Are women with more of 'social capital' more empowered? A cross-sectional analysis from rural Wardha, Central India.

Background: Despite international commitment and government policies and programmes, a vast majority of women of rural India are socially and economically deprived in the patriarchal sociocultural framework. Strategies to improve women empowerment need context-specific field-based evidence.
Objectives: This study was undertaken to address the research question: "Are women with a higher level of social capital more empowered than their counterparts?"
Methods: A community-based cross-sectional analytical study was carried out in the Wardha district of Maharashtra, where a two-stage cluster random sampling technique was used to select the 300 study subjects. The outcome variable, women empowerment was assessed using four domains: mobility, inter-spouse communication, household decision-making and asset ownership. The independent variables included social capital along with potential sociodemographic confounders. Multivariate linear regression was applied using the backward stepwise method.
Results: The participants had better women empowerment percentage score in the "Household decision-making" domain compared to the other domains. The women empowerment score was found to be significantly higher in the participants with increasing age, higher education, those involved in business/service, above poverty line (ABL), nuclear family, married, and those having at least one living child. In the adjusted analysis, the social capital was found to be a significant determinant of women empowerment, along with age, education, religion, marital status and family type.
Conclusion: The improvement of social capital in the form of community-based development projects through Self-Help Groups (SHG) and microfinance programmes need to be encouraged by the Government and NGOs, as this strategy has the potential to improve women empowerment through community-driven development. Copyright:

Introduction

The improvement of women’s social status is an international agenda which is also highlighted in the Sustainable Development Goals to be achieved by 2030.[1] It is much relevant in counties like India, where women are traditionally neglected due to the patriarchal socio-cultural framework, resulting in a considerable gender gap in the socioeconomic and development-related indicators.[2] Various interventions such as gender-sensitive legislative measures, increasing political participation by reservation, focus on women education have been adopted by the Government to overcome this social disadvantage.[34] Despite all the efforts, a huge proportion of women in India are to date socially and economically deprived. Empowering women is an important weapon to change the current scenario. Women empowerment is a foundation stone for the economic development and end of inequity and violence against women. As Kabeer said women empowerment is about acquiring “the ability to make strategic choices in life in the context in which they were previously denied this ability”.[5] Despite the international commitment and numerous policies and programmes, not much could be achieved as far as women empowerment is concerned. This could be attributed to the fact that the current policies and programmes are based on theoretical framework rather than field-based evidence.

Social capital is the amalgamation of “the links, shared values and understandings in a society that allow people and community members to have faith in each other so that they can work together”.[6] It is constituted by the extent to which people are embedded within their family, relationships, social networks and communities and have a sense of belonging and civic identity, which can considerably affect people’s life chances and trajectories.[7] In a patriarchal society, women are traditionally involved in forming informal and local networks, unlike men, who generally are involved in formal networks.[8] Such social networks established by women have a protective effect and work as risk reduction mechanisms. The community-based development projects and Self-Help Groups (SHG) are also found to be effectively increasing bonding, trust and solidarity among members of the target group, thereby, improving their social capital along with the economic growth.[9] Improving social capital could be an important intervention to achieve stagnated goals of women empowerment. Evidence from the existing literature revealed that different domains of social capital have the potential to affect all the constructs of women empowerment. A research gap is identified whether social capital is affecting women empowerment. The present study envisages the understanding and generating evidence about whether the social capital is associated with women empowerment.

Materials and Methods

Study design, setting and population

The present community-based cross-sectional analytical study was carried out in the Wardha district of Maharashtra which is situated in Central India. As per the 2011 India census, the Wardha district had a population of 1,300,774 with a majority (67.5%) residing in rural areas. The district is having a sex ratio of 946 (48.6% female population). Wardha has an average literacy rate of 86.99%, higher than the national average of 74.0%, the female literacy being 81.81%. Out of 27 Primary Health Centres (PHC) in the district, the Kharangana (Gode) PHC with a population of 36,769, spread over 22 villages, was purposively chosen for the study.[1011] All married women in the reproductive age group (15–49 years) residing in the area of the Kharangana PHC served as the study population.

Sample size and sampling technique

Considering 50% as a proportion of empowered women, 5% alpha error and 10% allowable error in the estimate of the proportion and design effect of 2, the sample size required was 210. To cover for non-response, 300 married women in the reproductive age group were planned to be studied. A two-stage cluster random sampling technique was used to select the 300 study subjects. In the first stage, 30 clusters were identified by probability proportionate to the size of the population technique. In the second stage, from each of the clusters, 10 eligible respondents were selected by random walk method, giving a total sample size of 300.

Measurements and data collection

the following data collection was done by house-to-house survey. Before declaring unavailable, three attempts were made so that the minimum drop-out rate is achieved.

Women empowerment:

For the purpose of measurement, we defined women empowerment as “the ability to exercise agency and acquire resources within a context of gender inequality”.[5] Four domains were included as described by in a World Bank report: “Mobility” (socio-cultural dimension: 3 questions); inter-spousal communication (psychological dimension: 4 questions); “Household decision making” (familial dimension: 8 items) and “Asset ownership” (economic dimension: 5 items).[12] Reliability in terms of Cronbach’s alpha was found to be 0.817 for the women empowerment scale.

Social capital:

The Core questions of the integrated questionnaire of the World Bank were used in this study to measure the social capital.[13] This 25-item questionnaire consists of six domains namely “Groups and networks” (7 items); “Trust and solidarity” (5 items); “Collective action and cooperation” (2 items); “Information and communication” (2 items); “Social cohesion and inclusion” (5 items) and “Empowerment and political action”. The language and wordings of the questions were modified from the original questionnaire considering the local context and reliability was tested. Reliability in terms of Cronbach’s alpha was found to be 0.662 for the women empowerment scale.

Socio-demographic characteristics of the respondent:

The information on the socio-demographic characteristics of the respondent such as age, years of schooling, occupation, socioeconomic status, religion, caste category, marital status, number of children, number of sons, and type of family was also collected, as they were expected to be potential confounders. The socioeconomic status was measured by the type of ration card the respondent had and was categorised as Antyodaya (very poor), Below Poverty Line (BPL) and Above Poverty Line (APL).[14]

Statistical analysis

Data were entered and analysed using SPSS 18.0. The continuous variables were tested for the assumption of normality and were expressed as mean and their standard deviation (SD) if the assumption of normality was met. The categorical variables were expressed in terms of percentages. The distribution of the social capital and women empowerment has been expressed in tertiles. The scores of different domains of women empowerment have been expressed as standardised scores using z-transformation as these domains have been measured on different scales (different number of questions were asked). The univariate analysis was carried out with empowerment as dependent variables and others (social capital and socioeconomic characteristics) as independent variables. Analysis of variance (ANOVA) and Student’s t-test of significance were used appropriately to test the associations of the dependent variable with independent variables. Multivariate linear regression was applied using the backward stepwise method to adjust the effects of the confounders after testing the data for the assumption of multicollinearity. The model was evaluated for stress. The R2 value for the model fit has been reported. Beta and its 95% confidence interval were used to express the strength of the association. For all statistical analyses, P value less than 0.05 was considered to be statistically significant.

Ethical issues

Written Informed consent had been obtained from all the participants before the interview. Anonymity and confidentiality were maintained throughout the research process. The study was initiated after obtaining clearance from the Institute Ethics Committee.

Results

Characteristics of the study subjects

In this study, among the 300 study participants, a majority were from the age group of 26–35 years (49%), homemakers (46.3%), educated up to middle school (between 5th and 10th standard, 56.3%), belonging to Hindu religion (91%), married (96.3%) and residing in a nuclear family (68.3%). Around one-fourth of the participants were BPL [Table 1].

Table 1

Sociodemographic characteristics of the study subjects (n=300)

Variable	Frequency (N=300)	Percentage
Age (years)
< = 25	63	21.0
26–35	147	49.0
>35	90	30.0
Education
Up to 4th standard	33	11.0
5–10	169	56.3
>10	98	32.7
Occupation
Homemaker	139	46.3
Labourer	146	48.7
Business	10	3.3
Service	5	1.7
Socio-economic status
Below poverty line	81	27.0
Above poverty line	219	73.0
Religion
Hindu	273	91.0
Buddhist	25	8.3
Muslim	2	0.7
Marital status
Married	289	96.3
Unmarried	11	3.7
Have at the list one living child
No	15	5.0
Yes	285	95.0
Have at the least one living son
No	75	25.0
Yes	225	75.0
Family type
Nuclear	205	68.3
Joint	95	31.7

Magnitude of women empowerment

The mean (SD) score of women empowerment for the study subjects was 20.7 (4.5) with a median of 22.0. The distribution of the overall women empowerment score (median [Interquartile range]) is depicted in Figure 1. The mean percentage scores in different domains of women empowerment revealed that the participants had a better percentage score in the “Household decision-making” domain compared to the other domains [Figure 2].

Figure 1

Distribution of Women empowerment score among study participants (n=300)

Figure 2

Distribution of Mean percent scores of different domains of Women empowerment (n=300)

Determinants of women empowerment—Univariate analysis

The women empowerment score differed significantly in the different age groups (P < 0.001). It was the lowest in the youngest age group (<25 years) and highest in the oldest age group (>35 years). Similarly, the score of women empowerment increased significantly as the education level increased. The mean scores were 20.1, 20.2 and 21.8 for women with education up to the fourth standard, 5th–10th standard, and >10th standard, respectively. the empowerment scores were 19.3 in the BPL group as against 21.3 in the APL group. There was no statistically significant difference between the empowerment scores of women by religion (P = 0.199). Women, who were married, had at least one child, had at least one son, and who were from the nuclear family had significantly higher empowerment scores as compared to their counterparts [Table 2].

Table 2

Association of women empowerment with sociodemographic factors (n=300)

Variables	Frequency	Women empowerment		P-value
		Mean	SD
Overall	300	20.7	4.5
Age (years)
< = 25	63 (21.0)	16.8	4.9
26–35	147 (49.0)	21.8	3.8	<0.001
>35	90 (30.0)	21.8	3.5
Education
Up to 4th standard	33 (11.0)	20.1	3.2
5–10 standard	169 (56.3)	20.2	4.5	0.012
>10 standard	98 (32.7)	21.8	4.6
Occupation
Homemaker	139 (46.3)	20.1	4.0
Labourer	146 (48.7)	20.8	4.7	<0.001
Business	10 (3.3)	26.4	0.9
Service	5 (1.7)	25.8	1.1
Socioeconomic status
Below poverty line	81 (27.0)	19.3	4.1	0.001
Above poverty line	219 (73.0)	21.3	4.5
Religion
Hindu	273 (91.0)	20.6	4.5	0.199
Buddhist	25 (8.3)	22.3	3.2
Muslim	2 (0.7)	20.5	7.8
Marital status
Married	289 (96.3)	20.9	4.5	<0.001
Unmarried	11 (3.7)	16.1	0.8
Have at least one living child
No	15 (5.0)	17.1	5.0	0.001
Yes	285 (95.0)	20.9	4.4
Have at least one living son
No	75 (25.0)	19.1	4.8	<0.001
Yes	225 (75.0)	21.3	4.2
Family type
Nuclear	205 (68.3)	21.9	3.6	<0.001
Joint	95 (31.7)	18.3	5.17655

Determinants of women empowerment—the role of social capital

The significant determinants of women empowerment were social capital, age, education, religion, marital status and family type. With a 1-year increase in age, the women empowerment score increased by 0.136 units. Similarly, for every standard of education, the score increased by 0.354 units. As religion moved from Muslim to Buddhist and then to Hindu, the empowerment increased by 1.438 units at each step. Being unmarried decreased the score by 1.948 and being from a joint family decreased by 3.689 units. The role of social capital in women empowerment was significant. When adjusted for confounding effect of age, education, religion and other determinants, as social capital increased by one unit, the empowerment score increased by half unit [Table 3].

Table 3

Determinants of women empowerment—multivariate linear regression using backward stepwise method (n=300)

Variable	Beta	95% CI of beta	P-value
Constant	9.758	5.812 to 13.703	<0.001
Social capital	0.501	0.353 to 0.649	<0.001
Age	0.136	0.060 to 0.211	<0.001
Education	0.354	0.204 to 0.505	<0.001
Religion–Hindu	1.438	0.242 to 2.633	0.019
Marital status—Unmarried	-1.948	-2.989 to -0.907	<0.001
Family type—Joint	-3.689	-4.638 to -2.740	<0.001

Footnote: Negelkarke R2=0.487%

Domains of women empowerment: distribution and determinants

The significant determinants of women’s mobility were age, marital status, having a child, having a son and type of family. There is a significantly higher probability of mobility of women who are older, unmarried, have at least one child, have at least one son and belong to a nuclear family. Similarly, the determinants of interspousal communication were age, education, occupation, socioeconomic status and having a son. Older women, more of education, women from higher the socioeconomic strata, working women, and those who have at least one living son have significantly higher chances of establishing interspousal communication as compared to their counterparts. Women’s involvement in household decisions is significantly higher in older women, those who have a living child or/and son and those who belong to a nuclear family as compared to their counterparts. The labourer women have a lesser say in household decision-making as compared to the other women. Older, educated, non-Hindu, unmarried women and women who have at least one living child/son and those who belong to a nuclear family have a significantly higher probability of possession of an asset as compared to their counterparts. Again, the labourer women have a significantly lower probability of having an asset possession as compared to the women in the other occupations including housewives [Table 4].

Table 4

Domains of women empowerment: distribution and determinants (n=300)

Variables	Mobility		Inter-spousal communication		Household decision-making		Asset ownership
	Z-score	P-value	Z-score	P-value	Z-score	P-value	Z-score	P-value
Age (years)
< = 25	−0.991 (1.091)	<0.001	−0.213 (0.918)	<0.001	−0.905 (1.234)	<0.001	−0.846 (1.045)	<0.001
26–35	0.156 (0.863)		0.259 (0.937)		0.099 (0.719)		0.112 (0.846)
>35	0.441 (0.604)		0.285 (0.352)		0.513 (0.743)		0.413 (0.848)
Education
Up to 4th standard	0.117 (0.891)	0.134	−0.858 (0.979)	<0.001	−0.164 (0.928)	<0.001	−0.203 (1.052)	0.018
5–10 standard	−0.102 (1.015)		−0.159 (0.943)		0.015 (1.107)		0.053 (0.977)
>10 standard	0.135 (0.996)		0.522 (0.816)		0.457 (0.081)		0.323 (0.851)
Occupation
Homemaker	0.024 (0.943)	0.20	−0.475 (0.985)	<0.001	0.148 (0.808)	0.015	0.252 (0.911)	<0.001
Labourer	−0.099 (1.068)		0.309 (0.845)		−0.175 (1.156)		−0.376 (0.934)
Business	0.759 (0.193)		1.029 (0.488)		0.457 (0.026)		1.366 (0.409)
Service	0.698 (0.273)		0.875 (0.691)		0.342 (0.099)		1.171 (0.532)
Socioeconomic status
Below poverty line	−0.171 (1.078)	0.073	−0.425 (0.970)	<0.001	−0.042 (0.896)	0.674	0.084 (0.884)	0.375
Above poverty line	0.062 (0.964)		0.142 (0.971)		0.014 (1.034)		−0.031 (1.039)
Religion
Hindu	−0.024 (1,018)	0.391	−0.008 (1.015)	0.480	−0.026 (1.031)	0.353	−0.065 (0.994)	0.001
Buddhist	0.259 (0.766)		0.047 (0.817)		0.266 (0.528)		0.588 (0.793)
Muslim	0.021 (-)		1.184 (-)		0.457 (-)		1.560 (-)
Marital status
Married	−0.028 (1.005)	0.008	−0.168 (1.078)	0.070	−0.165 (0.994)	0.053	−0.032 (0.994)	0.004
Unmarried	0.821 (0.061)		0.698 (0.273)		0.361 (1.010)		0.853 (0.764)
Have at the list one living child
No	−1.255 (1.301)	<0.001	0.154 (0.848)	0.540	−0.983 (1.166)	<0.001	−0.707 (1.081)	0.005
Yes	0.066 (0.938)		−0.009 (1.008)		0.054 (0.963)		0.037 (0.983)
Have at the least one living son
No	−0.383 (1.174)	<0.001	−0.210 (1.013)	0.039	−0.308 (1.150)	0.003	−0.163 (0.979)	0.102
Yes	0.126 (0.903)		0.071 (0.987)		0.102 (0.925)		0.054 (1.003)
Family type
Nuclear	0.210 (0.817)	<0.001	−0.011 (1.031)	0.773	0.481 (0.486)	<0.001	0.245 (0.875)	<0.001
Joint	−0.451 (1.193)		0.024 (0.938)		−0.978 (1.062)		−0.526 (1.051)

Discussion

In this study, women empowerment was assessed using four domains: mobility, interspouse communication, household decision-making and asset ownership. Empowerment is context-specific, and there is no ‘fit for all’ instrument to measure it. The literature reveals different conceptual frameworks and domains used to measure women empowerment.[1516] The domains used in this study were considered suitable by the subject experts in the local context. In the present study, social capital, age, education, religion, marital status and family type are significant determinants of women empowerment.

In this research, social capital was found to be significantly predicting women empowerment, after adjustment with potential socioeconomic confounders. In the published literature, the association between social capital and empowerment is complex and less studied. In a report of the World Bank, it was argued that social capital is an expression of the social groups that allows for collective action, and thereby, citizen empowerment. Further, community-driven development (CDD) was expressed as a sub-form of social capital that can have a positive effect on citizen empowerment and development, and SHGs can act as facilitators of CDD.[17] In another report of the same organisation, Grootaert C commented that “building social capital is one way to facilitate empowerment”. According to the investigator, empowerment, social capital and CDD act through a two-way process. The CDD builds a social capital and makes communities empowered which in turn take charge of their development.[18] In a study on rural Northern Ethiopia, Nega F et al.[19] observed a positive association between empowerment and social capital. In the above-mentioned study, researchers tried to find out the link between social capital and empowerment, but none are specific for women empowerment.

There is a paucity of literature on how social capital specifically influences women empowerment. Mayoux L found in his study at Cameroon that microfinance programmes which build social capital can indeed make a significant contribution to women’s empowerment.[20] In a study conducted in selected villages of Bihar, Janssens W observed a significant impact of a women’s empowerment program on social capital. She also observed that the effects on social capital were correlated with the empowerment of women in the study villages.[21]

In this study, it was observed that all the domains of women’s empowerment increased positively with the age of the participant. In a study using the Demographic and Health Survey (1998–1999) data, Gupta K et al.[22] found that household autonomy and freedom of movement significantly increased with increasing age. Wiklander J et al.[23] also got similar results in a study at Uttar Pradesh and Tamil Nadu, India. According to Indian culture, most of the newly married women used to stay along with their in-laws in a joint family and were not expected to participate in decision-making or moving freely outside the home. So, culturally, they are in a disadvantaged condition. Poorer mobility of the participants residing in a joint family (compared to the nuclear family) and married (compared to unmarried) as observed in this study, can also be explained by the same cultural disadvantage.

Inter-spousal communication, household decision-making and asset ownership were found to be significantly associated with the educational status of the study participants in this study. The result was in line with that by Gupta K et al.[22] Educating a woman brings about self-esteem and confidence, thereby, promoting active participation in her family and society.[24]

In this study, it was also observed that the overall women empowerment scores significantly increased with income generation. Those who were working outside were having higher scores (compared to homemakers) in inter-spouse communication, household decision and asset ownership, but not in mobility. In contrast, Gupta K et al. and Wiklander J et al.[23] observed a higher women empowerment score in all domains (including mobility) among the working women.[22] the working women have financial security and contribute in household asset development, thereby increasing her social status in her family, which leads them to have greater control over the household decision-making. This reflects the argument of Amartya Sen.[25]

In this study, the overall empowerment score was significantly higher in the participants from APL compared to their BPL counterparts. Gupta K et al.,[22] and Wiklander J et al.[23] also observed household standard of living as a significant predictor of women empowerment.

Women who have at least one living child in the family were found to be having a higher women empowerment score. This finding is in congruence with the empowerment model of Wiklander J et al.[23] It emphasises that a woman’s status within the family declines when she “fails” to reproduce. He also reported that when a woman has a stillbirth, it creates a negative effect on her power and authority in the family. On the other hand, a higher empowerment score in women having at least one living son, as observed in this study, indicates the patriarchal social system. In this study, it was also observed that women having no living child in the family were having a poor women empowerment score.

It was also observed that the score for asset ownership was significantly high among unmarried persons compared to married, which also reflects the patriarchal pattern of inheritance and the opinion or tradition that women are considered less suitable to own property than men. Wiklander J et al.[23] observed women with fewer possibilities inheriting land or other property with sons in the household in Tamil Nadu, which may also be explained by the same reasoning.

The main strength of this community-based study was that it tried to enlighten a pertinent social issue where, to the best of our knowledge, limited evidence was available to date, especially in the current study settings. However, the findings of this study need to be interpreted with limited external validity. Further, some domains used in the previous studies, like freedom from domestic abuse, ability to express opinion,[23] attitude towards gender preference[22] were not used in this study. Again, exposure to mass media,[22] history of stillbirth, husband’s educational status, ownership of a house, age at marriage,[23] social psychology, psychological patterns of the family (patrilocality, patrilineality and dowry system), and psychological patterns of individual women[26] were found to be significant predictors of women empowerment in the previous studies, but could not be included in this study.

Social capital was found to be a significant determinant of women empowerment. To achieve Sustainable Development Goals by 2030, planned efforts need to be instituted to initiate community-driven action, which has the potential to improve the social capital, and thereby, promote empowerment. Evidence shows that community-based development projects using SHGs[27] and microfinance programmes[28] have a significant role in improving the social capital in different parts of the world.[21] So, the implementation of such projects either through the government or NGO will be greatly beneficial for the study population.

As the educational status was found to be a significant predictor of women empowerment, collective efforts in creating equal opportunities, increasing the enrolment of the girl-child into schools and preventing school drop-out need to be strengthened. Significant improvement of the empowerment of women involved in occupational groups reflects the need for financial upliftment of women for their overall development. It reinforces the need for SHGs and microfinancing, as measures of income generation for those rural women.

Women having no living child in the family were having poor women empowerment scores. This emphasises the need to strengthen the reproductive and child health services to enhance women empowerment. A higher empowerment score among those having a baby boy reflects the patriarchal pattern of thinking and attitude of family members, reflecting the need for enhancing gender-sensitive education and counselling. Significantly, a lower social capital among unmarried/divorced women can point towards their social vulnerability, which emphasises the need to involve them in community-based development programmes.

To conclude, social capital is a significant determinant of women empowerment. Planned efforts need to be instituted to augment women empowerment which is indispensable to fulfil the Sustainable Development Goals by 2030. Women with less education, having lower socioeconomic status and not involved in gainful occupation/involved in manual labour (maid) were found to be less empowered compared to their corresponding categories. All efforts should be made to improve their social status and empowerment. Community-based development projects using SHG and microfinance programmes need to be promoted by the government through intra-sectoral coordination, as they have a significant role in improving the social capital, and thereby, promoting women empowerment.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Key Messages

Social capital was found to be a significant determinant of women empowerment. To achieve Sustainable Development Goals, planned efforts need to be instituted to improve the social capital in the form of community-based development which in turn will improve women empowerment.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.