Background : Previous research suggests that both network contagion and connection contribute to social network effects. Interventions that integrate network-based approaches may be the most effective. Objective : This dissertation aims to test network effects on two diverse outcomes in two diverse settings : substance abuse within high schools in the United States, and latrine ownership within villages in Southern India. Methods : We analyzed data from two separate datasets. For Chapter 2 we used data from the National Longitudinal Study of Adolescent Health, a nationally representative multi-wave study that includes full social network data on adolescents in the United States. We examined the effect of friends' parents' authoritative parenting on substance abuse outcomes in adolescents. For Chapters 3 and 4 we used data from the Social Networks and Microfinance Study to test social network predictors of individual latrine ownership in rural South India. In Chapter 3 we examined the correlation between direct social contact latrine ownership and individual latrine ownership. We also explored interactions of social contact latrine ownership with individual network position and demographic characteristics. In chapter 4 we compared the effects of social network latrine ownership proportions on individuals' latrine ownership at three levels: that of their direct social contacts, their network communities, and their villages. Results : In Chapter 2, adolescents whose friends' parents were authoritative were significantly less likely to abuse substances, and this effect was only partially mediated through friends' substance abuse. In Chapter 3, the latrine ownership of individuals was significantly correlated with that of their direct social contacts. Furthermore, those more central to the network were more likely to own latrines although the direct effect of the latrine ownership of social contacts was also smallest for this group. In Chapter 4, while both the latrine ownership proportions of direct social contacts and of those within the village significantly predicted individual latrine ownership, the effect of the network community was the strongest. Discussion : The results of this body of work provide support for incorporating network strategies into health behavior interventions, while highlighting some novel approaches beyond what has been traditionally used

Since the 1970s sociologists have explored the best means for measuring social networks, although few name generator analyses have used sociocentric data or data from developing countries, partly because sociocentric studies in developing countries have been scant. Here, we analyze 12 different name generators used in a sociocentric network study conducted in 75 villages in rural Karnataka, India. Having unusual sociocentric data from a non-Western context allowed us to extend previous name generator research through the unique analyses of network structural measures, an extensive consideration of homophily, and investigation of status difference between egos and alters. We found that domestic interaction questions generated networks that were highly clustered and highly centralized. Similarity between respondents and their nominated contacts was strongest for gender, caste, and religion. We also found that domestic interaction name generators yielded the most homogeneous ties, while advice questions yielded the most heterogeneous. Participants were generally more likely to nominate those of higher social status, although certain questions, such as who participants talk to uncovered more egalitarian relationships, while other name generators elicited the names of social contacts distinctly higher or lower in status than the respondent. Some questions also seemed to uncover networks that were specific to the cultural context, suggesting that network researchers should balance local relevance with global generalizability when choosing name generators.

The purpose of this study was to assess via longitudinal analysis whether women's economic empowerment and financial inclusion predicts incident IPV. This prospective study involved analysis of three waves of survey data collected from rural young married women (n = 853 women) in Maharashtra at baseline and 9&18 month follow-ups. This study, which was in the field from 2012 to 2014, was conducted as part of a larger family planning evaluation study unrelated to economic empowerment. Participants were surveyed on economic empowerment, as measured by items on women's income generation and joint decision-making of husband's income, and financial inclusion, as measured by bank account ownership. Women's land ownership and participation in microloan programs were also assessed but were too rare (2-3% reporting) to be included in analyses. Longitudinal regression models assessed whether women's economic empowerment predicted incident IPV at follow-up. At Wave 1 (baseline), one in ten women reported IPV in the past six months; 23% reported income generation; 58% reported having their own money; 61% reported joint control over husband's money, and 10% reported bank ownership. Women's income generation and having their own money did not predict IPV over time. However, women maintaining joint control over their husband's income were at a 60% reduced risk for subsequent incident IPV (AOR = 0.40; 95% CI = 0.18, 0.90), and women gaining joint control over time were at a 70% reduced risk for subsequent incident IPV (AOR = 0.30; 95% CI = 0.13, 0.72), relative to women whose husbands maintained sole control over his income. Women who initiated a new bank account by Wave 3 also had a 56% reduced likelihood of reporting incident IPV in this same wave (AOR = 0.44; 95% CI = 0.22, 0.93), relative to those who maintained no bank account at Waves 1 and 3. These findings suggest that women's joint control over husband's income and her financial inclusion as indicated by bank ownership appear to reduce risk for IPV, whereas her income generation or control over her own income do not. Awareness of and participation in financial inclusion services may help reduce women's risk for IPV in rural India and elsewhere.

We conducted longitudinal analyses examining the associations between intimate partner violence (IPV) attitudes and women's reported IPV in couples (N = 762) using 3 waves of data from a randomized controlled trial in Maharashtra, India. We found that, between Waves 1 and 2, men's and women's acceptance of IPV in the overall population decreased significantly while reports of IPV increased. These changes, we hypothesize, are evidence of an exogenous shock, possibly a high profile rape in Delhi in December 2012, that may have impacted the entire population. Cross-sectional associations between men's attitudes towards IPV and reported IPV were not significant in Wave 1, while positively and significantly associated in Waves 2 and 3. Longitudinal analysis showed that reduction in men's acceptance of IPV between Waves 1 and 2 was associated with a lower likelihood of reported IPV in Wave 3. Women's Wave 1 acceptance of IPV was positively associated with reported IPV in the Wave 1 cross-sectional analysis, while Wave 2 and Wave 3 measures of IPV acceptance were negatively associated with reported IPV in Waves 2 and 3 respectively. Longitudinal analyses of the change in women's attitudes towards IPV from Wave 1 to 2 and reported IPV in Wave 3 were insignificant. However, When women first reported IPV in Waves 2 or 3 they were less likely to report acceptance of IPV in that same wave. Findings suggest that changes in husbands' IPV acceptance is predictive of subsequent IPV, while newly experienced IPV predicts decreased IPV acceptance for women. Wave 2 and Wave 3 results were significant for the control group only, evidence that the intervention affected those associations, potentially changing attitudes more quickly than behavior. We recommend interventions that expose community opposition to IPV as a new social norm, and analysis of how the 2012 Delhi rape case may have affected these norms.

Background

Intimate partner violence (IPV) is a complex global problem, not only because it is a human rights issue, but also because it is associated with chronic mental and physical illnesses as well as acute health outcomes related to injuries for women and their children. Attitudes, beliefs, and norms regarding IPV are significantly associated with the likelihood of both IPV experience and perpetration.

Methods

We investigated whether IPV acceptance is correlated across socially connected individuals, whether these correlations differ across types of relationships, and whether social position is associated with the likelihood of accepting IPV. We used sociocentric network data from 831 individuals in rural Honduras to assess the association of IPV acceptance between socially connected individuals across 15 different types of relationships, both within and between households. We also investigated the association between network position and IPV acceptance.

Results

We found that having a social contact that accepts IPV is strongly associated with IPV acceptance among individuals. For women the clustering of IPV acceptance was not significant in between-household relationships, but was concentrated within households. For men, however, while IPV acceptance was strongly clustered within households, men's acceptance of IPV was also correlated with people with whom they regularly converse, their mothers and their siblings, regardless of household. We also found that IPV was more likely to be accepted by less socially-central individuals, and that the correlation between a social contact's IPV acceptance was stronger on the periphery, suggesting that, as a norm, it is held on the periphery of the community.

Conclusion

Our results show that differential targeting of individuals and relationships in order to reduce the acceptability and, subsequently, the prevalence of IPV may be most effective. Because IPV norms seem to be strongly held within households, the household is probably the most logical unit to target in order to implement change. This approach would include the possible benefit of a generational effect. Finally, in social contexts in which perpetration of IPV is not socially acceptable, the most effective strategy may be to implement change not at the center but at the periphery of the community.

Introduction

Despite global progress on many measures of child health, rates of neonatal mortality remain high in the developing world. Evidence suggests that substantial improvements can be achieved with simple, low-cost interventions within family and community settings, particularly those designed to change knowledge and behaviour at the community level. Using social network analysis to identify structurally influential community members and then targeting them for intervention shows promise for the implementation of sustainable community-wide behaviour change.

Methods and analysis

We will use a detailed understanding of social network structure and function to identify novel ways of targeting influential individuals to foster cascades of behavioural change at a population level. Our work will involve experimental and observational analyses. We will map face-to-face social networks of 30 000 people in 176 villages in Western Honduras, and then conduct a randomised controlled trial of a friendship-based network-targeting algorithm with a set of well-established care interventions. We will also test whether the proportion of the population targeted affects the degree to which the intervention spreads throughout the network. We will test scalable methods of network targeting that would not, in the future, require the actual mapping of social networks but would still offer the prospect of rapidly identifying influential targets for public health interventions.

Ethics and dissemination

The Yale IRB and the Honduran Ministry of Health approved all data collection procedures (Protocol number 1506016012) and all participants will provide informed consent before enrolment. We will publish our findings in peer-reviewed journals as well as engage non-governmental organisations and other actors through venues for exchanging practical methods for behavioural health interventions, such as global health conferences. We will also develop a 'toolkit' for practitioners to use in network-based intervention efforts, including public release of our network mapping software.

Trial registration number

NCT02694679; Pre-results.