Prevalence of spousal violence among married women in a rural area in North Kerala

Abstract

Background:

Domestic violence has substantial public health consequences. More than one-third of women globally suffer from physical and sexual violence, mostly by an intimate partner with a lifetime prevalence of 10%–69%. In India, 28.8% of the ever-married women in the reproductive age group, reported domestic violence. Although Kerala, in southern India, performs better than other Indian states on a host of human development indicators for women, 15.8% among the women in this group have ever experienced physical or sexual violence in the past 12 months, the most common perpetrator being the husband.

Methodology:

A community-based cross-sectional study was carried out among the ever-married women, residing in a selected area of northern Kerala, who have been married for at least 1 year. One to one structured interview was conducted with a pretested structured questionnaire by the investigator. Spousal violence was defined as proportion of ever-married women with a lifetime experience of violence perpetrated by their spouse, manifested through acts of physical, sexual, and/or emotional violence, even threat.

Results:

A total of 290 women were interviewed. The age of the participants ranged from 18 years to 64 years; majority were Muslims. Only 16.5% were college-educated and over 90% were homemakers. 29% reported having experienced spousal violence; with the most common being emotional violence (19%) closely followed by sexual violence (18.6%) and physical violence (14.8%). A little over one-third (39.7%) reported controlling behaviour of their husband. Age of the participant (P = 0.019), absence of social support (P = 0.034), employment status of husband (P = 0.026), controlling behaviour of husband (P < 0.001), alcohol abuse by husband (P = 0.004) and extra-marital relationship of spouse (P < 0.001) were found to have significant influence on spousal violence experience.

Conclusion:

More than a quarter of the married women in this rural setting have experienced one or the other form of violence.

Recommendation:

A multifaceted approach needs to be employed which takes into consideration legal measures, social sciences, women empowerment, mental health systems and educate the men to curb this menace.

Introduction

As stated by the UN Secretary-General, Ban Ki-Moon “There is one universal truth, applicable to all countries, cultures and communities: violence against women is never acceptable, never excusable and never tolerable.” During the last few decades, both in developing and developed countries, violence against women and domestic violence in particular, has increasingly been acknowledged as violation of basic human rights.

Domestic violence has substantial public health consequences, such as chronic pain, injuries, fractures, disability, unwanted pregnancies, sexually transmitted diseases, and higher rates of HIV. The victims also suffer from significant psychological disturbances.[1] In addition, the economic impact due to increased health care costs and loss of women's labour hours must be also considered.[2]

According to WHO, more than one-third of women suffer from physical and sexual violence globally, with a lifetime prevalence of violence ranging from 10-69%, mostly by an intimate partner.[1] As per WHO estimates for South-East-Asia, around 40% are exposed to domestic violence in India. National Family Health Survey (NFHS)-4 (2015-2016) reports the prevalence of spousal violence among the ever-married women in the reproductive age group to be 28.8% ranging from 2.6% in Sikkim to 53.1% in Manipur.[3] In most states, the spousal violence is more common in the rural areas.

More than 6 in every 10 women victims of violence refrain from asking for help or support or protection of any sort.[4] The remaining ones who do speak up turn to their family and friends for support. As per NFHS 4, more than two-thirds of women 69% believe it's justifiable for a husband to beat his wife under some circumstances. Even among those who completed at least 12 years of schooling, 67% justified a husband beating his wife for specified reasons.[3]

Kerala is an interesting state in India for analyzing the prevalence of violence against women, as it performs better than other Indian states on a host of human development indicators for women. It has a favorable female sex-ratio of 1,058 females to 1,000 males (India's 933), a high female literacy rate of 92.1% (65.5% at national level), and a high female life-expectancy of 77 years (67.7 years nationally).[5] Though the situation is much better in the state than rest of India, economic impoverishment, malnourishment, low political participation rates, and low labor force participation rates are prevalent among women in Kerala.[6] As per NFHS-4 survey (2015-2016) in Kerala, 15.8% among the ever-married women aged 15-49 years have ever experienced physical or sexual violence in the preceding 12 months, the most common perpetrator being the husband for the ever-married women and this increases to 16.5% when emotional violence is included. According to Pradeep Panda, both physical and psychological violence were relatively higher in rural areas than in urban areas.[5] The prevalence and perception of domestic violence varies with the local social norms and literacy level of women; hence, it is important to assess the problem in a given geographical area. Hence, this study was carried out to assess the prevalence and types of spousal violence among ever-married women residing in a selected rural area in Malappuram district of North-Kerala, South-India; and find out the association between spousal violence and selected socio-demographic variables.

Methodology

A community-based cross-sectional study was carried out among the ever-married women residing in a selected rural area of North Kerala who have been married for at least 1 year. Women were excluded if they refused to be interviewed or there was of lack of privacy or gave incomplete responses to the spousal-violence module of the questionnaire. Considering the prevalence of spousal violence of 16.5% as per NFHS-4, sample size was calculated to be 243, accommodating a 10% non-response rate.[3] Sampling unit was household and were selected using systematic random sampling. One person was administered the spousal violence module in each selected household; if there were more than one eligible respondent in a sampled household, one was selected by lottery method. In absence of an eligible respondent in the sampled household, the next house was visited for the survey. The houses that were locked or those in which eligible women were not present at the time of the visit were noted and were excluded if the woman could not be contacted even after three revisits, and the next household in sample list was approached.

Study tools

The women were interviewed with a pretested structured questionnaire, adapted from NFHS-3. The questionnaire was translated into Malayalam and then retranslated into English to ensure originality and meaning was retained. The questionnaire was pre-tested among married women of reproductive age group during a mothers’ meeting and necessary amendments were made. The questionnaire consisted of socio-demographic characteristics of the women and questions to determine whether the respondent had experienced any form of spousal violence. Different forms of spousal violence (physical, psychological, sexual) were assessed, using questions that asked for presence and frequency of specific experiences. Five questions were used to assess presence of physical violence, a set of three questions for sexual violence and emotional violence was assessed, using a set of three questions. An affirmative answer to one or more of these questions implied presence of that form of violence. The frequency of occurrence of each individual experience was asked for, in case of an affirmative answer. Victims of SV were further asked whether they reported the violent episodes, and if they did, to whom.

Spousal violence (outcome variable) was defined as proportion of ever-married women with the lifetime experience of violence perpetrated by her husband, manifested through acts of physical, sexual, and or emotional violence even threat, at any point in her life. An affirmative response to any of the aforementioned 11 questions was considered as having experienced spousal violence irrespective of the type of violence. A woman is considered to have ‘no spousal violence’ when the answer to all the questions in the set is negative.

Independent variables/factors at individual, family and community levels expected to have association with spousal violence were selected based on extensive literature search and assumptions derived from subject matter knowledge. Women's characteristics-age in completed years; education level; occupation; and possession of any property was reported as present or absent. Husband's characteristics-age in years; employment status (employed or not); educational qualification; alcohol consumption habit. The respondents were asked whether their husband has any extramarital relationship or stayed away from home for days together. Husband's controlling behavior was also assessed by asking a set of 9 questions. If the answer to one or more of these questions is ‘Yes’, the woman is considered to have a controlling husband. Family level- Duration of marriage in completed years; age of the respondent at marriage; living arrangement (nuclear/joint); number of children. Community level variables- Religion, social support.

Data Collection was completed over a period of 3 months. After explaining the purpose of study and obtaining written informed consent, one-to-one interview was conducted by the co-investigator after developing trustworthy relationship. As this is a very sensitive matter, adequate privacy and absence of a third party was ensured and the location for the interview was selected based on the respondent's preference.

Ethical concerns

The study protocol was approved by the Institutional Scientific Committee and Institutional Ethics Committee. Informed written consent was obtained from the participants prior to the interview. If the participant had mental breakdown during the interview, help was offered in the form of counseling by the co-investigator, who had undergone training for 1 week in the Psychiatry department, prior to the study.

Data analysis

310 women who were approached, agreed to be interviewed. We excluded 20 participants with missing data. Data were analyzed using SPSS-16 trial-version. Descriptive analysis was done. Distribution of categorical variables were reported as frequency counts whilst crude associations between dependent and independent variables were tested using Chi-square or fisher's exact test. Odds ratio (ORs) and 95% confidence interval (CI) were computed for those factors that were found to have a statistically significant association (P < 0.05).

Results

Out of 290 eligible women who participated in the study, 84 (29%) were victims of at least one type of violence by their husband at any point in their married life. The most commonly occurring form was emotional violence (19%); followed by sexual violence (18.6%) and physical violence (14.8%). The different forms of violence were overlapping in nature [Figure 1], with 7.9% having experienced all three forms of violence concurrently, while 13.4% experienced only one form. As shown in Table 1 depicting the frequency distribution of each act of violence, the single most common act of violence reported was forceful sexual intercourse (15.17%). The most commonly occurring act of emotional violence was the husbands not reacting when his family insults them (14.5%), followed by 13.1% reporting being insulted by their husbands in front of others. Slapping (11.38%) was reported to be the most common form of physical violence, followed by being beaten up or having things thrown at them (8.62%).

Figure 1.

Distribution of different types of Domestic violence

Table 1.

Frequency distribution of different forms of violence (n=290)

Form of violence n (%)	Never n (%)	Once or twice n (%)	Rarely n (%)	Regularly n (%)	Always n (%)
Emotional violence			55 (19%)
Doesn’t react when his family insults you	248 (85.5)	17 (5.9)	14 (4.8)	5 (1.7)	6 (2.1)
humiliated you in front of others	252 (86.9)	15 (5.2)	16 (5.5)	5 (1.7)	2 (0.7)
Threatens to harm you physically	263 (90.7)	15 (5.2)	7 (2.4)	1 (0.3)	4 (1.4)
Sexual violence	54 (18.6%)
Ignored you purposefully and avoided sex with you even if you wanted	266 (91.7)	17 (5.9)	1 (0.3)	2 (0.7)	4 (1.4)
forced you to have sexual intercourse with him without your consent or Forced you to perform any sexual acts you did not want to	246 (84.8)	17 (5.9)	10 (3.4)	10 (3.4)	7 (2.4)
Has force you to have sex with others	288 (99.3)	0 (0)	0 (0)	1 (0.3)	1 (0.3)
Physical violence	43 (14.8%)
Has ever slapped you	257 (88.6)	23 (7.9)	9 (3.1)	0 (0)	1 (0.3)
Has ever punched you with his fist or with something or beat you up	265 (91.4)	14 (4.8)	9 (3.1)	1 (0.3)	1 (0.3)
Has ever twisted your arms/pulled your hair	274 (94.5)	9 (3.1)	2 (0.7)	1 (0.3)	4 (1.4)
Has ever kicked you/dragged you/stamped you	281 (96.9)	5 (1.7)	1 (0.3)	0 (0)	3 (1.0)
Has choked you/inflected burns on you	286 (98.6)	2 (0.7)	0 (0)	0 (0)	2 (0.7)

The age of the participants ranged from 18 to 64 years with a mean of 33.7 years (SD 8.6). Only 1.4% was under 20 years. Women aged ≥35 years were found to be at twice the risk for experiencing spousal violence compared to women <35 years (COR = 1.843; 95%CI 1.103-3.078). Though majority who experienced domestic violence had not been to college (85.7%), no significant association (P = 0.507) was found between educational status of the women and falling victim to SV. Majority (90.3%) of the participants were home makers. Of the 28 employed participants in our study, only seven reported being exposed to spousal violence; however, the employment status of the women and exposure to spousal violence were not found to be associated (P = 0.626). No relation was seen between religion and spousal violence. [Table 1]

The mean age at marriage was 18.9 years (SD 4.3) with nearly half (44.8%) married before the age of 18 years. The average duration of marriage was 14.9 years (SD 8.6) with majority (64.5%) having been married for more than 10 years. One of our participants was divorced and another's husband had passed away. More than half were living in joint families. No significant association could be established between spousal violence and age at marriage or years of marriage or living arrangement (nuclear/joint). [Table 2]

Table 2.

Socio-Demographic details of the participants (n=290)

Characteristics	Frequency	Percent	Test of significance	p value	Prevalence of SV	OR (95% CI)
AGE (years)
<20 years	4	1.4
20-29 years	101	34.8			<35yrs(23.7%) vs >35yrs(36.3%)	1.843 (1.103-3.078)
30-39 years	106	36.6	t 2.168	0.03
40-49 years	65	22.4
>50 years	14	4.8
Religious distribution
Hindu	99	34.1
Muslim	183	63.1	χ2 1.405	0.495	NA	NA
Christian	8	2.8
Educational qualifications
Primary	18	6.2	Upto Sen.sec.school vs college educated
Secondary	156	53.8
Senior Secondary	68	23.4	χ2 0.440	0.507	NA	NA
Diploma/Degree	25	8.6
PG / Professional degree	23	7.9
Employment status
Unemployed	262	90.3	χ2 0.237	0.626
Employed	28	9.7			NA	NA
Living arrangement					NA	NA
Joint family	161	55.5	χ2 2.987	0.084
Nuclear family	129	44.5			NA	NA
Number of children
0	16	5.5	None vs >1
1-2	169	58.3	χ2 0.859	0.354	NA	NA
>3	105	36.2
Currently pregnant/Lactating
Non Pregnant Non Lactating	275	94.8
Pregnant / Lactating	15	5.2	χ2 0.618	0.432	NA	NA
Social support
Yes	51	17.6	χ2 4.485	0.034	41.2%	0.511 (0.273 – 0.958)
Suicidal thoughts
Yes	10	3.4	χ2 2.227	0.136	NA	NA
Owns Property
Yes	43	14.8	χ2 2.741	0.098	NA	NA

The mean age of the spouse was 41.4 years (SD = 9.5) with the age difference ranging from - 2 to + 10 years. A significant difference in age between spouses was noted between those who experienced domestic violence [mean age 35.44 years; SD = 9.1] and those who did not [mean age 33.05; SD = 8.3] (t = 2.168; P = 0.031). [Table 3] Only 11.8% spouses had been to college. Two spouses were unemployed. Unemployment status of spouse was found to put women at 3.5 times higher risk for spousal violence. No significant association was noted between age or educational status of spouse and spousal violence [Table 4]

Table 3.

Controlling Behaviour of spouse

CONTROLLING HUSBAND	Never		Once or twice		Rarely		Regularly		Always
	n	%	n	%	n	%	n	%	n	%
Did not permit to meet friends	212	73.1	40	13.8	8	2.8	3	1	27	9.3
Restricted interaction with own family	245	84.5	15	5.2	19	6.6	2	0.7	9	3.1
Doesn’t permit to handle money	223	76.9	18	6.2	21	7.2	5	1.7	23	7.9
No value for one’s choice	216	74.5	21	7.2	21	7.2	7	2.4	25	8.6
Doesn’t permit to talk with male friends	250	86.2	20	6.9	6	2.1	3	1	11	3.8
Accused of being unfaithful	270	93.1	7	2.4	9	3.1	2	0.7	2	0.7
Did not allow to pursue a career	256	88.3	18	6.2	5	1.7	2	0.7	9	3.1
Treated like a servant	250	86.2	12	4.1	14	4.8	10	3.4	4	1.4
Doesn’t allow to take part in decision making	223	76.9	26	9	15	5.2	9	3.1	17	5.9

Table 4.

Distribution of socio-demographic characteristics of spouses of the participant (n =290)

Characteristics	Frequency	Percent	Test of significance	p value	Prevalence Of SV	OR (95% CI)
SPOUSE’s AGE (years)
21-30	31	10.7
31-40	122	42.1
41-50	86	29.7	t 1.262	0.208	NA	NA
51-60	41	14.1
>61	9	3.1
Died	1	.3
Educational qualifications
Primary	39	13.4	Upto Sen.sec.school vs college educated
Secondary	185	63.8
Sen. Secondary	26	9.0	χ21.813	0.178	NA	NA
Diploma/Degree	28	9.7
PG/Professional	12	4.1
Employment status
Unemployed	2	0.69			100% vs 28.5%	3.512 (2.925 – 4.218)
Employed	288	99.31	χ24.939	0.026
Alcohol abuse of husband
Yes	29	10.0	χ28.111	0.004	51.7%	2.981 (1.369-6.494)
Controlling Husband
Yes	114	39.3	χ2101.327	<0.001	62.3%	20.703 (10.488-40.868)
Kept away from home for days
Yes	16	5.5	χ228.199	<0.001	87.5%	20.4 (4.523-92.001)
Has extra-marital relations
Yes	10	3.4	χ218.75	<0.001	90%	24.6 (3.065 -197.471)

Sixteen participants did not have any children. No significant association could be established between spousal violence and not having children. Among the 12 pregnant participants, only three had ever experienced spousal violence and none among the three lactating women reported spousal violence exposure. Current pregnancy or lactation was not associated to experiencing spousal violence. [Table 3]

A little over one-third (39.3%) reported controlling behaviour of their husband. Most of the participants reported they were not allowed to meet their friends (26.89%), followed by a quarter (25.52%) reporting that their husbands did not value their choice. [Table 4] The controlling behavior of husband and experiencing spousal violence was found to be significantly associated (P < 0.001), with 84.52% of those who experienced spousal violence reporting to have controlling husbands. They were 20 times likely to experience spousal violence than the women having non-controlling husband (COR = 20.703, 95% CI 10.488-40.868). [Table 5]

Table 5.

Distribution of Marital characteristics of the married women in (n =290)

Characteristics	Frequency	Percent	Test of significance	p value
Age at marriage (in years)
< 18 years	130	44.8	<18 vs >18 χ2 0.029	0.865
18-30 years	155	53.5
>30 years	5	1.7
Duration of marriage (years)
<1 year	6	2.1
1-5 years	40	13.8	t 1.842	0.067
5-10 years	57	19.7
>10 years	187	64.5
Age difference
+/- 2 years	19	6.6
3-5 years	50	17.2	t -1.943	0.053
5-10 years	186	64.1
>10 years	35	12.1

Twenty-nine (10%) reported that their husbands consumed alcohol. Women whose husbands consumed alcohol were almost at 3 times more risk of experiencing spousal violence compared to women whose husbands don’t consume alcohol (COR = 2.981, 95% CI 1.369-6.494). A few (5.52%) reported their husbands stayed away from home without informing them while 10 participants (3.45%) reported their husbands have extra-marital relationships. Women whose husbands had extra-marital relationship had 25 times higher risk for spousal violence (OR = 24.6; CI 3.065-197.471). Similarly, women who reported husbands kept away from home regularly were at 20 times higher risk for spousal violence (OR 20.4; CI 4.523–92.001). [Table 5]

Ten participants reported they have had suicidal thoughts at one point or the other in their life; no significant association (P = 0.136) was noted between suicidal thoughts and experiencing spousal violence [Table 3].

Only 17.6% reported having social support and presence of social support was found to be protective against spousal violence (OR = 0.511, CI 0.273–0.958) [Table 3].

Five victims of spousal violence reported having sought help, three sought help of their parents/family and one gave a complaint at the police station. None of the pregnant or lactating participants who experienced domestic violence sought help.

Discussion

In this study, more than one-fourth of the married women were victims of spousal violence. This was in concordance with the national findings as per NFHS-4 (28.8%), but higher than the findings for Kerala during NFHS-4 (17%).[3] Similar findings have been reported from West Bengal (23.4%), Nepal (26.4%), NFHS-3 (37.2%), Saudi Arabia (34%), Ghana (33.6%) and Pakistan (33%).[7,8,9,10,11,12] Study done in Puducherry (56.7%) showed almost twice the prevalence.[13] The exact reasons for the differences are not known. Domestic violence varies with the local sociocultural norms such as acceptability of physical violence at the hands of husbands, literacy level of women, women's autonomy in decision making, and limited freedom of expression. These differences could be also attributable to the methodological variations of the studies.

In this study emotional violence was the predominant form (19%) of spousal violence whereas NFHS-4 reported physical violence was predominant (13%) and only 9% experienced emotional violence in Kerala. Higher proportion reported experiencing emotional violence in rural Nepal (31%), North Bengal (54.5%), rural Puducherry (51.3%) and Eastern India (52%).[8,14,15,16] The complexities in quantifying emotional abuse unlike physical or sexual abuse could be the reason for the wide differences noted.

In the study, 14.8% of the ever-married women have experienced some form of physical violence, similar to the findings in Kerala (13%) as per NFHS-4 and rural Nepal (29.6%).[3,8] Slapping was the most common single act reported similar to most other studies. But NFHS-4, in Kerala, identified pushing/being shaken was the commonest (9%) form of physical violence.

Sexual violence was reported by 18.6% in the study similar to study done in Puducherry (13.5%).[15] But lower prevalence were reported by NFHS-4 in Kerala (3%) and rural Nepal (6.8%).[3,8] The lower occurrence of sexual violence observed may be due to the underreporting in rural settings. People generally don’t complain about male members at any cost or are scared to disclose their personal issues especially in rural areas, because of cultural norms and conjugal affairs being perceived as private matter.

As seen in multiple studies, most of the women were subjected to more than one type of violence. There were overlaps between different types of violence, with 7.9% experiencing all three forms of violence concurrently, which is higher than the findings from rural Vietnam whereas significantly lower than what was seen in Nepal and Ethiopia.[8,17,18] About 8% experienced any two forms in combination, while 13.4% experienced only one form. This study reported combination of sexual and emotional abuse as the most commonly occurring violence forms in contrast to combination of physical and emotional in most studies. These combinations with emotional violence might be explained as physical/sexual violence being often accompanied by psychological attacks, threatening, and controlling behaviours.

Age of the participant was found to have a significant influence on spousal violence in the study, similar to studies done in Puducherry and Bangladesh.[15,19] Employment status of the husband was found to be associated with spousal violence, where women with unemployed husbands were at 3.5 times higher risk for spousal violence. Similar findings were seen in Puducherry and rural Nepal.[8,15] Many studies identified age of the spouse and his educational status associated to spousal violence, but these associations could not be elicited in this study.[20] Alcohol consumption by husband has been identified as risk factor for spousal violence similar to other studies done in India (NFHS-3), Nepal, Jeddah and Ghana.[8,9,10,11] Many researchers believe that alcohol operates as a situational factor increasing the likelihood of violence by reducing the inhibitions clouding of judgment and ability to interpret cues. Controlling behaviour of the husband has been acknowledged as a crucial issue in the marital relationships and was identified to be a significant risk factor in the study. Economic dependency led them to tolerate violence and lack of awareness about existing laws. The situation is even worse due to deep-rooted sociocultural norms, acceptance as normal male behavior in male-dominated society, subordinate position of women. There are arguments about controlling behavior, whether it is a contributing factor or part of the violence acts.[20]

Only a minor proportion (7.9%) sought help on being abused similar to Nepal study (9.7%) but 28% in Kerala (NFHS-4) and 34% in Bangladesh reported seeking help. And most studies point that the women sought help mostly from their parents.[3,8,19]

Some potential limitations need to be acknowledged. Like any study based on self-reporting, recall bias may have been associated with disclosure of acts of violence. Being cross-sectional in nature, only statistical association can be proved but casual association can’t be established. Since the study was conducted in a small geographical area, the findings may not be generalizable to other rural areas. The extraction of details regarding personal or husband's or family income was not successful. Presence of male child and reason for not reporting was not elicited in this study.

To summarize, more than a quarter of the married women in this rural study setting have experienced one or other form of violence. The different forms are overlapping in nature. Age of the participant, unemployment status of husband, controlling behaviour of husband, extra-marital relationship of husband, husbands keeping away from home without informing and alcohol consumption by husband were found to have a significant influence on spousal violence. Presence of social support was found to be protective against spousal violence.

If women are not safe in their homes especially from their husband, then how can we assure their safety outside home and from others?

To conclude, despite enforcement of law against domestic violence and several governmental and non-governmental organizations working to eliminate, every form of discrimination against women is still prevailing in our society. The patterns and prevalence of spousal violence vary with time, person and locality.

Recommendations

A multifaceted approach needs to be employed to curb this menace, taking into consideration legal measures, social-cultural factors, women empowerment, integration of mental health to primary health care and education of the men.

Women are unaware of their rights or lack knowledge of how and where to seek help and many at times fear reprisals for reporting. Improving overall education of girl child will help in strengthening self-esteem of women and girls and thus empowering them against the domestic violence. Awareness programs regarding where and whom to seek help in case of violence need to be conducted keeping the societal context of India in mind. Woman self-help groups in the rural areas may be involved in these programs.

Opportunistic screening for domestic violence by physicians at primary health care facilities will be useful. Community-level workers such as ASHA and anganwadi workers can be involved in targeted screening of married women since they have an existing rapport with them and routinely interact with the women for the provision of services. Helplines should be initiated for women at every primary health-care centre. There should be mental health professionals as well as professional social workers in the primary health centres to address this issue effectively.

The services should be focusing not only woman but also family members so that they have the right attitude regarding domestic violence. There is a strong need to challenge the social norms and bring changes in attitudes that foster violence and promoting equity in marital relationships. In addition, men should be educated early in life, through schools, to instil values such as respect for women, which is likely to have a positive effect on the boys in the long term.

Ethical concerns

Ethical clearance was obtained from the Institutional Ethics Committee as well as written informed consent was taken from all participants.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.