ship between female empowerment and intimate partner violence cannot be deter- mined by the use of binary logistic regression. Moreover, the study only focused on two forms of intimate partner violence; physical and sexual violence, without considering other forms of violence. It could be possible that sexual violence was underreported due to the tendency of social desirability bias where the females would not be comfortable to share sensitive information with strangers. Social desirability affects surveys since respondents might respond in a socially accepta- ble manner instead of being authentic (Kaminska and Foulsham 2013). The under reports of sexual violence in this study could have been because of the traditional culture in Zimbabwe which instils beliefs among females that sexual abuse by their partners is not violence (Tom et al. 2013; Chireshe 2015). One of the proxies of female empowerment used in the study participation in decision-making was not asked to females not in a union in the Zimbabwe Demographic and Health Survey. This could have affected the results since females not in a union had three variables (educational level, ownership of assets, employment status) being used to measure empowerment, while females in a union used four variables (females’
participation in decision-making, educational level, ownership of assets, employ- ment status). It is a limitation of the DHS to have asked married women only questions on women empowerment.
Recommendations
The very nature of empowerment employed in this study does not fully address what constitutes power itself, being able to participate in decision-making and being edu- cated does not mean that one is empowered, and this further explains why educated females continue to be battered. It is recommended for further studies to assess the issue of empowerment among females and to examine whether females feel empow- ered and document the reasons or attributes that make them acknowledge that they are empowered. There is a need for further discussion among intimate partner violence experts and researchers to understand how empowering females needs to happen in cer- tain contexts so that the desired outcome which is to reduce intimate partner violence
can be achieved before programmes are put into place. It is recommended that empow- erment of females should also involve education of both males and females on the social ills of the patriarchal society which inhibits females to attain power in the same way as the males.
Data availability The data sets generated and/or analysed during the current study are available on the Demographic Health Survey website [https:// dhspr ogram. com/ data/ avail able- datas ets. cfm].
Declarations
Conflict of interest The authors declare that there are no conflict of interest.
References
Abramsky T, Watts CH, Garcia-Moreno C, Devries K, Kiss L, Ellsberg M, Jansen HA, Heise L (2011) What factors are associated with recent intimate partner violence? findings from the WHO multi- country study on women’s health and domestic violence 11(109):1–17
Bedi AS, Chhachhi A, Bhattacharyya M (2011) Marital violence and women’s employment and prop- erty status: evidence from North Indian villages. World Dev 39(9):1976–1689
Biemer PP, Christ S (2008) Weighting survey data. International handbook of survey methodology Castro R, Casique I, Brindis CD (2008) Empowerment and physical violence throughout women’s
reproductive life in Mexico. Viol against Women 14(6):655–677
Chireshe E (2015) Article christian women’s experiences of domestic violence in Zimbabwe. Affilia 30(3):380–394
Dalal K (2011) ‘Does economic empowerment protect women from intimate partner violence? J Injury Viol Res 3(1):35–44
Fidan A, Bui HN (2016) Intimate partner violence against women in Zimbabwe. Violence against Women 22(9):1075–1096
García-Moreno C, Pallitto C, Devries K, Stöckl H, Watts C, Abrahams N (2013) Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence. World Health Organization
Green EP, Blattman C, Jamison J, Annan J (2015a) Women’s entrepreneurship and intimate partner violence: a cluster randomized trial of microenterprise assistance and partner participation in post-conflict Uganda. Soc Sci Med 133:177–188
Green EP, Christopher B, Jamison J, Jeannie A (2015) Women’s entrepreneurship and intimate partner violence: a cluster randomized trial of microenterprise assistance and partner participation in post-conflict Uganda. 133:177–188
Hove K, Gwazane M (2011) A study to determine factors associated with domestic violence among concordant and discordant couples in Zimbabwe. Spec Issue Behav Soc Sci 1(7):156–162 Jewkes R (2002) Intimate partner violence: causes and prevention. Lancet 359(9315):1423–1429 Kabeer N (1999) Resources, agency, achievements: Reflections on the measurement of women’s
empowerment. Dev Chang 30(3):435–464
Kabeer N (2005) Gender equality and women’s empowerment: a critical analysis of the third millen- nium development goal 1. Gend Dev 13(1):13–24
Kaminska O, Foulsham T (2013) Understanding sources of social desirability bias in different modes:
Evidence from eye-tracking (No. 2013–04). ISER Working Paper Series
Kwagala B, Wandera SO, Ndugga P, Kabagenyi A (2013) Empowerment, partner’s behaviours and intimate partner physical violence among married women in Uganda. BMC Public Health 13(1):1
Lawoko S (2006) Factors associated with attitudes toward intimate partner violence: a study of women in Zambia 21(5):645–656
Mahmud S, Shah NM, Becker S (2012) Measurement of women’s empowerment in rural Bangladesh.
World Dev 40(3):610–619
Mashiri L (2013) Conceptualisation of gender based violence in Zimbabwe. Int J Humanit Soc Sci 3(15):94–103
Mukhopadhyay M (2007) Gender justice, citizenship and development, IDRC
Nyamayemombe C, Mishra V, Rusakaniko S, Benedikt C, Gwazane M, Mukweza P (2010) The associa- tion between violence against women and HIV: evidence from a national population-based survey in Zimbabwe, Calverton, Maryland, USA: ICF Macro
Oduro AD, Deere CD, Catanzarite ZB (2015) Women’s wealth and intimate partner violence: insights from Ecuador and Ghana. Fem Econ 21(2):1–29
Okenwa L, Lawoko S (2010) Social indicators and physical abuse of women by intimate partners: a study of women in Zambia. Viol Vict 25(2):278–288
Osirim MJ (2003) Crisis in the state and the family: violence against women in Zimbabwe. Afr Stud Q 7(2):153–169
Pambè MW, Gnoumou B, Kaboré I (2014) Relationship between women’s socioeconomic status and empowerment in Burkina Faso: A focus on participation in decision-making and experience of domestic violence. Afr Popul Stud 28:1146–1156
Panda P, Agarwal B (2005) Marital violence, human development and women’s property status in India.
World Dev 33(5):823–850
Presser H, Sen G (2000) Women’s empowerment and demographic processes: moving beyond Cairo.
Oxford University Press
Rahman M, Hoque MA, Makinoda S (2011) Intimate partner violence against women: is women empow- erment a reducing factor? A study from a national Bangladeshi sample. J Fam Viol 26(5):411–420 Resko SM (2007) Intimate partner violence against women: exploring intersections of race, class and
gender–PhD thesis, The Ohio State University
Stewart S (1995) Working with a radical agenda the Musasa project, Zimbabwe. Gender Dev 3(1):30–35 Tom T, Musingafi MCC, Campus MR (2013) Domestic violence in urban areas in Zimbabwe: a case
study of Glen Norah (Harare). Res Humanit Soc Sci 3(3):45–50
United Nations Women, 2014. New research paints poignant picture of intimate partner violence in Zim- babwe, Headquarters
Wekwete NN, Sanhokwe H, Murenjekwa W, Takavarasha F, Madzingira N (2014) The association between spousal gender based violence and women’s empowerment among currently married women aged 15–49 in Zimbabwe: evidence from the 2010–11 Zimbabwe demographic and health survey. Afr Popul Stud 28(3):1413–1427
Zacarias AE, Macassa G, Svanstrom L, Soares JJ, Antai D (2012) Intimate partner violence against women in Maputo city, Mozambique. BMC Int Health Human Rights 12(35):1–16
Zimbabwe National Statistics Agency and ICF International (2012) Zimbabwe Demographic and Health Survey
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
