Sage Journals: Discover world-class research

Abstract

Intimate partner violence (IPV) against women and girls is recognized as a human rights problem with little understanding of its ripple effects on the workplace. While the literature has attempted to understand the impacts of IPV on the workplace in high-income countries, relatively, literature on low and middle-income countries is very scant. Using primary survey data collected from 16,921 workers in 257 businesses in Ghana, South Sudan, Bolivia, and Paraguay, this is the largest ever study of its kind that highlights the invisible costs businesses incur due to IPV experienced by female employees. IPV’s economic impact on labor productivity is based on tardiness, absenteeism, and presenteeism. Unlike the common perception that violence affects only the survivors, this study also estimates the effects on perpetrators. The results show that IPV exists in all the businesses surveyed, leading to enormous productivity losses due to both the experience and perpetration of IPV.

Keywords

intimate partner violence productivity loss for businesses presenteeism costs of perpetration economic costs

Introduction

Intimate partner violence (IPV) against women and girls is a pandemic affecting every country in the world, and particularly in the current context of COVID-19. Globally, for every 10 ever partnered women, 3 women have experienced physical and/or sexual violence by an intimate partner in their lifetime (World Health Organization [WHO], 2013); a figure that increases significantly if psychological and economic violence is included. IPV is widely considered an issue of conflict, whose impacts remain behind the closed doors of the home. Its impact, however, more often than not seeps into public spaces in a variety of ways (Martinez, 2015). For example, preventing or interrupting women’s employment is a strategy that abusive partners often use (Gupta et al., 2018). Violent experiences in the home can also intensify the workplace stress of employees.

The workplace is a particularly important space to consider given women’s participation rate has steadily increased globally in the context of international policy commitments to women’s empowerment. It is thus important to understand if IPV impacts businesses’ productivity via its effects on the physical and mental health of the woman survivor. For businesses, rates of absenteeism, tardiness, and work distraction (or presenteeism) among employees are seen as the three dimensions of productivity loss, which is a critical concern given the impact of declining productivity on their bottom line. Equally, the impacts of IPV at the workplace are important from a social reputation perspective as distressed employees can undermine client relationships, which represent the social capital of a business. Based on primary survey data collected from Ghana, South Sudan, Bolivia, and Paraguay, this study is the largest of its kind that not only establishes but also quantifies, the productivity loss for both survivors and perpetrators of IPV.

In the case of high-income countries, research has consistently established the negative work consequences of IPV. For the US, numerous studies have recognized IPV related work impacts including unemployment, job instability, missed workdays, tardiness, work disruption, reduced job performance (Anderson et al., 2014; Blodgett & Lanigan, 2018; Borchers et al., 2016; Crowne et al., 2011; Kimerling et al., 2009; Kulkarni & Ross, 2016; Logan et al., 2007; Pollack, Cummiskey, et al., 2010; Pollack, McKay, et al., 2010; Potter & Banyard, 2011). IPV and similar work-related impacts have also been established in other developed countries, including Norway (Alsaker et al., 2016), Australia (McFerran, 2011), New Zealand (Rayner-Thomas, 2013), Canada (Wathen et al., 2015).

On the other hand, research on work-related consequences of IPV in low and middle-income countries is relatively scant. Amongst the few studies that exist, Naved et al. (2018) document the overlaps between IPV and violence in the workplace for female garment workers in Bangladesh. Based on a survey in Mexico City, work-related disruption due to IPV is also reported by 40.6% of low-income women working (Gupta et al., 2018). For Vietnam, Nata Duvvury et al. (2012) estimated that an average of 5.5 workdays per incident of violence are lost as a consequence of IPV. In an important study on Mongolia, Chadha et al. (2020) estimated significant economic costs of domestic violence, including foregone income and productivity loss. In a Papa New Guinea study of women and men businesses employees, Darko et al. (2015) find, on average, each staff member experiencing gender-based violence loses 11.1 days of work as a result of the violence. In two recent studies, the costs to businesses of both IPV and sexual violence in the form of time spent by colleagues supporting survivors, survivors being absent, survivors arriving late to work and being less productive have been estimated for Fiji and the Solomon Islands (IFC, 2019a, 2019b). More recently, Chadha et al. (2022) investigated management’s outlook on economic costs of violence against women and girls.

The literature suggests that the deteriorating physical and mental health impact on survivors is one of the prime pathways to IPV’s work-related impacts (Meisel et al., 2003; Parvin et al., 2018; Wathen et al., 2018). Survivors in Canada with any lifetime IPV experience report considerably poorer general health, mental health, and quality of life (Wathen et al., 2018). Survivors of both recent IPV and IPV experienced in the last 12 months have the poorest health (Wathen et al., 2018). The negative association between the need for services and posttraumatic stress resulting from abuse coupled with working at least 32 hours a week has also been established (Meisel et al., 2003). A relationship has also been uncovered between the need for services and working fewer weeks in a year, losing jobs during the year and having a lower income due to IPV (Meisel et al., 2003). Similarly, Parvin et al. (2018) also find IPV leads to increased work-related stress and depression.

To address IPV against women and girls, researchers have also attempted to evaluate the health status of male perpetrators (Askeland & Heir, 2014; Machisa & Shamu, 2018; Shorey et al., 2012). In Norway, approximately 71% of male perpetrators fulfilled the diagnostic criteria for at least one ongoing psychiatric disorder (Askeland & Heir, 2014). Three categories of disorders were revealed as most prevalent: anxiety disorders including post-traumatic stress disorder (PTSD), depressive disorders, and alcohol/substance abuse (Askeland & Heir, 2014). Machisa and Shamu (2018) have similarly established an association between depressive symptoms and lifetime IPV perpetration in Zimbabwe. IPV perpetration in the last 12 months is also associated with PTSD, binge drinking, and power in a sexual relationship (Machisa & Shamu, 2018). In addition to PTSD and depression among male perpetrators, Shorey et al. (2012) find the prevalence of generalized anxiety disorder, social phobia, panic disorder, and alcohol and drug disorders in male perpetrators of IPV. In an important literature review, Sesar et al. (2018) discuss additional studies that have associated various mental disorders with perpetrators of violence.

As with the case of IPV survivors, an effort has been made in recent times to establish not only the mental effects on perpetrators, but also the productivity impacts at their workplaces (Mankowski et al., 2013; Rothman & Corso, 2008). For example, Mankowski et al. (2013) revaled that men in the extreme abuse cluster of work-related IPV are roughly four times more likely to experience significant impacts on their job performance than those in the low-level tactics cluster (Mankowski et al., 2013). Similarly, Rothman and Corso (2008), based on a small sample (N = 61), found a positive association between greater propensity for abusiveness and missing work, as well as experiencing lower work productivity, while controlling for income, marital status, level of education, part-time versus full-time employment status, and age. Problems with concentration at work, with an adverse effect on perpetrators’ performance and sometimes also leading to workplace accidents, have further been reported by Schmidt and Barnett (2012).

However, there is a dearth of studies that move beyond a single sector or single firm focus, to empirically estimate productivity loss including understanding the level of presenteeism and exploring the productivity impacts for both survivors and perpetrators of IPV in low and middle-income countries. This study makes four major crucial contributions to the literature. Firstly, while the literature has established the negative work-related impacts of IPV, this study is the largest of its kind that not only establishes but, unlike most previous studies, also quantifies the productivity loss. Secondly, in the limited studies that have estimated productivity loss, the focus has predominantly been on absenteeism and tardiness, with not much attention given to presenteeism. This study by quantifying the estimate of presenteeism provides a rationale for businesses to incorporate IPV as one of the major factors affecting their employees’ performance. Thirdly, IPV has traditionally been assumed to have an impact on survivors alone, with very limited studies exploring the impacts on perpetrators, particularly in low and middle-income countries. This study provides empirical evidence quantifying the productivity impacts for survivors and perpetrators. Importantly, the study establishes the external validity of these impacts across different cultural contexts across low and middle-income countries.

In this paper, we examine more closely the level of productivity loss due to IPV among women and men employed in the formal private sector in Ghana, South Sudan, Bolivia, and Paraguay to understand the implications for businesses. These four countries fall within the World Bank’s classification of low to middle-income countries. Paraguay stands apart as upper-middle-income country with a higher score of human development and a relatively lower prevalence of IPV (see Table 1). All the countries have a sizable amount of workers in informal employment, with 64% in Paraguay to almost 89% in Ghana.

Table 1.

Economic and Gender Context of the Studied Countries.

	Ghana	South Sudan	Bolivia	Paraguay
Population	30 million (2018)	11 million (2018)	11 million (2017)	7 million (2017)
Economic × parameters
GDP	GDP: 66 Billion USD (2018)	GDP: 3 Billion USD (2016)	38 Billion USD (2018)	30 Billion USD (2018)
Income level	Lower middle-income country	Amongst the poorest countries in the world; conflict-affected since December 2013	Lower middle-income country	Upper middle-income country
Human Development Index (HDI) Rank (2017)	140; medium human development category	187; low human development category	118; medium human development category	110; high human development category
Male and female HDI values	male HDI-0.619; female HDI-0.563	male HDI-0.422; female HDI-0.348	male HDI-0.716; female HDI-0.665	male HDI-0.710; female HDI-0.690
% of informal employment	89% (Ghana Statistical Service, 2015)	80% (World Bank, 2011)	87% (INE, 2019)	64% (International Labour Organization [ILO], 2014)
Gender-based violence
Prevalence of IPV	2 out of every 10 Ghanian women in a relationship have experienced physical and/or sexual IPV in the last 12 months (Asante et al., 2019)	5 out of every 10 South Sudanese women in a relationship have experienced physical and/or sexual IPV in the last 12 months (Elmusharaf et al., 2019)	3 out of every 10 Bolivian women in a relationship have experienced physical and/or sexual IPV in the last 12 months (Bott et al., 2019)	1 out of every 10 Paraguayan women in a relationship have experienced physical and/or sexual IPV in the last 12 months (Bott et al., 2019)
Impacts of IPV	The national loss in productivity among all working women in Ghana was approximately 65 million days annually, equivalent to 4.5% of employed women in effect not working (Asante et al., 2019)	The overall productivity loss due to any form of violence experienced by working women comes to 8.5 million lost days of work in the year or equivalent to 6% of the total employed women not working in the regions of South Sudan covered in this study (Elmusharaf et al., 2019)	Not estimated	The overall productivity loss due to any form of violence experienced by working women comes to 3.7 million lost days of work in the year or equivalent to 2.5% of the total employed women not working in the country (Vara-Horna, 2018)

Source. United Nations Development Programme, 2017; World Bank, 2017.

Methods

Selection of Countries

This paper draws on two sets of studies undertaken in low and middle-income countries using comparable sampling strategies and questionnaires. The paper focuses on two countries from Africa (Ghana and South Sudan) and two countries from South America (Bolivia and Paraguay). As such, the analysis enables us to understand and validate the costing model in countries with dissimilar cultural contexts and violence rates, Ghana and South Sudan were part of a larger project funded by the UK Department for International Development. Bolivia and Paraguay were part of the COMVOMUJER program of studies on the business costs of violence against women in South America. Bolivia and Paraguay were selected for the current analysis as the research was undertaken in the same period of time as the two African studies.

Sampling Strategy

A purposive quota sample was used within selected cities of each country to gather information from business employees about IPV and its impact on their work. As most formal business sectors are based in the cities, national researchers selected the capital city of each country as well as other cities which had a high prevalence of formal employment. As the sampling was a purposive quota, the sampling was not reflective of the entire cities’ formal employment or population. National researchers drafted a sampling strategy to include businesses in the formal market. In Ghana, business sectors were chosen based on their contribution to GDP, where the overall service industry contributed 49.6% of the GDP—manufacturing 28.4% and agriculture 22% in 2014. Accordingly, two-thirds (65) of businesses selected were from the services sector and one-third (35 businesses) were from industry. Within the services sector, businesses were selected from five sub-sectors which accounted for 69% of the overall contribution to GDP from this sector—trade, hotels and restaurants, transport and storage, financial services, and real estate. For the industry sector, the selected subsectors were construction, mining and quarrying, and manufacturing. Together, these three subsectors accounted for almost all of the industrial contribution to GDP in 2014 (25.6%). In South Sudan, GDP data was not available due to the conflict. Based on the African Enterprise Survey conducted in 2014 for the World Bank by Tango Consult and in consultation with the National Bureau of Statistics, sectors most relevant to the economy of South Sudan were identified. Table 2 below provides the cities, business sectors, and sample sources used to select businesses for participation.

Table 2.

Cities and Sectors Covered in the Surveys by Country.

Country	Cities (% of population)	Business sectors	Sample source
Ghana	Accra and Kumas (19%)	Trade; Hotels and Restaurants; Transport and storage; Financial Services; Real Estate; Manufacturing; Mining and Quarrying; Construction	Association of Ghana Industries; Business directories (Ghanayello.com/ghanaweb.com); Dun and Bradstreet
South Sudan	Juba and Yei (7%)	Manufacturing; Retail and Wholesale; Hotels and Restaurants; Communication and IT; Transport and Logistics; Construction; Financial Services	Africa Enterprise Survey
Bolivia	La Paz, Cochabamba, and Santa Cruz (34%)	Retail and Wholesale, manufacturing; financial services, Communication and IT, Transport and Logistics	Bolivian National Chamber of Commerce, German-Bolivian Chamber of Commerce AHK.
Paraguay	Asuncion, Central, Alto Parana, Itapua, and Ñeembucu (58.9%)	Retail and Wholesale, manufacturing; financial services, Communication and IT, Transport and Logistics; Agricultural and farming	Department of Statistics, Surveys, and Censuses

Source. Authors’ own.

In Bolivia and Paraguay, a similar method of identifying key sectors in terms of GDP contribution was used. In Paraguay, the sample design focused on medium and large companies in the major cities across the main regions as follows: Asunción (43.9%), Alto (28%), Paraná (12.4%), Itapúa (8.8%), and Ñeembucú (6.9%). The selected businesses were from the retail (24%), manufacturing (32%), and services (40%) sectors. In Bolivia, the sample was limited to businesses headquartered in the three core cities. In 2014, these cities (La Paz, Cochabamba, and Santa Cruz) contributed 67.2% of the GDP of the national economy and accounted for 84.3% of the economically active employed population. The sample thus comprised five business sectors—manufacturing (47.3%), banking (24.5%), telecommunications (11%), services (9.6%), and companies dedicated to retail and wholesale (7.7%).

Questionnaire

A Survey of women and men employees based on a questionnaire used in the 2013 Peru study undertaken by the University of San Martin de Porres (USMP) on the financial consequences of IPV was conducted in the two sets of country studies. For male employees, the self-completion questionnaire covered perpetration of violence, and in Ghana and South Sudan, it also included experience(s) of IPV. For female employees, the questionnaire covered experience(s) of violence.

For the two African studies, the National University of Ireland, Galway (NUIG) led the questionnaire development, supported by Ipsos MORI. The questions were translated from Spanish to English and then adapted to the respective country’s contexts. As a result, some new questions were added, such as those relating to attitudes to working in the organisation. Similarly, the ComVoMujer program of The Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) led the adaption of the original questionnaire for Bolivia and Paraguay, supported by USMP.

Overall, the surveys used in Africa and South America were similar with only minor differences regarding regional language and context. The later sections on “Estimation of Intimate Partner Violence” and “Estimation of Productivity Loss” provide, in detail, the minor adjustments made to the surveys for each region.

Data Collection

Primary survey data was collected from 100 businesses using anonymous and confidential self-completion questionnaires across all four countries. The selected businesses were communicated by telephone, letter, and in-person to seek their participation in the research. The questionnaire was introduced as a crucial survey about employee quality of life and wellbeing.

Among those businesses that agreed to take part, a day and time were agreed for a member of the field research team to visit their sites to provide and later collect the self-completed questionnaire. Upon arrival, interviewers asked to talk to the head/representative of the business who had initially consented to the survey being undertaken. Employees were then randomly selected for participation, with all employees of the business eligible to complete the questionnaire. The nature and purpose of the survey were described to each employee who agreed to participate.

Participants were asked to complete the questionnaire in private and to hand it back to an interviewer once completed. However, in a few cases, interviewers were not allowed to remain on-site for the entire workday, having to return to take completed questionnaires. Where interviewers could not stay onsite, employees were provided with an envelope and asked to seal their questionnaire upon completion. A member of the interviewing team then returned to collect completed questionnaires at the agreed time, which was usually at the end of the business day. Apart from some key demographics (age and length of service for analysis purposes), the survey did not collect any personal identifying information such as name, address, or contact details. This helped to ensure confidentiality.

Ethical approval was also granted at the overall level by the Research Ethics Committee at the National University of Ireland, Galway. Additionally, in-country approval was granted by the University of Ghana (Ethics Committee for the Humanities) and by the National Bureau of Statistics, South Sudan. Both the Bolivia and Paraguay studies underwent internal review by the University St. Martin des Porres and GIZ to ensure alignment with standard WHO ethical and safety guidelines for research on violence against women (WHO, 1999).

It was not possible to weigh the data collected via the business surveys. This is because the information on the business universe in each country was not available, comprehensive or recent enough to provide accurate data that could be used for this purpose. Furthermore, given the purposive/quota sampling strategy used and the small number of responses obtained, the results from the surveys should not be regarded as representative of the target population. Rather, the results are indicative of the situation in businesses in these countries.

A total of 16,921 employees, comprising 7,062 women and 9,859 men in 257 private companies, completed an employee questionnaire. The response rate for businesses contacted in Ghana and South Sudan was 27% and 23%, respectively, while in Bolivia and Paraguay, the response rate was 31% and 25% respectively. In terms of the response rate of employees, it was around 20% in the two African countries whereas the response rate was higher in Latin America with 48% and 70% in Bolivia and Paraguay respectively.

Estimation of Intimate Partner Violence

Intimate Partner Violence (IPV) is defined as at least one episode of any form of violence: economic, psychological, physical, or economic in the last 12 months. The questions used in the survey to measure IPV have been modified from the Conflict Tactics Scale 2 (Straus, 2007), 2005 WHO questionnaire used in the first cross-country study of IPV, and the National Intimate Violence Against Sexual Violence Survey by the Center for Disease Control and Prevention (Breiding et al., 2015). In Bolivia and Paraguay, the questionnaires covered the following types of IPV:

Economic violence (taking their income)

Psychological violence (threats, insults, and intimidation)

Physical violence (blows, slaps, kicking, biting, seizing, physical attacks with objects like knives, belts, sticks and firearms, and physical violence leading to medical attention or rest)

Sexual violence (unwanted sexual contact)

Female employees were asked about their experience(s) of IPV, while male employees were asked about the perpetration of IPV. The answers were noted in the form of frequencies of indents, ranging from never to more than 20 times.

In Ghana and South Sudan, all four forms of violence (economic, psychological, physical, and sexual) were also taken into consideration. However, the survey questions provided a little more detail in terms of violence categories. For example, Psychological violence included insulting her, belittling or humiliating her in front of other people, making her feel bad, making her feel worthless, or doing things to scare or frighten her.

As in the case of Bolivia and Paraguay, female employees were asked about their experience(s) of IPV, while male employees were asked about the perpetration of IPV. However, male employees were also asked about their experience(s) of IPV. Answers were again recorded on the following scale: never; once or twice; between 3 and 5 times; between 6 and 10 times; between 11 and 20 times; and more than 20 times.

Estimation of Productivity Loss

Productivity loss has been estimated by absenteeism, tardiness, and presenteeism, with items and weighting based primarily on Duvvury et al. (2022). The items, weightings, and algorithms used are provided in Table 3.

Table 3.

Items, Weightings, and Algorithms of the Scales and Dimensions (Duvvury et al., 2022).

Dimensions	Items		Weightings	Algorithms
Absenteeism: Health reasons	A1	I was unwell at home	1	Annual Absenteeism (days)=
Absenteeism: Health reasons	A2	I had to go to a hospital or a health clinic because I was unwell	1	∑ [(Mean (A1, A2)), A3, A4, A5] *12
Absenteeism: Other reasons	A3	I had to look after a child or other family member because they were unwell	1
	A4	I had to attend to legal, financial, or personal matters	1
	A5	I did not have enough money for transport to and from work	1
Tardiness: Health reasons	T1	I was unwell at home	0.125	Annual Tardiness (days)=
Tardiness: Health reasons	T2	I had to go to a hospital or a health clinic because I was unwell	0.125	∑ [(Mean (T1, T2)) * 0.125, T30.125, T40.125] *12
Tardiness: Other reasons	T3	I had to look after a child or other family member because they were unwell	0.125
Tardiness: Other reasons	T4	I had to attend to legal, financial, or personal matters	0.125
Presenteeism: Low performance by distraction/exhaustion	P1	Difficulties concentrating on my work	0.25	Annual Presenteeism (days)=
	P2	Working more slowly than I would normally	0.25	∑ [(Mean (P1, P2, P3)) * 0.25, P4, P52] 12
	P3	Feeling exhausted at work	0.25
Presenteeism: Zero productivity	P4	Had to stop working because something was bothering him/her	1
Presenteeism: Zero productivity	P5	Made mistakes or had work-related accidents	2

Source. Authors’ own.

Absenteeism Loss

This is measured as the cost of missed workdays. The measurement items are based on those developed by Reeves and O’Leary-Kelly (2007, 2009). Male and female employees were asked to record the number of days they had been absent from work during the last 4 weeks (last month). Replies were recorded using a 5-item scale with six alternative answers: no days; 1 day; 2 days; between 3 and 5 days; between 6 and 10 days; 10 or more days. To calculate the total days lost, only the lower point of the range of each answer has been taken (Duvvury et al., 2022)

The scale comprises two dimensions: absenteeism for health reasons and absenteeism for other reasons. As an absence is equivalent to a whole workday, items measuring absenteeism are given a weight of 1. This is because 1 day missed is treated as one whole workday missed (Duvvury et al., 2022). To determine the number of days lost due to absenteeism per year all items are added and then multiplied by their relevant weightings. The resultant estimate is further multiplied by 12 months to provide an estimate of missed workdays in a year.

Tardiness Loss

Similar to absenteeism loss, tardiness loss is estimated based on the questionnaire developed by Reeves and O’Leary-Kelly (2007, 2009). As shown in Table 3, tardiness is measured on a 4-item scale comprising tardiness for health reasons and tardiness for other reasons. Male and female employees were asked to record the number of days they were tardy during the last 4 weeks (last month). As with the case of absenteeism loss, to calculate the total days lost due to tardiness, only the lower points of each answer have been used. As tardiness measures getting late to work by 1 hour and assuming 8 hours workday, it is given a weighting of 1/8th of a workday or 0.125 (Duvvury et al., 2022). To determine the annual number of days lost due to tardiness per year, all items were added, multiplied by their relevant weights and then multiplied by 12 months. Due to a higher number of employee data collected in Bolivia and Paraguay, compared to Ghana and South Sudan, the questionnaire only incorporated a one-item scale, whereby employees were asked “if they have been late/tardy by an hour in the workday.”

Presenteeism Loss

This is defined as the cost of unproductive work time, measured during a 4-week period (last month). As seen in Table 1, to measure presenteeism in Ghana and South Sudan, this study uses five items that are part of two dimensions: the low work performance due to distraction and exhaustion dimension, and the zero-productivity dimension days stopped due to accidents. The low performance by distraction and exhaustion dimension is based on the measurement items of the Work Distraction Scale devised by Stewart et al. (2003) and from the Work Limitations Questionnaire-WQL by Lerner et al. (2001).

The respondents were provided with the following options: no days; 1 day; 2 days; between 3 and 5 days; between 6 and 10 days; and 10 or more days. As in the case of Absenteeism loss and Tardiness loss, the lower points of each answer are used for estimation. To determine the number of days lost due to presenteeism per year, all items were added and then multiplied by their relevant weightings. The weighting for some items like “Difficulties concentrating on my work” is given as 0.25 as it is expected that you miss at least 25% of your workday output if you are not concentrating (Duvvury et al., 2022). The resultant estimate was further multiplied by 12 months to provide an estimate of presenteeism in a year. Unlike Ghana and South Sudan, to measure presenteeism in Bolivia and Paraguay, two additional questions were used due to respective country’s context: “Were you worried about personal or family matters that were not related to work?” and “Did you not work despite being at your workplace?”.

Estimation of Productivity Loss Due to IPV

To address recall bias and underreporting of missed workdays due to IPV, this study did not ask survivors directly “how many workdays have you missed due to IPV?” (Duvvury et al., 2019). Instead, the productivity loss questions were asked to every employee, whether experiencing IPV or not. With IPV being a binary variable, a difference in the mean days lost for women experiencing IPV and those not experiencing IPV is attributed to workdays lost due to IPV. In this study, two workdays lost due to IPV have presented as statistically significant difference in mean days between those who experience IPV and those who do not, estimated by controlling for age, tenure of work, financial dependents and contract using Propensity Score Matching (PSM; Guo & Fraser, 2015; Pan & Bai, 2015).The same estimation techniques were used to estimate days lost due to the perpetration of IPV.

Results

Descriptive Statistics

Tables 4 and 5 provide descriptive statistics for the African and South American countries. As seen in Table 4, the average age of the participants in Ghana was 32 years for female employees and 33 years for male employees. In South Sudan, the average female worker was aged 28 years, while for male employees the average age was 30 years. The variable “education level” reveals that, in Ghana, 76% of female employees and 87% of male employees had higher education qualification. In South Sudan, the overall educational level is lower, as only 61% of female employees and 53% of male employees had higher education qualification. In Ghana, 63% of female employees and 62% of male employees had been working in the company for more than 2 years. In South Sudan, not unexpectedly given the conflict context, these figures were 48% and 42% respectively. These results are in line with a higher percentage of permanent full-time employees in Ghana (60.2% women, 65.2% men) than in South Sudan (47.8% women, 40.6%, men; Tables 6 –8).

Table 4.

Descriptive Characteristics of Participants Stratified by Sex in Ghana and South Sudan.

	Ghana		South Sudan
	Women	Men	Women	Men
	(n = 391)	(n = 414)	(n = 323)	(n = 357)
Age
M (SD)	31.6 (8.8)	33.1 (9.1)	28.37 (7.0)	29.64 (8.2)
Education level (%)
No formal qualifications	0.3	0.8	3.7	4.2
Primary school	1.4	—	4.4	6.4
Secondary school	22.2	12.4	30.8	36.1
Undergraduate diploma	10.3	5.0	18.0	14.6
Bachelor’s degree or higher	65.8	81.8	43.2	38.7
Labor tenure (%)
Less than 12 months	11.7	13.1	23.1	26.6
Between 1 and 2 years	25.3	24.6	29.6	31.7
More than 2 years	62.9	62.3	47.4	41.7
Type of contract (%)
Permanent contract—full-time	60.2	65.2	47.8	40.6
Permanent contract—part-time	9.9	5.1	11.3	10.5
Temporary contract—full-time	22.9	20.5	24.7	20.7
Temporary contract—part-time	2.3	4.5	10.9	13.4
No contract	4.7	4.8	5.3	14.8
Financial dependents (%)
Respondents having financial dependents	47.7	54.4	61.1	62.4
Hours of work per day
M (SD)	8.3 (1.5)	8.3(1.4)	8.6(1.3)	8.9(1.5)
Days of work per week
M (SD)	5.3 (0.5)	5.3 (0.6)	5.7 (0.7)	5.8 (0.7)
Monthly income
US$ (SD)	193.7 (99.7)	220.5 (103.2)	99.4 (96.3)	91.7 (85.3)

Source. Authors’ own.

Table 5.

Descriptive Characteristics of Participants Stratified by Sex in Bolivia and Paraguay.

	Bolivia		Paraguay
	Women	Men	Women	Men
	(n = 2,705)	(n = 4,638)	(n = 3,643)	(n = 4,450)
Age
Mean (S.D.)	31.44 (7.68)	32.48 (8.90)	28.65 (7.71)	30.87 (9.15)
Labor tenure (%)
Between 1 and 3 years	38.0	33.6	46.6	36.0
More than 3 years	40.9	45.6	31.2	44.2
Indefinite contract (%)
Respondents with contracts	79.3	78.3	74.0	76.9
Financial dependents
M (S.D.)	1.65 (1.66)	1.96 (1.91)	2.53 (2.36)	2.33 (2.34)

Source. Authors’ own.

Table 6.

Violence Items in Bolivia and Paraguay: Survivors and Perpetrators (Vara-Horna, 2015, 2018).

Violence items	Bolivia				Paraguay
Violence items	Experienced	Not experienced	Perpetrated	Not perpetrated	Experienced	Not experienced	Perpetrated	Not perpetrated
Control or take of the income or salary (wconomical violence)	64	2,313	74	4,122	28	3,367	38	4,022
Threatened, insulted, or attacked verbally (psychological violence)	403	1,751	233	3,559	259	2,894	153	3,731
Hit, slapped in the face, kicked, bit, or grabbed forcibly by the arm (physical violence)	190	2,112	78	3,954	109	3,159	26	3,990
Attacked with objects, sticks, sharp weapons. or firearms (extreme physical violence)	33	2,400	5	4,229	27	3,332	9	4,055
Forced to have sex without consent (Sexual violence)	92	2,285	37	4,154	65	3,728	23	4,009
Caused physical damage (serious bruises, sprains, fractures, lesions, and cuts) (extreme physical violence)	121	2,244	43	4,131	50	3,264	15	4,029

Table 7.

Violence Items in Ghana and South Sudan: Survivors (Asante et al., 2019; Elmusharaf et al., 2019).

Violence items	Ghana		South Sudan
Violence items	Experienced	Not experienced	Experienced	Not experienced
Insulted me, or made me feel bad	77	159	90	75
Belittled or humiliated me in front of other people	33	199	59	107
Did things to scare or frighten me	55	176	54	111
Made me feel worthless	26	192	61	104
Prevented me from getting a job, going to work, trading, or earning money	12	222	26	141
Threatened to hurt me or someone I care about	21	210	39	128
Taken my earnings or salary against my will	13	218	17	149
Thrown me out of the house	19	211	27	140
Spent money on alcohol, tobacco, or other things for himself when he knew I did not have enough for essential household expenses	16	211	53	114
Slapped me or thrown things at me which could hurt me	17	212	37	129
Pushed or shoved me	32	198	41	124
Hit me with a fist or something that could hurt me	14	213	33	133
Hit, kicked, dragged, beaten, choked, or burnt me	8	221	27	138
Attacked me with a weapon or instrument such as a gun, knife, or other weapon	7	220	20	146
Physically forced me to have sex when I did not want to	15	213	36	131
Used threats or intimidation to get me to have sex when I did not want to	8	219	25	141
Forced me to do something else sexual that I did not want to	11	219	17	149

Table 8.

Violence Items in Ghana and South Sudan: Perpetrators (Asante et al., 2019; Elmusharaf et al., 2019).

Violence items	Ghana		South Sudan
Violence items	Perpetrated	Not perpetrated	Perpetrated	Not perpetrated
Insulted her, or made her feel bad	78	150	109	97
Belittled or humiliated her in front of other people	27	199	66	139
Did things to scare or frighten her	35	189	59	146
Made her feel worthless	25	201	61	144
Prevented her from getting a job, going to work, trading, or earning money	19	202	27	179
Taken her income or salary against her will	15	205	18	185
Visited her place of work to harass or threaten her	13	207	14	189
Thrown her out of the house	18	201	31	173
Spent money on alcohol, tobacco, or other things for myself when I knew she did not have enough for essential household expenses	16	199	54	149
Threatened to assault her or someone she cared about	16	198	38	165
Slapped her or threw things at her which could hurt her	21	193	39	164
Pushed or shoved her	27	185	45	160
Hit her with a fist or something else that could hurt her	13	203	30	174
Hit, kicked, dragged, beat, choked, or burnt her	12	204	29	174
Attacked her with a weapon or instrument such as a gun, knife, or other weapon	9	209	18	186
Physically forced her to have sex when she did not want to	17	198	28	176
Made her have sex when she did not want to by threatening or intimidating her	14	202	17	187
Forced her to do something else sexual that she did not want to	10	205	10	195

Hours of work per day and days of work per week were slightly different in both countries, with Ghana more inclined toward a 5-day working week and 8-hour workday, while South Sudan tended toward a 6-day working week and 9-hour workday. Given its conflict context, South Sudanhad a much lower average monthly wage (Women—US$ 91.7, Men—US$ 99.4) was than Ghana (Women—US$ 193.7, Men—US$ 220.5). Furthermore, a higher percentage of employees reported having financial dependents in South Sudan (61.1% women, 62.4% men) than in Ghana (47.7% women, 54.4 % men).

As detailed in Table 5, the average age of personnel in Bolivia was 31 years for women and 32 years for men, while in Paraguay, it was 29 years for women and 31 years for men. The percentage of employees with labor tenure between 1 and 3 years was higher for women than men in both Bolivia and Paraguay. In addition, the percentage of employees with a labor tenure of more than 3 years was higher for men than women in both countries. Approximately 79% of women and 78% of men had a permanent job contract in Bolivia, whereas, in Paraguay, this figure is slightly lower with 74% of women and 77% of men having a permanent contract. Bolivian and Paraguayan participants reported having two financially dependents on average.

The coefficients of internal consistency of the scales used for each country are displayed in Table 9. The Cronbach’s α coefficient for the scales (ranged from .657 to .923), the McDonalds’Ω (ranged from .678 to .927), and the composite reliability (ranged between .768 and .933) indicate adequate levels of internal consistency, values more the commonly used threshold (Hair et al., 2016). The table also shows that the average variance explained is between 45.7% and 95.9% for the second-order construct (theoretical dimensions of each scale), indicating convergent validity. The scales have also been tested for construct and criterion validity for all the studied countries (Duvvury et al., 2022; Vara-Horna, 2015, 2018).

Table 9.

Reliability and Validity Measures for Scales.

Scales	Internal consistency			Average variance extracted (AVE)
Scales	Ω McDonald	Composite	σ Cronbach	First order	Second order (range)
Ghana and South Sudan
Absenteeism and tardiness	.806	.849	.805	.349	.713; .734
Presenteeism	.807	.856	.802	.542	.740;0.959
IPV	.927	.933	.923	.455	.561; .682
Bolivia
Absenteeism and tardiness	.794	.828	.790	.445	.566; .605
Presenteeism	.785	.833	.775	.423	0.546; .629
IPV	.790	.846	.739	.484	.583; .594
Paraguay
Absenteeism and tardiness	.678	.768	.657	.364	.457; .593
Presenteeism	.747	.806	.730	.387	.478; .603
IPV	.803	.845	.744	.448	.502; .602

Prevalence of Survivors and Perpetrators of Intimate Partner Violence

Figure 1 provides the annual prevalence of surviving and perpetrating IPV. Given the cultural contexts, the prevalence of IPV survivors in Bolivia (19.4%) and Paraguay (10.7%) was much less compared to Ghana (43.2%) and South Sudan (66.7%).

Figure 1.

Prevalence of survivors and perpetrators of intimate partner violence in the last 12 months.

The prevalence of IPV perpetrators in Bolivia (8.2%) and Paraguay (6.5%) is much lower than in Ghana (40.9%) and South Sudan (60.2%). While the cultural context may partly explain the relatively low prevalence rates in the two South American countries, in the case of perpetrators, another possible reason is the shame and guilt associated with acknowledging the perpetration of violence. In Ghana and South Sudan, IPV is widely accepted and thus normalized; men do not necessarily consider what they are doing as wrong, which may not be the case in the South American countries (Sardinha & Catalán, 2018; Tausch, 2019; Tran et al., 2016). Other research has shown that in contexts where IPV is normalized, there is high reporting by men of perpetrating such violence (Babu & Kar, 2009). Moreover, perpetrators are often more honest with their responses when they have a non-socially sanctioning justification, such as blaming their partner. This kind of behavior could be explained because perpetrators tend to use denial, engage in personal attacks on victim credibility, and assume a victimized role to deflect blame (Harsey et al., 2017; Valor-Segura et al., 2011).

Productivity Impacts

Intimate Partner Violence: A Determinant of Absenteeism and Presenteeism

Using logistic regression, Tables 10 and 11, provide odds ratios which suggest that employees who either survive or perpetrate IPV have higher odds of missing workdays or being less productive. For example, in Ghana and South Sudan, the odds of IPV survivors missing work due to accessing health care in the last 12 months are 315% and 187% higher than for female employees who do not experience IPV, respectively. Similarly, the odds of perpetrators being distracted at work are 172% higher in Ghana and 175% higher in South Sudan than for men who do not perpetrate violence.

Table 10.

Odds Ratios of Survivors and Perpetrators Experiencing Absenteeism and Presenteeism in Ghana and South Sudan.

	Ghana		South Sudan
	Surv	Perp	Surv	Perp
Absenteeism
Illness	5.09***	2.01**	1.34	1.01
Health care	4.15***	1.71*	2.87***	1.76
Illness in the family(a)	2.23**	1.41	3.93**	1.47
Legal or financial difficulties	1.85*	1.98**	2.32	0.8
No money for transport	4.03***	1.97**	4.18*	1.38
Presenteeism
Distraction at work	4.65***	2.722***	1.84*	2.75***
Slowness at work	4.24***	3.44***	2.56**	2.42***
Physical exhaustion	3.94***	5.51***	2.51**	2.59***
Had to stop work	4.61***	3.35***	2.04	1.36
Accidents or incidents	2.01**	2.36***	2.06	0.82

***

p < .01. **p < .05. *p < .1.

Table 11.

Odds Ratios of Survivors and Perpetrators Experiencing Absenteeism and Presenteeism in Bolivia and Paraguay.

	Bolivia		Paraguay
	Surv	Perp	Surv	Perp
Absenteeism
Illness	1.24**	1.19	1.34**	1.47***
Health care	1.27**	1.31**	1.49***	1.52***
Illness in the family(a)	1.53***	1.46***	1.63***	1.8***
Legal or financial difficulties	1.41***	1.31**	1.62***	1.59***
No money for transport	1.39***	1.31**	1.62***	1.87***
Presenteeism
Distraction at work	2.04***	2.09***	2.62***	2.44***
Slowness at work	1.52***	2.04***	2.38***	2.36***
Physical exhaustion	2.02***	2.21***	2.16***	2.07***
Concerns about family	2.1***	2.65***	3.17***	2.28***
No work done despite being present	1.62***	1.94***	1.6***	2.19***
Had to stop work	1.5***	1.61***	1.97***	2.09***
Accidents or incidents	1.43***	1.49***	1.67***	1.36*

***

p < .01. **p < .05. *p < .1.

In Bolivia, the odds of survivors missing work due to illness in the family in the last 12 months are 53% higher for female employees who experience IPV than for female employees who do not experience IPV. Similarly, the odds of survivors in Paraguay being distracted at work are 162% higher for female employees who experience IPV. In both countries, perpetrators also have higher odds of absenteeism and presenteeism compared to non-perpetrators. For instance, in Bolivia, the odds of perpetrators working at a slower pace were 104% and in Paraguay 136% higher than for men who did not report perpetrating IPV.

The Elasticity of Labor Supply to Violence Exposure/Perpetration

Log-level regressions between absenteeism and IPV exposure/perpetration have been run, as shown in Tables 12 and 13. As shown in Table 12, the log-level regression models between absenteeism and IPV exposure were statistically significant for South Sudan, Bolivia, and Paraguay at 10%, 5%, and 1% levels of significance, respectively. For South Sudan, the estimate shows that IPV survivors have 36% higher absenteeism days compared to non-survivors. Similarly, in Bolivia and Paraguay, survivors have 14% and 23% higher absenteeism days, respectively, compared to non-survivors. The model was statistically insignificant for Ghana. The log-level regressions between absenteeism and IPV perpetration were statistically significant only for Bolivia and Paraguay at 5% and 1% levels of significance, respectively. In Bolivia and Paraguay, perpetrators have 14% and 27% higher absenteeism days, respectively, compared to non-perpetrators.

Table 12.

Absenteeism and IPV Exposure.

	Ghana	South Sudan	Bolivia	Paraguay
IPV exposure	0.03	0.36*	0.14**	0.23***
N	97	78	1,551	1,719
F	0.03	2.92*	6.09**	11.28***
R ²	.0003	.04	.0039	.0065

***

p < .01. **p < .05. *p < .1.

Table 13.

Absenteeism and IPV Perpetration.

	Ghana	South Sudan	Bolivia	Paraguay
IPV perpetration	0.19	−0.08	0.14**	0.27***
N	98	80	2,541	1,987
F	1.13	0.14	6**	13.42***
R ²	.01	.0018	.0024	.0067

***

p < .01. **p < .05. *p < .1.

Productivity Impacts Using Propensity Score Matching

Table 14 provides the lost productivity days of IPV survivors due to absenteeism, tardiness and presenteeism. The control variables for each country are age, tenure of work, financial dependents, and contract type. While the control variables are the same for all the countries, the measurement of some control variables is slightly different. For example, “financial dependents” is a binary variable in African countries, while it is a continuous variable in South American countries. This can be seen more explicitly in Tables 5 and 6. This is due to the data collection in different countries.

Table 14.

Lost Productivity Days of Survivors using PSM in the Last 12 Months.

	Coefficient (days)	Robust standard error	z	p > [z]	95% confidence interval
Ghana
Absenteeism	12.50	6.19	2.02	.04	[0.36, 24.64]
Tardiness	1.13	0.47	2.39	.02	[0.21, 2.06]
Presenteeism	17.07	6.88	2.48	.01	[3.58, 30.56]
Overall	30.70	11.94	2.57	.01	[7.30, 54.10]
South Sudan
Absenteeism	9.64	3.66	2.63	.01	[2.47, 16.82]
Tardiness	0.69	0.33	2.13	.03	[0.06, 1.33]
Presenteeism	6.24	3.11	2.00	.05	[0.14, 12.34]
Overall	16.57	5.90	2.81	.01	[5.02, 28.13]
Bolivia
Absenteeism	5.91	2.11	2.80	.01	[1.77, 10.04]
Tardiness	0.41	0.17	2.46	.01	[0.08, 0.73]
Presenteeism	4.78	1.46	3.28	.00	[1.93, 7.64]
Overall	11.10	3.19	3.48	.00	[4.84, 17.35]
Paraguay
Absenteeism	4.10	2.12	1.94	.05	[−0.05, 8.25]
Tardiness	0.38	0.14	2.70	.01	[0.10, 0.66]
Presenteeism	5.47	1.43	3.83	.00	[2.67, 8.26]
Overall	9.95	3.02	3.30	.00	[4.04, 15.86]

Note. The control variables were age, tenure of work, financial dependents, and contract type.

All survivors across the four countries reported a significant number of lost workdays: Ghana (M = 30.7, p = .01), South Sudan (M = 16.6, p < .01), Paraguay (M = 9.95, p < .01), and Bolivia (M = 11.1, p < .01). The difference in absenteeism days was also statistically significant for all the countries: Ghana (12.5), South Sudan (9.64), Paraguay (4.1), and Bolivia (5.9). Lost days because of presenteeism were the greatest for Ghana (17.07), followed by Paraguay (5.47), South Sudan (6.24), and Bolivia (4.78).

The productivity loss by perpetrators is also quite large, as seen in Table 15. In the case of absenteeism, the largest number of days lost was in Ghana (16.9), followed by Paraguay (7.96), Bolivia (7.01), and South Sudan (3.2). Likewise, lost days because of presenteeism were also higher for Ghana (19.08), followed by Bolivia (8.8) and Paraguay (5.47). The presenteeism results of perpetrators for South Sudan are statistically insignificant.

Table 15.

Lost Productivity Days of Perpetrators using PSM in the Last 12 Months.

	Coefficient (days)	Robust standard error	z	p > [z]	95% confidence interval
Ghana
Absenteeism	16.90	5.30	3.19	.00	[6.50, 27.30]
Tardiness	0.86	0.48	1.81	.07	[−0.07, 1.80]
Presenteeism	19.08	3.92	4.87	.00	[11.40, 26.77]
Overall	36.85	8.10	4.55	.00	[20.98, 52.72]
South Sudan
Absenteeism	3.16	5.49	0.58	.57	[−7.60, 13.93]
Tardiness	0.22	0.43	0.51	.61	[−0.62, 1.06]
Presenteeism	4.77	7.05	0.68	.50	[−9.04, 18.58]
Overall	8.15	11.21	0.73	.47	[−13.82, 30.11]
Bolivia
Absenteeism	7.01	2.19	3.20	.00	[2.71, 11.31]
Tardiness	0.71	0.19	3.65	.00	[0.33, 1.09]
Presenteeism	8.80	1.72	5.12	.00	[5.43, 12.16]
Overall	16.52	3.33	4.97	.00	[10.00, 23.04]
Paraguay
Absenteeism	7.96	2.96	2.69	.01	[2.16, 13.76]
Tardiness	0.68	0.17	4.09	.00	[0.35, 1.00]
Presenteeism	5.47	1.64	3.33	.00	[2.24, 8.69]
Overall	14.11	4.00	3.53	.00	[6.26, 21.95]

Note. The control variables were age, tenure of work, financial dependents, and contract type.

Comparing the four countries, Ghana incurred the largest loss in the case of both survivors and perpetrators, whereas Paraguay lost the least number of days for both survivors and perpetrators

The extreme variation in the number of days lost for survivors and perpetrators was expected as IPV does not, and cannot, lead to the same amount of productivity loss in each country; many factors such as the cultural, political, and economic situation of the countries play a role. Nonetheless, what is crucial is that, despite the intensity of productivity loss is different in each country, it is still quite significant for all the countries.

The productivity loss due to IPV found in this study for both survivors and perpetrators is very similar to that observed in previous studies (Blodgett & Lanigan, 2018; Kimerling et al., 2009; Mankowski et al., 2013; Rothman & Corso, 2008). The results of this study are also similar to the limited studies that exist that have attempted to quantitatively establish the productivity loss due to IPV for survivors alone. For example, in Fiji and the Solomon Islands, the equivalent of 10 workdays and 17 workdays respectively are lost per employee (whether they have experienced violence or not) due to employees feeling distracted, tired or unwell, being late for work, being absent, or because of helping others respond to domestic and sexual violence (IFC, 2019a, 2019b). Similarly, for employees in Papa New Guinea, Darko et al. (2015) found that each staff member lost almost 11 work days due to gender-based violence.

Discussion

This is the largest study of its kind which not only establishes but, unlike most previous studies, also quantifies the gigantic invisible costs that businesses incur due to IPV experienced by female employees in Africa and South America. While, as expected, there are significant variations between countries in prevalence rates and productivity loss, the key finding is that IPV does lead to significant productivity losses for each of the businesses surveyed in all of the countries.

The new ILO (2019) Convention on Violence and Harassment places an obligation on employers to address both violence at home and in the workplace. However, most companies still tend to underestimate IPV primarily due to three myths that this study addresses. First, there is the belief that violence against women is a private affair that does not affect the organization. As this study shows, businesses incur significant productivity losses due to IPV. Secondly, there is the belief that only physical violence leading to physical injuries should be a cause for concern as it increases the rates of absenteeism. However, as this study demonstrates, in all four countries, the costs of presenteeism for both survivors and perpetrators are as crucial as the costs of absenteeism. Finally, there is the belief that only female employees who are survivors of violence produce costs for businesses when this research clearly shows that male employees who are perpetrators of IPV are as costly as survivors or more so. Therefore, regardless of the gender composition of the workforce, companies should take an equal interest in programs to prevent and eradicate IPV.

In addition, the productivity loss to businesses ultimately leads to loss at the level of the economy, and thus investments by both businesses and the government are required to reduce IPV. While employee assistance programs have provided some evidence on supporting IPV survivors (Lindquist et al., 2010; Pollack, Austin, & Grisso, 2010; Pollack, Cummiskey, et al., 2010), they are far from sufficient. Interventions to reduce IPV need to be undertaken at a larger scale, as discussed extensively in the literature (Adhia et al., 2019; Katula, 2012). In the context of low and middle-income countries, a recent study by Al Mamun et al. (2018) has validated the intervention “HERrespect” in reducing both IPV and workplace violence against female garment workers in Bangladesh. Similar interventions that can be adapted to low and middle-income country contexts need to be undertaken.

In the present-day context of the coronavirus pandemic, the lockdown has brought into sharp relief the ongoing critical pandemic of IPV, as IPV related calls spiked across the globe (UN WOMEN). It is all the more essential for businesses to understand that the productivity loss due to IPV may further intensify as economies begin to reopen. A sustained effort to address IPV by governments, businesses, and communities has become more urgent than ever.

Limitations

The study has the following limitations. Firstly, though expected to be significantly less, this study has not included productivity loss where females are the perpetrators of IPV in a heterosexual relationship. Secondly, despite this study only using the difference in days of productivity loss for survivors/perpetrators and those who have not experienced/perpetrated violence to measure the days lost due to IPV, it may be possible that the days estimated are a slight overestimation. This is because of the extrapolation of days lost from the last 4 weeks to 1 year. Finally, though every effort has been made to account for the assumptions associated with the method used (PSM), the methods still have limitations and thus the days lost should be seen as an approximation rather than a perfectly accurate estimate. Fourthly, our sample was not representative of the cities selected and the business sectors covered. Therefore, the results should be treated as indicative and not perfectly reflective of the sampled sectors.

Future Research

Future research in this area can compare the average income of the formal versus informal sector at the country level and compare this with the prevalence of exposure to IPV in each country for the corresponding income categories. This will help in highlighting the economic strata where violence against women is more prevalent. Future research can also incorporate more variables into the PSM analysis as well as have a sample which is representative of the business sectors.

Research Data

sj-docx-1-sgo-10.1177_21582440231205524 – for Productivity Impacts of Intimate Partner Violence: Evidence From Africa and South America

sj-docx-1-sgo-10.1177_21582440231205524 for Productivity Impacts of Intimate Partner Violence: Evidence From Africa and South America by Nata Duvvury, Arístides Vara-Horna, Christine Brendel and Mrinal Chadha in SAGE Open

This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).

Footnotes

Author Note

This research was conducted while Dr. Mrinal Chadha was at the University of Galway. He is now working at the South East Technological University, Waterford and may be contacted at mrinalchadha07@gmail.com.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by Department for International Development (DFID), UK and COMVOMUJER Program by German Society for International Cooperation (GIZ).

Ethics Statement

Ethical approval was granted at the overall level by the Research Ethics Committee at the University of Galway. Additionally, in-country approval was granted by the University of Ghana (Ethics Committee for the Humanities) and by the National Bureau of Statistics, South Sudan. Both the Bolivia and Paraguay studies underwent internal review by the University San Martin de Porres (USMP) and COMVOMUJER Program (GIZ) to ensure alignment with standard WHO Safety and Ethical Guidelines for research on violence against women.

ORCID iDs

Nata Duvvury

Arístides Vara-Horna

Mrinal Chadha

References

Adhia

Gelaye

Friedman

L. E.

Marlow

Mercy

J. A.

Williams

M. A.

(2019). Workplace interventions for intimate partner violence: A systematic review. Journal of Workplace Behavioral Health, 34(3), 10.

Al Mamun

Parvin

Wan

Willan

Gibbs

Jewkes

Naved

R. T.

(2018). The HERrespect intervention to address violence against female garment workers in Bangladesh: Study protocol for a quasi-experimental trial. BMC Public Health, 18(1), 512.

Alsaker

Moen

B. E.

Baste

Morken

(2016). How has living with intimate partner violence affected the work situation? A qualitative study among abused women in Norway. Journal of Family Violence, 31(4), 479–487.

Anderson

D. G.

Fallin

Al-Modallal

(2014). Workplace violence experiences of homeless women and women residing in battered women shelters. Affilia, 29(1), 56–65.

Asante

Fenny

Dzidzor

Chadha

Scriver

Ballantine

Duvvury

(2019). Economic and social costs of violence against women and girls in Ghana: Country technical report. https://www.whatworks.co.za/documents/publications/302-10079-ghana-technical-report-final-web-file/file

Askeland

I. R.

Heir

(2014). Psychiatric disorders among men voluntarily in treatment for violent behaviour: A cross-sectional study. BMJ Open, 4(4), e004485.

Babu

B. V.

Kar

S. K.

(2009). Domestic violence against women in eastern India: A population-based study on prevalence and related issues. BMC Public Health, 9(1), 129.

Blodgett

Lanigan

J. D.

(2018). The prevalence and consequences of intimate partner violence intrusion in the workplace. Journal of Aggression, Maltreatment & Trauma, 27(1), 15–34.

Borchers

Lee

R. C.

Martsolf

D. S.

Maler

(2016). Employment maintenance and intimate partner violence. Workplace Health & Safety, 64(10), 469–478.

10.

Bott

Guedes

Ruiz-Celis

A. P.

Mendoza

J. A.

(2019). Intimate partner violence in the Americas: A systematic review and reanalysis of national prevalence estimates. Revista Panamericana de salud Publica, 43, e26.

11.

Breiding

Basile

K. C.

Smith

S. G.

Black

M. C.

Mahendra

R. R.

(2015). Intimate partner violence surveillance: Uniform definitions and recommended data elements [Version 2.0.]. Centers for Disease Control and Prevention.

12.

Chadha

Forde

Duvvury

(2020). Economic costs of domestic violence in Mongolia: Country report. UNFPA Mongolia. https://www.genderhub.mn/contents/view/115

13.

Chadha

Kennedy

Duvvury

(2022). Economic costs of violence against women and girls in low-and middle-income countries: A pilot study on management’s outlook. Workplace Health & Safety, 70(9), 405–411.

14.

Crowne

S. S.

Juon

H.-S.

Ensminger

Burrell

McFarlane

Duggan

(2011). Concurrent and long-term impact of intimate partner violence on employment stability. Journal of Interpersonal Violence, 26(6), 1282–1304.

15.

Darko

Smith

Walker

(2015). Gender violence in Papua New Guinea: The cost to businesses. The Overseas Development Institute.

16.

Duvvury

Carney

Nguyen

H. M.

(2012). Estimating the costs of domestic violence against women in Vietnam. UN Women Vietnam Country Office.

17.

Duvvury

Scriver

Forde

Chadha

Raghavendra

O’Brien

Sabir

Ballantine

(2019). Guidance on methods for estimating economic and social costs of violence against women and girls in low and middle income contexts. NUI Galway.

18.

Duvvury

Vara-Horna

Chadha

(2022). Development and validation of lost days of labor productivity scale to evaluate the business cost of intimate partner violence. Journal of Interpersonal Violence, 37, 2912–2943.

19.

Elmusharaf

Scriver

Chadha

Ballantine

Sabir

Raghavendra

Duvvury

Kennedy

Grant-Vest

Edopu

(2019). Economic and social costs of violence against women and girls in south Sudan (Country technical report). NUI Galway.

20.

Ghana Statistical Service. (2015). Ghana labour force survey 2015.

21.

Guo

Fraser

M. W.

(2015). Propensity score analysis. Sage.

22.

Gupta

Willie

T. C.

Harris

Campos

P. A.

Falb

K. L.

Moreno

C. G.

Díaz-Olavarrieta

Okechukwu

C. A.

(2018). Intimate partner violence against low-income women in Mexico City and associations with work-related disruptions: A latent class analysis using cross-sectional data. Journal of Epidemiology and Community Health, 72(7), 605–610.

23.

Hair

J. F.

Jr. Hult

G. T. M.

Ringle

Sarstedt

(2016). A primer on partial least squares structural equation modeling (PLS-SEM). Sage.

24.

Harsey

S. J.

Zurbriggen

E. L.

Freyd

J. J.

(2017). Perpetrator responses to victim confrontation: DARVO and victim self-blame. Journal of Aggression, Maltreatment & Trauma, 26(6), 644–663.

25.

IFC. (2019a). The business case for workplace responses to domestic and sexual violence in Fiji. https://www.ifc.org/wps/wcm/connect/de991ebc-a7f0-4118-9d13-daa8fcfead9d/IFC_Fiji_DSV_Report.pdf?MOD=AJPERES&CVID=mMYY8KA

26.

IFC. (2019b). The impact of domestic and sexual violence on the workplace in Solomon Islands. (Survey report). https://www.ifc.org/wps/wcm/connect/36087fa5-b699-453d-a558-f7a79e08906e/Survey+report+Solomon+Islands+impact+of+violence+on+workplace.pdf?MOD=AJPERES

27.

Instituto Nacional de Estadística de Bolivia. (2019). Encuestas de Hogares 2015-2019. http://anda.ine.gob.bo/index.php/catalog/84

28.

International Labour Organization. (2014). Thematic labour overview 1: Transition to formality in Latin America and the Caribbean (Revised version, p. 76). https://n9.cl/1ysdu

29.

International Labour Organization. (2019). C190: Violence and harassment convention. https://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:C190

30.

Katula

S. L.

(2012). Creating a safe haven for employees who are victims of domestic violence. Nursing Forum.

31.

Kimerling

Alvarez

Pavao

Mack

K. P.

Smith

M. W.

Baumrind

(2009). Unemployment among women: Examining the relationship of physical and psychological intimate partner violence and posttraumatic stress disorder. Journal of Interpersonal Violence, 24(3), 450–463.

32.

Kulkarni

Ross

T. C.

(2016). Exploring employee intimate partner violence (IPV) disclosures in the workplace. Journal of Workplace Behavioral Health, 31(4), 204–221.

33.

Lerner

Amick

III Rogers

W. M.

Malspeis

Bungay

Cynn

(2001). The work limitations questionnaire. Medical Care, 39(1), 72–85.

34.

Lindquist

C. H.

McKay

Clinton-Sherrod

A. M.

Pollack

K. M.

Lasater

B. M.

Walters

J. L. H.

(2010). The role of employee assistance programs in workplace-based intimate partner violence intervention and prevention activities. Journal of Workplace Behavioral Health, 25(1), 46–64.

35.

Logan

Shannon

Cole

Swanberg

(2007). Partner stalking and implications for women’s employment. Journal of Interpersonal Violence, 22(3), 268–291.

36.

Machisa

Shamu

(2018). Mental ill health and factors associated with men’s use of intimate partner violence in Zimbabwe. BMC Public Health, 18(1), 376.

37.

Mankowski

E. S.

Galvez

Perrin

N. A.

Hanson

G. C.

Glass

(2013). Patterns of work-related intimate partner violence and job performance among abusive men. Journal of Interpersonal Violence, 28(15), 3041–3058.

38.

Martinez

K. S.

(2015). Not just domestic violence: The role of the workplace in mitigating abusers. University of Maryland Law Journal of Race, Religion, Gender & Class, 15, 170.

39.

McFerran

(2011). National domestic violence and the workplace survey. Australian Domestic and Family Violence Clearinghouse, Australian Government.

40.

Meisel

Chandler

Rienzi

B. M.

(2003). Domestic violence prevalence and effects on employment in two California TANF populations. Violence Against Women, 9(10), 1191–1212.

41.

Naved

Rahman

Willan

Jewkes

Gibbs

(2018). Female garment workers’ experiences of violence in their homes and workplaces in Bangladesh: A qualitative study. Social Science & Medicine, 196, 150–157.

42.

Pan

Bai

(2015). Propensity score analysis: Fundamentals and developments. Guilford Publications.

43.

Parvin

Al Mamun

Gibbs

Jewkes

Naved

R. T.

(2018). The pathways between female garment workers’ experience of violence and development of depressive symptoms. PLoS One, 13(11), e0207485.

44.

Pollack

K. M.

Austin

Grisso

J. A.

(2010). Employee assistance programs: A workplace resource to address intimate partner violence. Journal of Women’s Health, 19(4), 729–733.

45.

Pollack

K. M.

Cummiskey

Krotki

Salomon

Dickin

Gray

W. A.

Grisso

J. A.

(2010). Reasons women experiencing intimate partner violence seek assistance from employee assistance programs. Journal of Workplace Behavioral Health, 25(3), 181–194.

46.

Pollack

K. M.

McKay

Cumminskey

Clinton-Sherrod

A. M.

Lindquist

C. H.

Lasater

B. M.

Walters

J. L.

Krotki

Grisso

J. A.

(2010). Employee assistance program services for intimate partner violence and client satisfaction with these services. Journal of Occupational and Environmental Medicine, 52(8), 819–826.

47.

Potter

Banyard

(2011). The victimization experiences of women in the workforce: Moving beyond single categories of work or violence. Violence and Victims, 26(4), 513–532.

48.

Rayner-Thomas

M. M.

(2013). The impacts of domestic violence on workers and the workplace (pp. 1–137). The University of Auckland.

49.

Reeves

O’Leary-Kelly

A. M.

(2007). The effects and costs of intimate partner violence for work organizations. Journal of Interpersonal Violence, 22(3), 327–344.

50.

Reeves

O’Leary-Kelly

A. M.

(2009). Study of the effects of intimate partner violence on the workplace (Report). Office of Justice Programs.

51.

Rothman

E. F.

Corso

P. S.

(2008). Propensity for intimate partner abuse and workplace productivity: Why employers should care. Violence Against Women, 14(9), 1054–1064.

52.

Sardinha

Catalán

H. E. N.

(2018). Attitudes towards domestic violence in 49 low-and middle-income countries: A gendered analysis of prevalence and country-level correlates. PLoS One, 13(10), e0206101.

53.

Schmidt

M. C.

Barnett

M. A.

(2012). Effects of domestic violence on the workplace. Center for Rural Studies at the University of Vermont.

54.

Sesar

Dodaj

Šimić

(2018). Mental health of perpetrators of intimate partner violence. Mental Health Review Journal, 23(4), 221–239.

55.

Shorey

R. C.

Febres

Brasfield

Stuart

G. L.

(2012). The prevalence of mental health problems in men arrested for domestic violence. Journal of Family Violence, 27(8), 741–748.

56.

Stewart

W. F.

Ricci

J. A.

Chee

Hahn

S. R.

Morganstein

(2003). Cost of lost productive work time among US workers with depression. JAMA, 289(23), 3135–3144.

57.

Straus

(2007). Conflict tactics scales in Encyclopedia of domestic violence, NA Jackson. Routledge and Taylor & Francis Group.

58.

Tausch

(2019). Multivariate analyses of the global acceptability rates of male intimate partner violence (IPV) against women based on World Values Survey data. The International Journal of Health Planning and Management, 34(4), 1155–1194.

59.

Tran

T. D.

Nguyen

Fisher

(2016). Attitudes towards intimate partner violence against women among women and men in 39 low-and middle-income countries. PLoS One, 11(11), e0167438.

60.

United Nations Development Programme. (2017). Human development reports. http://hdr.undp.org

61.

Valor-Segura

Expósito

Moya

(2011). Victim blaming and exoneration of the perpetrator in domestic violence: The role of beliefs in a just world and ambivalent sexism. The Spanish Journal of Psychology, 14(1), 195–206.

62.

Vara-Horna

(2015). Los costos empresariales de la violencia contra las mujeres en Bolivia. German Cooperation Agency GIZ. https://n9.cl/fhwk8

63.

Vara-Horna

(2018). Los costos-país de la violencia contra las mujeres en Paraguay. Una estimación causal-multinivel del impacto económico de la violencia contra las mujeres en relaciones de pareja. German Cooperation Agency GIZ. https://n9.cl/qqxyn

64.

Wathen

C. N.

MacGregor

J. C.

MacQuarrie

B. J.

(2015). The impact of domestic violence in the workplace: Results from a pan-Canadian survey. Journal of Occupational and Environmental Medicine, 57(7), e65.

65.

Wathen

C. N.

MacGregor

J. C.

MacQuarrie

B. J.

(2018). Relationships among intimate partner violence, work, and health. Journal of Interpersonal Violence, 33(14), 2268–2290.

66.

World Bank. (2011). International finance corporation. 2011. Doing business in Juba 2011. Doing business subnational. https://openknowledge.worldbank.org/handle/10986/13441 License: CC BY-NC-ND 3.0 IGO

67.

World Bank. (2017). World development indicators. https://www.worldbank.org

68.

World Health Organization. (1999). Putting women’s safety first: Ethical and safety recommendations for research on domestic violence against women (No. WHO/EIP/GPE/99.2).

69.

World Health Organization. (2013). Global and regional estimates of violence against women: Prevalence and health effects of intimate partner violence and non-partner sexual violence.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB